Congress onIntelligent Systems(ICCIS 2020)Prediction of bearing fault detection using comparative...
Transcript of Congress onIntelligent Systems(ICCIS 2020)Prediction of bearing fault detection using comparative...
-
Sponsored by
2020
Jointly Organized by
Global Institute of Technology,
Jaipur and
Rajasthan Technical University,
Kota
in Association with
Birla Institute of Applied
Sciences, Uttarakhand and
Soft Computing Research
Society
December 26-27, 2020
2nd International Conference on Communication and Intelligent
Systems (ICCIS 2020)
Souvenir
-
1
TABLE OF CONTENTS
Chief Patron ..................................................................................................................... 8
Patron ............................................................................................................................... 8
General Chair ................................................................................................................... 8
Organising Chair .............................................................................................................. 8
Organising Secretary ........................................................................................................ 8
Finance Chair & Treasurer ............................................................................................... 9
Publicity Committee ........................................................................................................ 9
Publication Committee ................................................................................................... 10
Session Management Committee ................................................................................... 10
Organizing Committee ................................................................................................... 11
Advisory Committee ...................................................................................................... 13
Special Tracks ................................................................................................................ 18
Abstract of Accepted Papers .......................................................................................... 19
Neural Network Imitation Model of Realization of the Business Analysis Process ..... 19
Thermal Modeling of the GaN HEMT device Using Decision Tree Machine Learning
Technique ....................................................................................................................... 19
Low-cost FPGA based onboard computer ..................................................................... 20
A Survey on Solution of Imbalanced Data Classification Problem using SMOTE and
Extreme Learning Machine ............................................................................................ 20
Thermal Imaging Assisted Infection Classification (BoF) for Brinjal Crop.................. 21
Spam Review Detection using k-Means Artificial Bee Colony..................................... 21
Preterm Delivery Prediction Using Gradient Boosting Algorithms .............................. 22
Mutual Learning based Spider Monkey Optimization for Constraint Optimization ..... 22
Analysis Urban Traffic Vehicle Routing Based on Dijkstra Algorithm Optimization .. 23
-
2
A Comprehensive Overview of Quality Enhancement Approach based Biometric Fusion
System using Artificial Intelligence Techniques ........................................................... 24
Topic Modelling and Theme Mining for effective Chat Script Labelling ..................... 24
Rainfall Prediction Using Deep Neural Network........................................................... 25
A Comparative Analysis of Supervised Word Sense Disambiguation in Information
Retrieval ......................................................................................................................... 25
Real-Time Bangladeshi Currency Recognition Using Faster R-CNN Approach for
Visually Impaired People ............................................................................................... 26
Real time deep learning face mask detection model during COVID-19 ....................... 26
Prediction of California Bearing Ratio of Subgrade Soils using Artificial Neural Network
Principles ........................................................................................................................ 27
Prediction of bearing fault detection using comparative analysis of Random Forest, ANN
and Autoencoder methods .............................................................................................. 28
Selection of a mesh network routing protocol for underground mines .......................... 28
Budget oriented reliable WDO Algorithm for Workow Scheduling in Cloud Systems 29
An Energy Efficient Communication Scheme for Multi-robot Coordination deployed for
Search and Rescue Operations ....................................................................................... 29
Butterfly Optimization Algorithm Based Optimal Sizing and Integration of Photovoltaic
System in Multi-Lateral Distribution Network for Interoperability .............................. 30
Document Classification in Robotic Process Automation Using Artificial Intelligence –
A Preliminary Literature Review ................................................................................... 31
Suppliers Selection Using Fuzzy AHP and Fuzzy TOPSIS Method - A Case Study of a
Bearing Manufacturing Company .................................................................................. 31
Artificial Intelligence Optimization Strategies for Invoice Management: A Preliminary
Study .............................................................................................................................. 32
Survey Analysis for Medical Image Compression Techniques ..................................... 32
-
3
A Comparative Study between Data-Based Approaches under Earlier Failure
Detection ........................................................................................................................ 33
Classification of fundus images based on non-binary patterns for the automated screening
of retinal lesions ............................................................................................................. 33
Performance Evaluation of SEIG Under Unbalanced Load Operations Using Genetic
Algorithm ....................................................................................................................... 34
A New Approach to Classify the Boolean Functions Based on Heuristic Technique ... 34
A Modulo (2n-2n-2-1) Adder Design ............................................................................... 34
Influence of Object-Oriented Software Design Measures on Reliability: Fuzzy Inference
System Perspective......................................................................................................... 35
Entity Based Knowledge Graph Information Retrieval for Biomedical Articles .......... 35
Test case prioritization based on Requirement .............................................................. 35
Mining and Predicting No-Show Medical Appointments: Using Hybrid Sampling
Technique ....................................................................................................................... 36
Adaptive Strategy for Environment Exploration in Search and Rescue Missions by
Autonomous Robot ........................................................................................................ 36
Investigating the Effect of Lockdown during COVID-19 on Land Surface Temperature
Using Machine Learning Technique by Google earth Engine: Analysis of Rajasthan,
India................................................................................................................................ 37
Human activity recognition using deep learning based approach .................................. 37
Emotion distribution profile for Movies Recommender systems .................................. 38
Prediction of Modulus of Subgrade Reaction using Machine Language Framework ... 38
Time Fractionalized Lattice Boltzmann Model Based Image Denoising ...................... 39
Distributed and Anonymous E-voting using Blockchain and Ring Signatures ............. 39
Enyo: A Multistage Partition and Transposition based cipher....................................... 40
Real-time multi-obstacle detection and tracking using a vision sensor for autonomous
vehicle ............................................................................................................................ 40
-
4
Healthcare Security: Usage of Generative Models for Malware Adversarial Attacks and
Defense ........................................................................................................................... 41
Exploring Cognitive Process in Extended Data Mining ................................................ 41
Sentiment Analysis from Bangla Text Review Using Feedback Recurrent Neural
Network Model .............................................................................................................. 42
Improved Vehicle Detection and Tracking using YOLO and CSRT ............................ 42
Neuronal Unit of Thoughts (NUTs): A Probabilistic Formalism for Higher-Order
Cognition ........................................................................................................................ 43
A Soft Computing Technique to Optimize Energy Consumption in Wireless Sensor
Networks ........................................................................................................................ 43
Human Identification System based on Latent Fingerprint ........................................... 43
Data quality requirements methodology for an adapted PHM implementation ............ 44
