Cyber-Physical-Social-Thinking Modeling and...
Transcript of Cyber-Physical-Social-Thinking Modeling and...
Cyber-Physical-Social-Thinking Modeling and Computing forGeological Information Service System
Yueqin Zhu∗†, Yongjie Tan∗†, Ruixin Li†‡§, Xiong Luo†‡§∗Development and Research Center, China Geological Survey, Beijing 100037, China.
†Key Laboratory of Geological Information Technology,Ministry of Land and Resources, Beijing 100037, China.‡School of Computer and Communication Engineering,
University of Science and Technology Beijing (USTB), Beijing 100083, China.§Beijing Key Laboratory of Knowledge Engineering for Materials Science, Beijing 100083, China.
Abstract:The serious geological hazards occurred frequently in the last few years. They haveinflicted heavy casualties and property losses. Hence, it is necessary to design a geological in-formation service system to analyze and evaluate geological hazards. With the development ofcomputer and Internet service model, it is now possible to obtain rich data and process the datawith some advanced computing techniques under network environment. Then, some technologies,including cyber-physical system (CPS), Internet of Things (IoT), and cloud computing, have beenused in geological information management. Furthermore, the concept of cyber-physical-social-thinking (CPST) as a broader vision of the IoT was presented through the fusion of those advancedcomputing technologies. Motivated by it, in this article a novel modeling and computing method forgeological information service system is developed in consideration of the complex data processingrequirement of geological service under dynamic environment. Specifically, some key techniquesof modeling the information service system and computing geological data via CPS and IoT areanalyzed. Moreover, to show the efficiency of proposed method, two application cases are providedduring the CPST modeling and computing for geological information service system.
Keywords:cyber-physical-social-thinking (CPST); Internet of Things (IoT); geological data pro-cessing; cloud computing.
I. INTRODUCTION
Recently, a geological hazard as one of the most important issues presents an obstacle to the social devel-
opment. It is necessary to design a geological information service system to analyze and evaluate geological
hazards. Among the available geological information service applications, a geographic information system
(GIS) and its location intelligence play an important role in analyzing and visualizing geological data for
the public. Furthermore, with the continuous improvement of Internet technology, the concept of cyber-
physical system (CPS) gradually appears in front of everyone [1], [2], [3]. CPS is a hyperspace system, in
which comprehensive computing, networking, and physical environment can be integrated. Meanwhile, CPS
can also achieve real-time sensing, dynamic control, and information services for large-scale engineering
Corresponding authors: [email protected] (Yueqin Zhu); [email protected] (Xiong Luo)This work was partially supported by the Special Funds Project for Scientific Research of Public Welfare Industry from the
Ministry of Land and Resources of China under Grant 201511079, and the National Key Technologies R&D Program of Chinaunder Grant 2015BAK38B01.
systems. Specifically, some techniques related to CPS, e.g., Internet-service-based multi-user accessing and
data service, have been developed in geological information service system (GISS) [4]. As an effective
architecture with the potential powerful capability of addressing complex issues, there are some research
topics to explore along this direction for GISS. Beyond the scope of the CPS, the cyber-physical society
system (CPSS) is developed. In addition to the cyber space and the physical space, it is deployed with
emphasis on humans, knowledge, society, and culture. Hence, it can connect nature, cyber space, and society
with certain rules [5].
With the development of computer and Internet service model, the Internet of Things (IoT) is becoming
increasingly common. Under the IoT architecture, objects can be sensed and controlled remotely across
existing network structure, while integrating the physical world and computer-based systems [27], [7],
[8]. Consequently, the efficiency, accuracy, and economic benefit are all improved. Meanwhile, the cloud
computing technology plays a prominent part in the IoT while making it easier for users to store and process
information. Furthermore, it also improves the computing performance by making the information available
over the web or other terminal ends. Hence, cloud computing is so popular in recent years due to its technical
benefits of the on-demand capacity management model used to efficiently share the resources, software, and
information among multiple devices [9].
More recently, through the fusion of CPS and IoT on the basis of cloud computing technology, a concept
cyber-physical-social-thinking (CPST) as a broader vision of the IoT was presented by merging the thinking
space into the CPS and highlighting the importance of the cognitive intelligence and social organization [10].
Generally, the paradigms of artificial intelligence and cognitive computing have been widely applied to many
fields. Those emerging technologies therefore have brought new development opportunity for IoT. In this
way, both the human cognition capacities and intelligent learning techniques from the society activities
are integrated into the implementation of IoT. Considering it, through the extension and mergence of
thinking space within the traditional CPS space, the CPST is accordingly developed to address more complex
application requirements in practice. Motivated by it, in this article a novel modeling and computing method
for GISS is developed. Considering the complex data processing requirement of geological service under
dynamic environment, it imposes very challenging obstacles to the implementation of GISS with the help
of those current computing technologies. Hence, on the basis of some successful applications of CPS in
GISS, CPST may play an important role in dealing with those difficult issues of designing an effective GIS
system.
The remainder of this article is organized as follows. Section 2 gives a description for related works,
including CPS, CPSS, CPST, IoT, and cloud computing. Section 3 and Section 4 respectively present
the CPST modeling and computing schemes for GISS while providing some application cases. Finally,
a conclusion and brief discussion regarding the open challenging science and technology issues along this
research direction in GISS are discussed in Section 5.
2
II. RELATED WORKS
A. CPS, CPSS, and CPST
1) CPS: CPS is a networked, component-based, real-time system that controls and monitors the physical
world [11]. From the structural point of view, a CPS includes the following parts: i) sensors designed to
perceive the physical world; ii) controllers or actuators designed to operate the physical entity; iii) computing
components designed to analyze and process physical information; iv) communication network designed to
integrate the above units and related information, objects, and events.
Currently, with the rise in popularity of smart phones, there is a growing interest in the mobile application
for CPS. Then, mobile cyber-physical system (MCPS) as a prominent subcategory of CPS is developed, in
which the physical system has inherent mobility [12].
2) CPSS: Generally speaking, the society environment is considered an autonomous, living, sustainable,
and intelligent variable within the CPS. As a result, the CPSS gathers and organizes resources into se-
mantically rich forms that both machines and people can easily use [13]. Such a system includes globally
distributed