A Comparative Analysis of Japan and India COVID-19 News Using Topic Modeling
Approach ........................................................................................................................ 44
Scaling Depression Level through Facial Image Processing and Social Media
Analysis .......................................................................................................................... 45
Witty City – Smart City on an Intelligent Conway Grid ............................................... 45
Double sided Split Ring Resonator Based Probe feed Patch Antenna with Enhanced
Bandwidth for 5G & Ku Band Applications .................................................................. 46
Reinforcement Learning based Clustering Algorithm for Cognitive Wireless Sensor
Networks ........................................................................................................................ 46
Classification of Social Media Users Based on Temporal Behaviors and Interests ...... 47
Automated Short Video Caption generation using Video Features and Audio Content 47
A Method of Polytexture Modeling in 3D Anatomy Simulators ................................... 48
An Impact of Different Uncertainties and Attacks on the Performance Metrics and
Stability of Industrial Control System............................................................................ 48
-
5
Experiences involving student assistants in interdisciplinary R&D projects using the
example of aerospace computing and bioeconomics: The "HONEYCLOUD" project 49
Parallel Algorithm for Matrix Sort ................................................................................. 50
Stability and Dynamic Power Analysis of Novel 9T SRAM Cell for IoT Applications 50
Leveraging Deep Learning Techniques on Remotely Sensing Agriculture Data .......... 51
A Novel Diagnosis system for Parkinson’s disease using k-means Clustering and
Decision Tree ................................................................................................................. 51
Multidimensional Ensemble LSTM for Wind Speed Prediction ................................... 51
An Investigation of Ground Barriers and Teachers’ Attitude towards Technology-
Enabled Education in Schools ........................................................................................ 52
An Improved Ant Colony Optimization with Correlation and Gini Importance for Feature
Selection ......................................................................................................................... 53
Social Network Analysis Based on Combining Probabilistic Models with Graph Deep
Learning ......................................................................................................................... 53
Unsupervised classification of zero-mean data based on L1-norm Principal Component
Analysis .......................................................................................................................... 54
Data Confidentiality and Integrity in Cloud Storage Environment ............................... 54
Social Media Analytics: Current Trends and Future Prospects ..................................... 55
Automated Sleep Staging using Convolution Neural Network based on Single-channel
EEG signal ..................................................................................................................... 55
Spark based FP-Growth Algorithm for Generating Association Rules from Big Data . 56
A Univariate Data Analysis Approach for Rainfall Forecasting ................................... 56
Improved Adaboost Algorithm with Regression Imputation for Prediction of Chronic
Type 2 Diabetes Mellitus ............................................................................................... 57
Automatic Generation Control of Multi-Area Multi-Source Deregulated Power System
Using Moth Flame Optimization Algorithm .................................................................. 58
A Study on Application of Interplanetary File System .................................................. 59
-
6
Automatic True Vessel Identification by Efficient Removal of False Blood Vessels for
Detection of Retinal Diseases ........................................................................................ 59
A hybrid LWT and DCT based lossless watermarking scheme for color images ......... 60
Kardex: Platformer ......................................................................................................... 60
Optimization of Solar Energy Storage in a Battery for Hybrid Photovoltaic System ... 61
Smart Locking in Home Automation using Internet of Things ..................................... 61
Neural Networks for Detecting Cardiac Arrhythmia from PCG Signals ....................... 62
Diagnosis of Covid-19 in X-ray and CT Images Using Online Clustering Framework 62
Intelligent Control of Pyro Coal Firing in the Energy Intense Cement Industry using
Machine Learning .......................................................................................................... 63
Analysis of GPSR Protocol with different mobility models in VANET Scenario ........ 63
Detection of Cardiac Disease with Less Number of Electrocardiogram Sensor Samples
using Chebyshev ............................................................................................................ 64
Impact Analysis on Essential Parameter of Dual Gate Organic Thin Film Transistor .. 64
Unsupervised Deep Learning Approach for the Identification of Intracranial
haemorrhage in CT images using PCA-Net and K-Means algorithm............................ 65
Do Tourists Act in a Responsible Manner at Anaimalai? Towards a Confirmatory
Perspective ..................................................................................................................... 65
A Survey on Methods and Techniques for Enhancing Storage Capacity of or Codes .. 66
Lexicon based Natural Language Processing and Semantic Analysis of Contractual Risks
on Engineering, Procurement & Construction projects ................................................. 66
Predicting Heart Disease with Multiple Classifiers ....................................................... 67
Ultraviolet Disinfection & Sterilizing Robot ................................................................. 67
Impact of COVID-19: Should Work From Home Be The New Normal? ..................... 68
Algebraic modelling of factors responsible for the process of decision-making ........... 68
-
7
Data labeling impact on deep learning models in digital pathology: A breast cancer case
study ............................................................................................................................... 68
-
8
Chief Patron
Prof. R. A. Gupta, Vice-Chancellor, Rajasthan Technical University,
Kota
Patron
Sh. Rajkumar Kandoi, Vice Chairman, Global Institute of Technology,
Jaipur, India
Sh. Naman Kandoi, CEO, Global Institute of Technology, Jaipur, India
Prof. Dhirendra Mathur, RTU (ATU) TEQIP-III Coordinator
General Chair
Nidhi Singhal, Director, Global Institute of Technology, Jaipur, India
I. C. Sharma, Principal, Global Institute of Technology, Jaipur, India
G. S. Tomar, Birla Institute of Applied Sciences, Bhimtal, Nainital
Uttarakhand, India
Lipo Wang, Nanyang Technological University, Singapore
Organising Chair
J. P. Agarwal, Dean Academics, Global Institute of Technology, Jaipur
Sayar Singh Shekhawat, Global Institute of Technology, Jaipur
Harish Sharma, Rajasthan Technical University, Kota, India
Robin Singh, Birla Institute of Applied Sciences, Uttarakhand, India
Organising Secretary
Girraj Khandelwal, Global Institute of Technology, Jaipur, India
-
9
Bishwajeet Pandey, Birla Institute of Applied Sciences, Uttarakhand,
India
Ponnambalam P, Vellore Institute of Technology, Vellore, India
Sowmya V., Amrita School of Engineering, Coimbatore, Amrita Vishwa
Vidyapeetham, India
Finance Chair & Treasurer
Sandeep Kumar, CHRIST (Deemed to be University), Bangalore
Publicity Committee
Rajesh Rajaan, Global Institute of Technology, Jaipur, India
Nand Kishor Yadav, Illinois Institute of Technology, Chicago, US
Vishal Gupta, AIACT&R, New Delhi, India
Sandesh Tripathi, Birla Institute of Applied Sciences, Uttarakhand, India
Kusum Kumari Bharti, Indian Institute of Information Technology,
Design and Manufacturing, Jabalpur, India
Anirban Das, University of Engineering & Management, Kolkata, India
Avinash Pandey, Jaypee Institute of Information Technology, Noida
Charu Gandhi, Jaypee Institute of Information Technology, Noida, India
Sumit Kumar, Amity School of Engineering and Technology, Noida
Mukesh Saraswat, Jaypee Institute of Information Technology, Noida
C. Rani, VIT Vellore, India
D. L. Suthar, Wollo University, Ethiopia
Faruk Ucar, Marmara University
-
10
Himani Bansal, Jaypee Institute of Information Technology, Noida, India
Himanshu Mittal, Jaypee Institute of Information Technology, Noida
Kusum Lata Agarwal, JIET Jodhpur, India
Neha, NIT Hamirpur, India
Ponnambalam P, VIT Vellore, India
Ramesh C. Poonia, Cyber-Physical Systems Laboratory, NTNU,
Ålesund, Norway
Shambhu Shankar Bharti, Computer Science and Engineering, LNJPIT
Chapra
V. K. Vyas, Sur University College, Oman
Vijay Kumar Bohat, Bennet University, India
Vijander Singh, Manipal University, Jaipur
Publication Committee
Ankur Goyal, Global Institute of Technology, Jaipur, India
Raju Pal, Jaypee Institute of Inormation Technology, Noida, India
Ashutos Bhatt, Birla Institute of Applied Sciences, Uttarakhand, India
Session Management Committee
Sunil Gupta, Global Institute of Technology, Jaipur, India
Dimple Jayaswal, Global Institute of Technology, Jaipur, India
Loveleen Kumar, Global Institute of Technology, Jaipur, India
Abhishek Verma, GLA University, Mathura, India
Ashish Tripathi, Malaviya National Institute of Technology, Jaipur, India
-
11
Ajay Sharma, Government Engineering College Jhalawar, India
Kamlesh Jangid, Rajasthan Technical University, Kota, India
Neeraj Jain, Jaypee Institute of Information Technology, Noida, India
Nirmala Sharma, Rajasthan Technical University, Kota, India
Sakshi Shringi, RTU Kota, India
Shimpi Singh Jadon, Govt. Rajkiya Engineering College Kannauj UP
Twinkle Tiwari, JIIT Noida
Organizing Committee
Nitin Jain, Global Institute of Technology, Jaipur, India
Sohan Gupta, Global Institute of Technology, Jaipur, India
Garima Sharma, Global Institute of Technology, Jaipur, India
Pooja Sharma, Global Institute of Technology, Jaipur, India
Rashi Jain, Global Institute of Technology, Jaipur, India
Hemant Mittal, Global Institute of Technology, Jaipur, India
Sujeet Gupta, Global Institute of Technology, Jaipur, India
Rishabh Sharma, Global Institute of Technology, Jaipur, India
Mohit Gautam, Global Institute of Technology, Jaipur, India
Amit Bohra, Global Institute of Technology, Jaipur, India
Ramkunwar, Global Institute of Technology, Jaipur, India
Kailash Ram, Global Institute of Technology, Jaipur, India
Dinesh Verma, Global Institute of Technology, Jaipur, India
-
12
Ayush Dogra, Punjab University, Punjab
Mostafa el Mallahi, Height Normal School, Sidi Mohamed Ben Abdellah
University, Fez, Morocco
Samrat Kumar Dey, Dhaka International University, Dhaka, Bangladesh
S. K. Thakur, Birla Institute of Applied Sciences, Uttarakhand, India
K. KALAISELVI, Vels Institute of Science, Technology and Advanced
Studies (VISTAS) (Formerly Vels University), Pallavaram, Chennai
Amrit Pal Singh, Jaypee Institute of Information Technology, Noida
Bindu Verma, Jaypee Institute of Information Technology, Noida, India
Rajani K Poonia, JECRC, University, Jaipur
Praveen Kumar Shukla, BBD University Lucknow
S. D. Purohit, Rajasthan Technical University, Kota, India
Dhiraj Sangwan, Sr. Scientist, CSIR-CEERI, PILANI
Devendra Kumar, University of Rajasthan, India
Harish V. Gorewar, RTM Nagpur University, Nagpur
Soniya Lalwani, BKIT, Kota
Jagdev Singh, JECRC University, Jaipur, India
Jayprakash, National Institute of Technology Calicut, India
K G Sharma, Government Engineering College Ajmer
Lokesh Chauhan, National Institute of Technology, Hamirpur, India
Nafis uddin Khan, Jaypee University of Information Technology, Solan
Pinkey Chauhan, Jaypee Institute of Information Technology, Noida
-
13
Ravindra N. Jogekar, RTM Nagpur University, Nagpur
Ritu Agrawal, Malaviya National Institute of Technology, Jaipur, India
Satya Narayan Tazi, Government Engineering College Ajmer India
Shantanu A. Lohi, SGB Amravati University, Amravati
Advisory Committee
Manoj Mahla, Executive Director, Global Institute of Technology, Jaipur
Praveen Sharma, VP – Liaison, Global Institute of Technology, Jaipur
A. K. Singh, Motilal Nehru National Institute of Technology Allahabad
Kedar Nath Das, NIT, Silchar
A K Verma, Western Norway University of Applied Sciences,
Haugesund, Norway
Manoj Thakur, IIT, Mandi, India
Abdel Salam Gomaa, Head of Student Data Management Section,
Department of Mathematics, Statistics and Physics, College of Art and
Sciences, Qatar University, Doha
Aboul Ella Hassanien, Cairo University, Egypt
Adarsh Kumar, UPES, Dehradun
Ajay Vikram Singh, AIIT, Amity University Uttar Pradesh
Akhil Ranjan Garg, MBM Engg. College, Jodhpur
Ali A. Al –Jarrah, Sur University College, Oman
Ali Mirjalili, Torrens University Australia
Alok Kanti Deb, Indian Institute of Technology Kharagpur
-
14
Anand Nayyar, Scientist, Graduate School, Duy Tan University, Da
Nang, Viet Nam
Anand Paul, Kyungpook National University, South Korea
Anuradha Ranasinghe, Liverpool Hope University, UK
Anurag Jain, GGSIP University, Delhi
Aruna Tiwari, Indian Institute of Technology Indore
Arun Solanki, Gautam Buddha University, Greater Noida
Ashish Kr. Luhach, The PNG University of Technology, PNG
Ashvini Chaturvedi, NIT Suratkal, India
Atulya K. Nagar, Liverpool Hope University, UK
Ayush Dogra, CSIR NPDF, CSIR-CSIO Research Lab, Government of
India
B. Padmaja Rani, JNTU Hyderabad
Basant Agarwal, IIIT Kota, Rajasthan India
Carlos A Coello Coello, Investigador CINVESTAV 3F
D.L. Suthar, Wollo University, Ethiopia
Dan Simon, Cleveland State University USA
Debasish Ghose, IISc Bangalore
Deepak Garg, Bennett University, India'
Dhirendra Mathur, RTU Kota
Dinesh Goyal, Poornima Institute of Engineering & Technology, Jaipur
Dumitru Baleanu, Cankaya University
-
15
Faruk Ucar, Marmara University
Garima Mittal, IIM Lucknow
Gonçalo Marques, University of Beira Interior, Portugal
Hanaa Hachimi, Ibn Tofail University, Morocco
J. Senthilnath, Scientist, Machine Intellection, Institute for Infocomm
Research (I²R) | Agency for Science, Technology and Research
(A*STAR), Singapore
Janmenjoy Nayak, Aditya Institute of Technology and Management
Tekkali, K. Kotturu Srikakulam, Andhra Pradesh-532201, India
Janos Arpad Kosa, Neumann Janos University, Hungary
K. S. Nisar, Prince Sattam bin Abdulaziz University Riyadh, Saudi
Arabia
Kapil Sharma, Head Department of IT, DTU, India
Kedar Nath Das, National Institute of Technology Silchar (NITS),
Silchar, India
Kusum Deep, Indian Institute of Technology, Roorkee, India
Lipo wang, NTU Singapore
Mahesh Bundele, Poornima College of Engineering, Jaipur
Manju, JIIT, Noida
Manoj Thakur, IIT Mandi
Mario José Diván, Data Science Research Group Universidad Nacional
de La Pampa Coronel Gil 353, Primer Piso - Santa Rosa (CP 6300) La
Pampa Argentina
-
16
Maurice Clerc, Independent Consultant, France
Mohammad S Khan, Director of Network Science and Analysis Lab
(NSAL) Department of Computing East Tennessee State University
Johnson City, TN 37614-1266, USA
N. R. Pal, Indian Statistical Institute, Kolkata, India
Neil Buckley, Liverpool Hope University, UK
Nilanjan Dey, Techno India College of Technology, India
Nishchal K. Verma, Indian Institute of Technology Kanpur, India
Noor Zaman, Taylor's University, Malaysia
P. Vijaykumar, University College of Engineering Tindivanam
Pankaj Srivastava, MNNIT, Prayagraj
Prashant Jamwal, Nazarbayev University, Kazakhstan
R. C. Mittal, Jaypee Institute of Inormation Technology, India
Ravinder Rena, NWU School of Business, North West University,
Mafikeng Campus, South Africa
Ravi Raj Choudhary, Central University of Rajasthan
S. Sundaram, IISc Bangalore
Said Salhi, Kent Business School | University of Kent
Sarbani Roy, Jadavpur University, Kolkata
Satish Chand, Jawaharlal Nehru University
Sanjeevikumar Padmanaban, Department of Energy Technology,
Aalborg University, Esbjerg, Denmark
-
17
Sudeep Tanwar, NIRMA University, Gujrat
Sunita Agrawal, Motilal Nehru NIT Allahabad
Suresh Satapathy, KIIT Deemed to be University, Bhubaneswar
Swagatam Das, Indian Statistical Institute, Kolkata, India
T. V. Vijay Kumar, Jawaharlal Nehru University
V. K. Vyas, Sur University College, Oman
Vivek Jaglan, Dean Research, GEHU, Dehradun
-
18
Special Tracks
1. Soft Computing and Smart Wireless Computing Using Internet of
Things
Chair: Dr. Nikhil Marriwala, Dr. Ram Avtar, Dr. Vishal Dattana
2. Artificial Intelligence for Medical, Biomedical and Health
Informatics (AIMBHI)
Chair: Dr. Sowmya. V, Dr. Vinayakumar Ravi, Dr. Gopalakrishnan.
E.A., Dr. Soman K.P, Dr. Tuan D. Pham
3. Intelligent and Advanced Control Strategies (IACS)
Chair: Dr. Praveen Kumar M, Dr. Ponnambalam P
4. Bio-Inspired Algorithms in Artificial Vision (BIAAV)
Chair: Dr. Himanshu Mittal, Dr. Avinash Chandra Pandey, Dr.
Satish Chand, Dr. Mukesh Prasad
-
19
Abstract of Accepted Papers
Neural Network Imitation Model of Realization of the
Business Analysis Process
Katerina Kolesnikova1, Olga Mezentseva1, Olena
Savielieva2
1Management Technology Department of Taras Shevchenko National
University, Ukraine
2South Ukrainian National Pedagogical University named after K. D.
Ushynsky, Ukraine
The article reflects the results of research related to the reorganization of the
business analysis processes of an IT enterprise by building a neural network
simulation model. For comparison, an analysis of the business process “as is” was
carried out using standard software tools on the basis of expert assumptions,
notations were made in BPMN. Ways to optimize the process are proposed. Then,
based on the same initial data, a simulation model is pro-posed for generating virtual
indicators based on various kinds of business process load scenarios and initial
independent conditions. Using neural net-work modeling based on simulation data,
a “as it should be” model is proposed. A possible economic effect was obtained from
each optimization project from the received 25 scenarios. The article offers a
mathematical description of the model and method for modeling these interactions
from the perspective of game theory. This allows you to build appropriate formal
models.
Thermal Modeling of the GaN HEMT device Using
Decision Tree Machine Learning Technique
Niketa Sharma1*, Yogendra Kumar Gupta1, and Ashish
Sharma2, Harish Sharma3
1Swami Keshvanand Institute of Technology Management & Gramothan,
Jaipur 302021, India
2Indian Institute of Information Technology, Kota 302021, India
3Rajasthan Technical University, Kota, 324010, India
In this paper, we have proposed electro thermal modeling of GaN-based HEMT
devices. A data-driven approach has been implemented for a temperature range
varying from 300 K to 600 K, based on one of the core methods of Machine learning
techniques, decision tree (DT). The performance of the proposed models was
validated through the simulated test examples. The attained outcomes show that the
-
20
developed models predict the HEMT device characteristics accurately depending on
the determined mean squared error between the actual and anticipated
characteristics. The paper also indicates that the decision tree technique could be
specifically beneficial when data is nonlinear and multidimensional, with the
different process parameters exhibited profoundly complex interactions.
Low-cost FPGA based onboard computer
Dirk van Wyk and Vipin Balyan
Cape Peninsula University of Technology, Cape Town, SA
Abstract. This paper discusses the feasibility for the use of commercially available
reconfigurable Field Programmable Gate Arrays as the main system processor for
small satellite systems and subsystems. FPGAs are in high demand as the space
industry and applications are rapidly increasing and evolving. The use of FPGAs
within the design of spacecraft systems reduce the design cost, as well as the
turnaround time. It is anticipated that the Single Board Computer can be used as a
configurable on-board computer with high flexibility allowing in-orbit
reconfiguration [1]. Modern FPGAs are designed with embedded processing
systems integrated inside the core allowing monotonous performance task to
execute more easily with flexibility. One single printed circuit board with the
computing power that will handle all the challenging requirements for performance
and functionality of the payload data handling.
A Survey on Solution of Imbalanced Data Classification
Problem using SMOTE and Extreme Learning Machine
Ankur Goyal1, Likhita Rathore2, and Sandeep Kumar3*
1Global Institute of Technology, Jaipur, India
2Yagyavalkya Institute of Technology, Jaipur, India
3CHRIST (Deemed to be University), Bangalore – 560074, Karnataka,
India
Imbalanced data are a common classification problem. Since it occurs in most real
fields, this trend is increasingly important. It is of particular concern for highly
imbalanced data sets (when the class ratio is high). Different techniques have been
developed to deal with supervised learning sets. SMOTE is a well-known method
for over-sampling that discusses imbalances at the level of the data. In the area,
unequal data is widely distributed, and ensemble learning algorithms are a more
efficient classifier in classifying imbalances. SMOTE synthetically contrasts two
closely connected vectors. The learning algorithm itself, however, is not designed
for imbalanced results. The Simple Ensemble idea, as well as the SMOTE algorithm,
works with imbalanced data. There are detailed studies about imbalanced data
problems and resolving this problem through several approaches. There are various
-
21
approaches to overcome this problem, but we mainly focused on SMOTE and
extreme learning machine algorithms.
Thermal Imaging Assisted Infection Classification (BoF)
for Brinjal Crop
Shubhangi Verma, Sachin Kumar, Sumita Mishra
Amity University Lucknow, India
In the development of economy, agriculture has always played an important role for
different nations; since it is considered to be the main source of income, food, and
employment to rural populations in the country, owing to diversified geographical
locations, environmental conditions and pest attacks, it is of prime importance to
devise technological assisted methods to monitor and provide early remedial actions
for the damage and infections to the crop. Algorithm proposed focuses on health
monitoring of Brinjal crop using digital thermal imaging. Paper aims to identify the
plant disease by analyzing thermal images of Brinjal leaves. Infrared images are rich
in important hidden details that are not visible due to their low contrast and blurring.
Experiment was conducted on two sets of images, first set comprising of healthy
and infected thermal images and second comprising of normal RGB capture of
healthy and infected images, 30 to 35 images per crop per set were acquired, total
dataset analyzed had 1160 images, the process of identification was implemented
via Bag of Features (BoF), under the umbrella; feature extraction was carried out by
SIFT operator and classification was performed using classification MLSTSVM.
Simulation was implemented using MATLAB 2018b. Results showed that duration
of the process was less for RGB images by a margin of approximately 6 secs but the
accuracy efficiency achieved was more for thermal images by margin of 3%, having
87% percent in all. From the results it can be concluded that however duration
required for the identification was more for thermal images but still percentage
accuracy is more for thermal images, thus thermal images assisted algorithm can be
employed for crops in remote scenarios where accuracy plays a vital role.
Spam Review Detection using k-Means Artificial Bee
Colony
Prateek Saini, Sakshi Shringi, Nirmala Sharma, and
Harish Sharma
Rajasthan Technical University, Kota, India
The current businesses which use internet for marketing depend on online reviews
and these online reviews can direct a customer towards or away from a product and
service. This effect of online reviews is the reason that business uses spam reviews
to either benefit their business or hinder their rival’s business. In this paper, a novel
solution k-means Artificial Bee Colony for feature selection and optimized clusters
using Artificial Bee Colony to detect spam reviews is presented. We report the
-
22
testing of our novel method on three different data sets. The findings of our testing
are encouraging and show a respectable performance on all three data sets.
Preterm Delivery Prediction Using Gradient Boosting
Algorithms
Monarch Saha1, Soumen Nayak2*, Nirjharini Mohanty1,
Vishal Baral1, Imlee Rout1
1Department of Computer Science and Information Technology, ITER,
S’O’A Deemed to be University, Bhubaneswar, India
2Department of Computer Science and Engineering, ITER, S’O’A
Deemed to be University, Bhubaneswar, India
Even with the scientific and medical advancements in today's world, the number of
preterm births keeps increasing. The preterm babies continue to face the
consequences of their early birth, lifelong. If predicted accurately, these agony
experiences with financial exhaustion can be dodged, with proper care and attention.
The Preterm delivery means when babies are born before the thirty-seventh week of
pregnancy, arise of infants, who has a greater risk of death than those born at fill
term. This work will intimate the mother about the preterm delivery, so that they
can take proper precaution and both mother and child will be health during the birth
time. The classification method in Machine Learning has enabled us to attain the
desired goal in most of the spheres of bio-medicine. The need for accuracy calls out
for the best possible classification method for a prediction that can help both the
mothers and their babies. The research is carried out with the central idea of this
paper is figuring out if the birth is going to be premature or not by implementing
Gradient Boosting classifiers of Machine Learning and conducting a comparative
study between two such classification algorithms, namely, XGB Classifier and
LGBM Classifier, to select the most precise algorithm for prediction. The paper
additionally highlights the factors on which the final deduction is made.
Mutual Learning based Spider Monkey Optimization for
Constraint Optimization
Meghna Singh, Nirmala Sharma, and Harish Sharma
Rajasthan Technical University, Kota
Spider Monkey Optimization (SMO) is a well-known population based algorithm
efficient for it's optimizing features. Though, SMO has always been a competitive
algorithm when compared with other optimization algorithms and has outperformed
them even at times but there are a few limitations associated with it as well like slow
convergence and stagnation. The purpose of this article is to overcome such
drawbacks and to set a trade-off between the expedition and exploitation. For the
same, a new alternative of SMO algorithm i.e., Mutual learning based spider
-
23
monkey optimization algorithm (MuLSMO) has been projected being inspired from
the mutual learning concept among individuals. To testify the achievement of the so
proposed algorithm, it is proved over a set of 20 standard optimization functions and
is compared with standard SMO as well as its recent variants namely Modified
limacon SMO (MLSMO), Fibonacci based SMO (FSMO), Power law based SMO
(PLSMO) and Levy flight SMO (LFSMO). Intensive statistical analysis of the
results shows that MuLSMO came out to be significantly better validating our newly
proposed approach.
Analysis Urban Traffic Vehicle Routing Based on Dijkstra
Algorithm Optimization
Truong-Giang Ngo1*, Thi-Kien Dao2, Jothiswaran
Thandapani2,3, Trong-The Nguyen2,4, Duc-Tinh Pham5,6,
Van-Dinh Vu7
1Faculty of Computer Science and Engineering, Thuyloi University,
Hanoi, Vietnam
2Fujian Provincial Key Laboratory of Big Data Mining and Applications,
Fujian University of Technology, Fuzhou, China
3Anna University, Chennai, India
4Haiphong University of Management and Technology, Haiphong,
Vietnam
5Center of Information Technology, Hanoi University of Industry, Hanoi,
Vietnam
6Graduate University of Science and Technology, Vietnam Academy of
Science and Technology, Vietnam 7Faculty of Information Technology, Electric Power University, Hanoi,
Vietnam
The transportation cost, running vehicle time, duration, and distance cost, the route
network of revealed urban vehicles, are considered to need to be analyzed
meaningfully and reasonably planned. This paper suggests an analysis of urban
traffic vehicle routing based on the Dijkstra algorithm optimization that is as a
solution to the road selection and optimization under various constraints jointly built
in the central metropolitan area. The leading metropolitan area road network's
various constraints are to consider the road condition factors and the risk resistance
in road planning. The analysis and realization are verified with the geocoding,
network topology, and network analysis. The results show that the integration of
network analysis technology and the Dijkstra algorithm realizes the urban vehicle
route's optimization decision. Still, the improved Dijkstra algorithm reduces the
number of node visiting and time complexity. Under the driving time and distance
constraints, the speed limit, road hierarchy, and road condition are the line
-
24
selection's restrictive factors. The graphical description could provide technical
support and reference for the driver's driving strip and traffic management
department for decision making.
A Comprehensive Overview of Quality Enhancement
Approach based Biometric Fusion System using Artificial
Intelligence Techniques
Gaurav Jindal and Gaganpreet Kaur
Sri Guru Granth Sahib World University Fatehgarh Sahib (Punjab), India
Biometric authentication has been reported to be one of the most emerging research
fields and their achievements are inseparable from the support of a variety of single-
modal and multi-modal biometric traits (e.g. Finger-print, hand geometry, iris, face,
ear, gait and so on). Generally, the characteristics of the biometric trait are used to
provide authentication information for any security systems. Sometimes the
characteristics of biometric traits are difficult to acquire in an appropriate way and
it is essential to use several pre-processing and post-processing algorithms to design
and process them and to use them on security systems. In this case, this review paper
presents a comprehensive overview of the Biometric Fusion System (BFS) with
some pre-processing and post-processing approaches using the concept of Artificial-
Intelligence/Machine-Learning techniques. In this regard, the following subject
matters are discussed: 1. Biometric traits quality improvement techniques in the
BFS; 2. Feature extraction and optimization approaches; 3. Analysis of classifiers
to improve biometric fusion accuracy; and 4. Existing challenges of BFS. In
Addition, a survey of existing work based on their classification accuracy is also
discussed. The main purpose of this survey is to provide a comprehensive overview
of BFS with the role of a different biometric trait in biometrics fusion.
Topic Modelling and Theme Mining for effective Chat
Script Labelling
Viplove Pandey1, Sai Krishna Reddy1 and Prakash
Selvakumar2
1Data Scientist, Genpact, Bengaluru, India
2Assistant Vice President, Genpact, Bengaluru, India
The recent technological development has brought about a revolution in the volume
of data and has brought forward the challenges associated with it. Especially,
unstructured and unlabeled data in call center, in the form of text, conversation or
transcriptional data, has seen immense growth. Identifying the concern of customer
and routing it to the correct service provider enables the effective query handling
and enhances the customer’s experience. Both are a crucial factor for efficient
-
25
performance of the customer service. However manual routing renders itself limited
when exposed to high data influx and the problem aggravates when the data is in
huge volume and without any labels. On the other hand, manual labelling of large
amount of data can be time consuming, exhaustive and prone to human errors. In
this paper, we present a framework to leverage topic modelling augmented with
theme mining to identify major topics and sub-topics along with their dominant
themes based on the corpus-topic distribution and chat pattern recognition in the
chat transcripts without any human effort.
Rainfall Prediction Using Deep Neural Network
Chitra Desai
National Defence Academy, Pune, India
A model when stated in simple terms is a mathematical equation, which is true when
it implies to any model in machine learning including deep neural network. Every
model will generate output for a given input, but important is to get output of desired
accuracy. In machine learning models are trained on training data and their best fit
is judged on testing data. Before fitting the training data to the model to predict on
unknown (test) data, pre-processing of data is essential to ensure model accuracy to
acceptable level. This paper presents steps involved in pre-processing raw labelled
dataset (Seattle weather) with 25551 records (from year 1948 to 2017) to make it
suitable for input to a deep neural network model. The data is split into 80% of
training data and 20% of testing data. Scaling is performed on the data before it is
passed to the deep neural network model. Deep neural network model that is
multilayer perceptron model using Sequential model API with dense layer is built
and compiled using Adam optimizer resulting accuracy of 97.33% in predicting
rainfall on a particular day.
A Comparative Analysis of Supervised Word Sense
Disambiguation in Information Retrieval
Chandrakala Arya1 Manoj Diwakar2 and Shobha Arya3
1Graphic Era Hill University, Uttarakhand, India
2Graphic Era (Deemed to be University), Uttarakhand, India
3Gurukul Kangri Vishwavidyalaya, Haridwar, Uttarakhand, India
As the amount of information increases every day there is a need of information
retrieval for finding useful information from large amount of information. This
paper present and evaluate supervised word sense disambiguation (WSD)
algorithms in Information Retrieval. Word ambiguity problem is a key issue in the
information retrieval systems. Supervised WSD is considered as the effective
method than other methods in information retrieval. The successful use of
supervised WSD in information retrieval is the main objectives of this research
-
26
paper. In this paper we describe the role of supervised WSD in information retrieval
and discuss about the best-known supervised algorithms in detail. We use WEKA
(Weikato Environment for Knowledge Analysis) tool to evaluate four Supervised
WSD algorithms Naïve Bayes, Support vector machine (SMO), Decision tree (J48),
and K-NN (IBK) through a series of experiments, the result conclude that
performance of the algorithms depends on the features of the datasets.
Real-Time Bangladeshi Currency Recognition Using
Faster R-CNN Approach for Visually Impaired People
Md. Tobibul Islam, Mohiuddin Ahmad, and Akash
Shingha Bappy
Khulna University of Engineering & Technology, Bangladesh
In this research work, we represent a new method for recognition of Bangladeshi
money using raspberry pi 4 and the Faster R-CNN approach. Here we represent a
microprocessor-based smart blind glass where one of the important features is
Bangladeshi money recognition. Money is a very important element for all the
people in the world including blind people. They cannot recognize the banknotes
without the help of others. Sometimes they cheat by wrong people in various money
issues. So, they suffer a lot in their day-to-day life. Through this paper, we introduce
a smart blind glass that can recognize money and provide audio information to the
blind man's ears. In this recognition, we introduced a new method faster R-CNN
approach which is widely used for object, and face recognition. By using a faster R-
CNN model in money dataset training we found about average 97.8% real-time
accuracy. Our recognition accuracy was high compared with other state-of-the-art
research work
Real time deep learning face mask detection model during
COVID-19
Amit Juyal1 and Aditya Joshi2
1Graphic Era Hill University, Dehradun, India
2Graphic Era (Deemed to be University), Dehradun, India
COVID-19 pandemic has affected the whole world not only physically but also
financially. Thousands of people have died due to this COVID-19 (Coronavirus)
epidemic and still the whole world is struggling to prevent this epidemic. Initially,
the government had only an option to stop this epidemic and that was to declare
complete lockdown. But during this time phase of lockdown many people became
unemployed. To stop this scenario and for the sake of employment and other
financial economy our government has decided to unlock various sectors. This
unlock phase has increased the risk of outbreak of COVID-19 so to avoid risk and
to stop spreading of this pandemic disease, the government has issued some
-
27
important guidelines to be followed mandatorily by each individual during this
unlock process. The face mask is one of the compulsory guidelines issued by the
government. Wearing a face mask may reduce the chances of spreading corona
virus. Many people roaming in public places without wearing a face mask. To
monitor such kind of people at a time is not easy manually. So in this paper we are
proposing deep learning based model that can automatically detects person wearing
a mask or not. In this method a model Convolutional Neural Network (CNN) was
trained over thousands of images to detect whether a person wearing a mask or not.
Proposed model achieved 93.36 validation accuracy and 98.71 training accuracy.
This model can be helpful to stop spreading of coronavirus in such organizations
where human interaction is necessary for smooth functioning like hospitals,
colleges, gyms, super markets etc.
Prediction of California Bearing Ratio of Subgrade Soils
using Artificial Neural Network Principles
T. V. Nagaraju1, R.Gobinath2, Paul Awoyera3 and Mohd
Abbas H. Abdy Sayyed4
1Department of civil engineering, S.R.K.R. Engineering College,
Bhimavaram, India
2Department of Civil engineering, SR Engineering College, Warangal,
India
3Department of civil engineering, Covenant University, Nigeria
4Department of civil engineering, SR University, Warangal, India
In geotechnical engineering, prediction of soil parameters is necessary due to the
large scale construction activities and time consuming testing. California Bearing
Ratio (CBR) is one of the soul parameter used as strength and stiffness indicator for
sub-grade soil. However, for investigating soil sub-grade in the field there is a need
of more soil samples to be tested, it may be time-consuming and cumbersome task.
Moreover, certain issues like lack of funding, unavailability of skilled labour and
poor lab infrastructure to handle large number of samples puts thrust on
development of models to predict strength with reference to certain amount of data.
Nowadays the potentiality of prediction models has been gaining importance in
every discipline. Numerous tools and techniques were evolved focusing on model
development; thanks to of increasing computing power researches are able to
perform iteration-based techniques. In this study, CBR values of sub grade along a
proposed road are collected. Nearly 480 samples were collected in which 15
Samples were used for comparison (Control Value). The results revealed that the
artificial neural networks (ANN) prediction models were significant promising tool
for predicting CBR of sub-grade soil by using index properties as input parameters.
-
28
Prediction of bearing fault detection using comparative
analysis of Random Forest, ANN and Autoencoder
methods
Pooja Kamat*, Pallavi Marni*, Lester Cardoz*, Arshan
Irani*, Anuj Gajula*, Akash Saha*,Satish Kumar**,Rekha
Sugandhi***
*Department of Computer Science and Information Technology,
Symbiosis Institute of Technology
**Department of Mechanical Engineering, Symbiosis Institute of
Technology
***Department of Information Technology, MIT School of Engineering
The manufacturing industry is currently witnessing a huge revolution in terms of the
Industry 4.0 paradigm, which aims to automate most of the manufacturing processes
from condition monitoring of the machinery to optimizing production efficiency
with automated robots and digital twins. One such valuable contribution of the
Industry 4.0 paradigm is the concept of Predictive Maintenance (PdM), which aims
to explore the contributions of Artificial Intelligence to get meaningful insights into
the health of the machinery to enable timely maintenance. As majority of these
machineries consist of bearings, bearing fault detection using Artificial Intelligence
has been a popular choice for researchers. This paper provides a systematic literature
survey of the existing research works in bearing fault detection. Further in this paper
we have done comparative analysis of bearing fault detection using the techniques
of Random Forest Classification, Artificial Neural Network and Autoencoder on the
benchmarked dataset provided by CWRU. The deep learning model of
Autoencoders provide the highest accuracy of 91% over the algorithms of Artificial
Neural Network and Random Forest.
Selection of a mesh network routing protocol for
underground mines
Prenern Reddy and Theo G. Swart
Center for Telecommunications, Department of Electrical & Electronic
Engineering Science, University of Johannesburg, Johannesburg, South
Africa
Research into determining the optimum routing protocol for underground mines was
conducted in a mock stope at the University of Johannesburg comparing
B.A.T.M.A.N., Babel and OLSR by transferring different types of network traffic.
It was found that the B.A.T.M.A.N. routing protocol offered the best performance
-
29
due to its layer 2 implementation, however OLSR a layer 3 routing protocol, was
also found to be an acceptable routing protocol. This may have been due to OLSR’s
ability to skip nodes when routing data. It was also found that the optimal internode
spacing was 60m as this allows for a reported -3dB RSSI enabling a good balance
between transfer speed and coverage distance.
Budget oriented reliable WDO Algorithm for Workow
Scheduling in Cloud Systems
Poonam Singh1, Maitreyee Dutta1, Naveen Aggarwal2
1NITTTR, Chandigarh India
2UIET,Panjab University Chandigarh India
The Use of Cloud computing for solving complex scientific workflow applications
has become very popular now a days. The cloud infrastructure is equipped with
elastic, heterogeneous and cost efficient resource provision features. It enhances the
computing by offering user an on demand services through Internet. Scheduling
these workflow applications under user defined budget constraint is one of the main
challenge in cloud environment. Also taking failure of computing resource into
consideration, it become imperative to generate reliable schedule while meeting
other scheduling attributes like cost, time and deadline. This paper presents a hybrid
wind driven optimization algorithm (HWDO) to generate reliable workflow
schedule by maintaining budget within specified limit. The algorithm is simulated
using WorkflowSim with real world scientific applications. The results achieved
substantiate 9-17% reliable schedule than other algorithms by meeting the budget
constraint.
An Energy Efficient Communication Scheme for Multi-
robot Coordination deployed for Search and Rescue
Operations
Rajesh M.1 and Nagaraja S.R.2
1Department of Computer Science and Engineering, Amrita School of
Engineering, Bengaluru, Amrita Vishwa Vidyapeetham, India
2Department of Mechanical Engineering, Amrita School of Engineering,
Bengaluru, Amrita Vishwa Vidyapeetham, India
Robots are assisting humans in various fields of life such as in Industry, Health care,
Defense, Entertainment etc. Rescue robotics is an evolving division of robotics
where robots are used to replace humans from hazardous situations to carry out
rescue operations. Examples of human fire fighters getting replaced by team of
robots to carry out search and rescue operation. The major challenge in the use of
-
30
multi-robot systems is the battery life of the robots. Efficient use of the battery
power is the essential task of robots. In a multi-robot system, coordination among
robots will be effective if there is proper communication scheme is available among
the robots. In this paper, an energy efficient communication scheme is proposed
which will avoid unnecessary messages being communicated among robots and
limit the message exchanges between interested parties only. Heterogeneous robots
will classify themselves into groups of robots with similar functionalities and they
will publish the details of services (data) which they can offer. Robots which need
those details will subscribe these groups and communication will be limited to
subscribed robots. This will reduce the number of messages exchanged in the system
and thus by reducing the use of energy for communication. Simulation of the
proposed system provides strong support for the claim that the proposed system is
much more effective than existing strategies in terms of energy efficiency. The
proposed system is able to reduce the energy consumption of the entire network by
around 24% compared to the traditional schemes.
Butterfly Optimization Algorithm Based Optimal Sizing
and Integration of Photovoltaic System in Multi-Lateral
Distribution Network for Interoperability
Thandava Krishna Sai Pandraju1, Varaprasad Janamala2
1Dept. of Electrical and Electronics Engineering, Dhanekula Institute of
Engineering & Tech-nology, Vijayawada – 521 139, Andhra Pradesh,
India
2Dept. of Electrical and Electronics Engineering, School of Engineering
and Technology, Christ (Deemed to be University), Bangalore – 560 074,
Karnataka, India
In this paper, a new and simple nature-inspired meta-heuristic search algorithm
namely butterfly optimization algorithm (BOA) is proposed for solving the optimal
location and sizing of solar photovoltaic (SPV) system. An objective function for
distribution loss minimization is formulated and minimized via optimally allocating
the SPV system on main feeder. At the first stage, the computational efficiency of
BOA is compared with various other similar works and highlighted its superiority
in terms of global solution. In second stage, the interoperability requirement of SPV
system while determining the location and size of SPV system among multiple
laterals in a distribution system is solved without compromises in radiality
constraint. Various case studies on standard IEEE 33-bus system have shown the
effectiveness of proposed concept of inter-line-photovoltaic (I-PV) system in
improving the distribution system performance in terms of reduced losses and
improved voltage profile via redistributing the feeder power flows effectively.
-
31
Document Classification in Robotic Process Automation
Using Artificial Intelligence – A Preliminary Literature
Review
Jorge Ribeiro, Rui Lima and Sara Paiva
Instituto Politécnico de Viana do Castelo, Portugal
In recent decades, combined with technological evolution, numerous operational
activities of companies are supported by information systems. Despite its
advantages, countless routine tasks of the organizations are done manually. In recent
years, Robotic Process Automation (RPA) has emerged allowing to create automatic
processes to deal with routine tasks. One typical feature of these systems is reading
documents via Optical Character Recognition (OCR) that are associated with the
classification of documents. This paper aims to present a general study on the
document classification process using OCR in RPA processes combined with the
application of Artificial Intelligence. It was intended to carry out a survey of the
state of the art of tools and approaches for the classification of documents using AI.
Conclusions show that despite the challenges associated with the classification and
categorization of documents, the applicability of AI techniques show good results
of accuracy to allow a better efficiency in the automation of RPA processes.
Suppliers Selection Using Fuzzy AHP and Fuzzy TOPSIS
Method - A Case Study of a Bearing Manufacturing
Company
Ramesh Karwal1, Pradeep Kumar2*, Manish Bhandari2,
M.L. Mittal3
1Water Resources Department, Rajasthan-323001, India. 2Department of Mechanical Engineering, MBM Engineering College, J.N.V
University, Jodhpur-342003, India. 3Department of Mechanical Engineering, MNIT, Jaipur-302017, India
Supplier selection, evaluation and monitoring are crucial for an industry to survive
in long term. Ranking of suppliers become complex when suppliers are evaluated
across multiple dimensions evaluation indices. The aim of this research Paper is to
determine the key factors of supplier selection and ranking of potential suppliers. A
bearing manufacturing Company was considering two criteria of suppliers selection
i.e. quality rating and service rating. In the current paper six criteria have been
considered instead of two for improving the supplier’s selection process which are
product quality, product cost, location, delivery time, information system and
service rating. First of all the key factors involved in supplier selection have been
identified, a survey has been conducted for data collection from purchase
department in the company. Fuzzy AHP method and fuzzy TOPSIS method are used
-
32
to calculate the criteria weights for the suppliers’ selection and to determine the
ranking of the suppliers. The rating has been represented by linguistic variables and
then parameterized by triangular fuzzy number. The contribution of this study is to
give improved suppliers’ selection process to the company.
Artificial Intelligence Optimization Strategies for Invoice
Management: A Preliminary Study
Rui Lima, Sara Paiva and Jorge Ribeiro
Instituto Politécnico de Viana do Castelo, Portugal
It is very common for companies to receive invoices (and other semi structured
documents) in paper and PDF files and someone have to manually enter that data
into a digital structure like a database or comma-separated values (CSV) file. This
type of work is very time consuming (making it expensive) and exhaustive (making
it prone to errors). Data entry activities also forces high paying specialized workers
to do repetitive tasks or to outsource that work, making it hard to manage the data
workflow. There is a need to automate this type of processes. In this context, the
following paper presents a preliminary study and review of technologies, tools and
recent research strategies for invoice management mainly in the scope of robotic
process automation tools.
Survey Analysis for Medical Image Compression
Techniques
Baidaa A. Al-Salamee and Dhiah Al-Shammary
College of Computer Science and Information Technology, University of Al-
Qadisiyah, Iraq
This paper presents a survey for medical image compression for both lossy and
lossless approaches. This survey discusses twenty-five publications with several
applied lossy and lossless compression techniques. All approaches are distributed
into seven groups based on the applied technique. Fractals, Wavelet, Region of
Interest (ROI) and Non-Region of Interest (Non-ROI), and other approaches
represent four lossy compression groups. Adaptive Block Size, Least Square, and
other approaches represent three lossless medical image compression techniques.
Technically, medical images communication requires large space to be represented
and sent over the network creating a number of challenges in terms of interaction,
processing, storage and transmission operations. Therefore, Significant
Compression Ratio (CR) and PSNR (Peak-Signal-to-Noise-Ratio) are always
targeted in image communication. Both CR and PSNR are considered in this survey
as the main metrics to investigate and evaluate models performance. As a result of
this survey analysis, ROI and Non-ROI has shown the best average CR with 91 and
Wavelet has shown the best average PSNR with 80.
-
33
A Comparative Study between Data-Based Approaches
under Earlier Failure Detection
Hadjidj Nadjiha, Benbrahim Meriem, Berghout Tarek,
Mouss Leila Hayet
Laboratory of Automation and Manufacturing Engineering
University of Batna2, 01 Rue Chahid Bokhlouf, Batna, 05000, Algeria,
A comparative study between a set of chosen machine learning tools for direct
remaining useful life prediction is presented in this work. The main objective of this
study is to select the appropriate prediction tool for health estimation of aircraft
engines for future uses. The training algorithms are evaluated using “time-varying”
data retrieved from C-MAPSS (Commercial Modular Aero-Propulsion System
Simulation) developed by NASA. The training and testing processes of each
algorithm are carried out under the same circumstances using the similar initial
condition and evaluation sets. The results prove that among the studied training
tools, SVM (Support Vector Machine) achieved the best results.
Classification of fundus images based on non-binary
patterns for the automated screening of retinal lesions
Mekhana Suresh, Sreelekshmi Indira, and Sivakumar
Ramachandran
Department of Electronics and Communication Engineering, College of Engineering
Trivandrum, Thiruvananthapuram, Kerala, 695016, India
The prevalence of lesions in the human retina has increased many folds during the
past few decades. The two major causes of retinal lesions that affect the visual
system are diabetic retinopathy (DR) and age-related macular degeneration (AMD).
Digital retinal images obtained from the fundus camera are typically used for lesion-
screening. Due to the large affected population, manual screening is not a feasible
solution for the early diagnosis of the disease. Hence, there exists a high demand for
automated computer-aided screening systems that help clinicians to handle the
enormous image data. The proposed screening technique relies on the texture
analysis of the retinal background using local ternary patterns (LTP). Also, we
compared the results obtained using the proposed approach with local binary
patterns (LBP) instead of LTP. Three experiments separating, DR from normal,
AMD from normal, and DR from AMD are conducted and tested using the proposed
pipeline. The feature vectors generated from the proposed technique are analyzed
using various classifiers and the discriminating capabilities of each classifier is
reported quantitatively. The results obtained show the effectiveness of LTP in
analyzing the retinal texture for diagnosing various lesions.
-
34
Performance Evaluation of SEIG Under Unbalanced
Load Operations Using Genetic Algorithm
Yatender Chaturvedi, Varun Gupta Arunesh Chandra,
Ankit Goel
KIET Group of Institutions, Ghaziabad, Uttar Pradesh-201206
In the current world, superiority of non-conventional over conventional power
sources are rapidly increasing in the area of powerful production. In this paper the
modeling of three-phase induction generator was done which turns into an objective
function that has been solved for their variables using genetic algorithm as
optimization tool in MATLAB. The program has been written in MATLAB and
being called in GA optimization window by setting limits to its variables. The
methodology as discussed has been adopted to analyze the performance of a 15 KW
induction machine in terms of voltage unbalanced factor, power output, losses and
efficiency of machine under different power factor loading.
A New Approach to Classify the Boolean Functions Based
on Heuristic Technique
Rajni Goyal and Harshit Grewal
Amity University, Noida UP 201313, India
Classification of Boolean functions is still remains an open problem for theoretical
cryptographers. In this paper, one method is introduced for systematic classification
of Boolean functions for n-variables. Our Classification method is an evolutionary
approach based on nonlinearity. In the method, we started with 1-variable functions
and found the classification formula for n-variables.
A Modulo (2n-2n-2-1) Adder Design
Ahmad Hiasat
Princess Sumaya University for Technology, Amman, Jordan
This paper presents a modulo (2n-2n-2-1) adder. The proposed structure uses a
parallel-prefix binary adder with an end-around carry as its skeleton. It splits the
parallel-prefix phase into lower (n-2)-bit part and upper 2-bit part. The proposed
structure requires the area and delay of a regular parallel-prefix binary adder with
an end-around input carry. Compared with functionally-similar modulo (2n-2n-2-1)
using VLSI synthesis tools, the suggested adder reduces area by (14.2 – 72.2)%,
time by (5.1-19.5)%, and area-time product by (36.5-92.0)%.
-
35
Influence of Object-Oriented Software Design Measures
on Reliability: Fuzzy Inference System Perspective
Syed Wajahat Abbas Rizvi
Department of Computer Science and Engineering, Amity University, Uttar Pradesh, Lucknow,
India
Looking at the lifestyle that every one of us is living, it can be easily noticed that
the impact of information technology is growing exponentially. Whether it is urban
or rural area everyone is heavily dependent on software directly or indirectly. This
dependence on software has been creating a pressure on the information technology
industry to meet this exponential demand. At the same time various software quality
attribute have been gaining its importance. The roots of a reliable software lie in the
careful and informed implementation of design stage. Therefore, author has focused
on the design stage and identified some object-oriented design measures and further
analyzed the impact of these measures on the identified quality attribute that is
reliability of application software using the fuzzy inference system.
Entity Based Knowledge Graph Information Retrieval for
Biomedical Articles
Vikash Kumar Prasad, Shashvat Bharti, Nishanth Koganti
GEP Worldwide Inc.
In this paper, we present an information retrieval system on a corpus of scientific
articles related to COVID-19 and biomedical. We build a heterogeneous entity-
based knowledge graph network, where edges are shared between biomedical
entities and paper names, where entities appear in abstract of the paper. The
biomedical entities are derived from the abstract of the scientific articles using a
fine-tuned Bio-BERT model. For a user query, entities are derived using a fine-
tuned Bio-BERT model and then semantic similarity to query is employed for the
return of the top-most relevant papers on the titles. We also provide a small set of
results for the information retrieval system.
Test case prioritization based on Requirement
Amrita1 and Prateek Gupta2
1Banasthali Vidyapith, Rajasthan, India, 2UPES, Dehradun, India
In the software development process, the main concern of the developer is to
develop a quality product and optimizing time and cost. Maintenance of the software
is considered a costly activity. Regression testing is performed to test the modified
software in the maintenance phase. In order to minimize the total time and cost, we
can arrange test cases according to the different requirements proposed in the
-
36
starting phase of the development. We can prioritize the test cases in such a manner
that the higher priority test cases will execute before the lower priority test cases.
This paper presents an approach to prioritize regression test cases based on
requirements. Firstly, we have selected some requirement factors; based on these
requirements, the weight for each specific requirement will be calculated. Further,
we will map the test cases to the requirements and calculate a parameter based on
the mapping, known as Test Case Weight (TCW) to the set of requirements. The
higher TCW will be executed first. We have compared our proposed work with other
prioritization techniques and found better results.
Mining and Predicting No-Show Medical Appointments:
Using Hybrid Sampling Technique
Albtool Alaidah1, Eman Alamoudi2, Dauaa Shalabi1, Malak
AlQahtani1, Hajar Alnamshan1, and Nirase Fathima Abubacker1
1School of Computing, Dublin City University, Dublin, Ireland 2Taif University, Saudi Arabia. School of Computing, Dublin City University, Dublin, Ireland
Clinics use scheduling systems for patients’ appointments. However, no-shows are
frequent in both general medical practices and specialties, and they can be quite
costly and disruptive. This problem has become more severe because of COVID-
19. The primary purpose of this study is to develop machine learning algorithms to
predict if patients will keep their next appointment, which would help with
rescheduling appointments. The main objective in addressing the no-show problem
is to reduce the false negative rate (i.e. Type II error). That occurs when the model
incorrectly predicts the patients will show up for an appointment, but they do not.
Moreover, the dataset encounters an imbalance issue, and this paper addresses that
issue with a new and effective hybrid sampling method: ALL K-NN and Adaptive
Synthetic (ADASYN) yield a 0% false negative rate through machine learning
models. This paper also investigates the leading factors that affect the no-show rates
for different specialties. The SHapley Additive exPlanation (SHAP) method reveals
several patterns to identify the target feature (patient no-shows). It determined that
a patient’s history of missed appointments was one of the leading indicators. It was
also found that greater lead times between booking the appointment and the
appointment date was associated with more no-show behaviour.
Adaptive Strategy for Environment Exploration in Search
and Rescue Missions by Autonomous Robot
Rokas Semenas and Romualdas Bausys
Department of Graphical Systems, Vilnius Gediminas Technical University, Vilnius, Lithuania
In this research, a new adaptive strategy is proposed for the autonomous mobile
robot, which explores the unknown search and rescue (SAR) environment. The
robot, which implements the proposed strategy, operates on the frontier-based
-
37
exploration approach and makes a decision of where to move next by applying a
total of eight new strategies for candidate frontier assessment. The fuzzy logic
controller is applied to determine the most suitable candidate frontier assessment
strategy regarding the current robot state and the discovered environment
information. The final decision of where to move next is made by the neutrosophic
interval-valued multi-criteria decision-making method, namely, WASPAS-IVNS,
which enables the modelling of vagueness present in the initial sensor data. The
proposed adaptive strategy is tested in the virtual Gazebo simulation. The obtained
test results show the increased efficiency when comparing the proposed adaptive
environment exploration strategy to the static environment exploration strategies
and the standard greedy environment exploration approach.
Investigating the Effect of Lockdown during COVID-19
on Land Surface Temperature Using Machine Learning
Technique by Google earth Engine: Analysis of
Rajasthan, India
Amita Jangid, Mukesh Kumar Gupta
Swami Keshvanand Institute of Technology Management & Gramothan, Jaipur Rajasthan, India
The COVID-19 in India is part of the global coronavirus pandemic (COVID- 19)
caused by severe acute breathing syndrome 2 (SERS-CoV-2). The first case of
COVID-19 in India, which originated from China, was reported on 30 January 2020.
This study analyzes the effects of Lockdown during COVID19 on Land Surface
Tempreature for the six categories of water, wetland, bare land, forest, cropland, and
urban. It is essential to examine the mean LST differences for each land cover type.
This study uses the SR data from Landset8. All Landsat level 1 and level 2 data is
directly available to Google Earth Engine, Including TOA(top of atmosphere) and
SR (surface Reflectance). The process is a comparative analysis, so data of the same
periods are analyzed for 2019 before lockdown and 2020. There are significant
changes that have been seen in Land surface temperature. Therefore, it is essential
to incorporate an investigation regarding LST differences for each land cover type
in various anthropogenic levels. So our results show mean LST differences between
during and before the emergence of COVID-19 for each land cover type regarding
lockdown policy in Rajasthan, India.
Human activity recognition using deep learning based
approach
Maruf Rahman and Tanuja Das
Department of Information Technology, GUIST, Guwahati, Assam, India
Human activity recognition (HAR) has a considerable demand in the area of human
-
38
perception and also encompasses many other purposes like healthcare monitoring,
assisted living for elders, and intelligent video surveillance. Many machine learning-
based approaches have been used for the objective of activity recognition, but these
techniques depend heavily on hand-crafted feature extraction which is unable to
perform well when dealing with complex scenarios. Deep learning techniques have
great potential for human activity recognition. This paper presents a neural network-
based approach in which a convolutional neural network (CNN) and a long short-
term memory (LSTM) are put together to classify and analyze human activities. The
CNN-LSTM model was applied on the UCI-HAR data-set for classifying the six
human activities, viz., walking, walking-upstairs, walking-downstairs, sitting,
standing, and laying. Results show that the proposed model is very efficient for
recognizing human activity.
Emotion distribution profile for Movies Recommender
systems
Mala Saraswat1 and Shampa Chakraverty2
1ABES Engineering College, Ghaziabad, U.P 2Netaji Subhas University of Technology, Delhi
Reviews, comments and feedbacks are user generated content which comprises of
insights regarding a given item or a thing and furthermore user’ emotions. Various
highlights of users created content incorporates feelings, opinions, survey
helpfulness that shows a promising research in the field of recommender systems.
Reviews contain various words and sentences that show their natural passionate
substance. Emotions are a significant component of human conduct. They enable us
for decision making by generating a liking or disliking towards a particular item.
This paper harnesses reviews as the content generated from user to exploit, emotion
as a basis for generating recommendations. Through experiments conducted on real
dataset, our proposed approach compares the performance with traditional item
based collaborative filtering approach. Experimental results show 173% increase in
prediction accuracy for top 25 recommendations as compared to prediction accuracy
based on rating based item similarity.
Prediction of Modulus of Subgrade Reaction using
Machine Language Framework
K. S. Grover, Jitendra Khatti and Amit Kumar Jangid
Rajasthan Technical University, Kota 324010, Rajasthan, India
The modulus of subgrade reaction test is also known as 𝑘𝑠 value test and it is essentially a plate bearing test. This test is generally used in the design of rigid
pavements and raft foundations. In the present research work, the modulus of
subgrade reaction has been predicted by an artificial neural network and principle
component analysis. For the prediction of 𝑘𝑠 value, the neural network (NN) models
-
39
of the different number of hidden layers and nodes have been developed in
MATLAB R2016b. The range of the number of hidden layers have been selected
from one to five and for each hidden layer, the range of the number of nodes have
been selected from two to eleven. Based on the training and validation performance
of NN models, the best architectural neural network model is selected. In the present
work, the neural network model of two hidden layers with eleven nodes on each
hidden layer has been selected as the best architectural neural network. The best
architectural neural network has been compared with principle component analysis.
From the comparison, it has been concluded that the principle component analysis
has predicted modulus of subgrade (𝑘𝑠 value) with 86.4% accuracy, which is ≈1.11 (1.105) times less than the accuracy of the neural network model.
Time Fractionalized Lattice Boltzmann Model Based
Image Denoising
P. Upadhyay1 and K.N. Rai2
1DST-CIMS Banaras Hindu University, Varanasi Uttar Pradesh India 221005 2Department of Mathematical Sciences IIT(B.H.U.) & DST-CIMS Banaras Hindu University,
Varanasi Uttar Pradesh India 221005
In this paper, we have proposed the Caputo sense time fractionalized lattice
Boltzmann (LB) model for time sub diffusion equation in Caputo sense recently
proposed by Rui Du et al. We have applied this model to denoise white Gaussian
noise corrupted images with different noise levels. The denoisng results have been
compared with some of the recently reported state of the art methods in terms of
PSNR and SSIM values. The results found, show considerable improvement over
other recently reported state of the art methods in terms of both PSNR and SSIM
values.
Distributed and Anonymous E-voting using Blockchain
and Ring Signatures
Nishay Madhani, Vikrant Gajria, and Pratik Kanani
Dwarkadas J. Sanghvi College of Engineering, Mumbai 400056, Maharashtra, India,
Digitization of bureaucratic processes has been a long suited method for the growth
of a country. E-Voting (Electronic Voting) is a method of casting, tallying &
verifying votes from citizens using electronic means. With the advent and boom of
Blockchain and Distributed Computing, many methods to improve E-voting have
come forward that allows us to assure the privacy of voters and security against voter
fraud. In this paper, we analyze the different types of Blockchains and which ones
would be suitable for these purposes. In the end, we propose a method using
Blockchain and Linkable Ring-Signature that provides the privacy of voters in a
large-scale election by deploying smart contracts for \rings" of registered voters and
analyze its performance on consumer devices.
-
40
Enyo: A Multistage Partition and Transposition based
cipher
Apratim Shukla, Mayank K Tolani, Dipan Polley, Abhishek
Thazhethe Kalathil, N. Subhashin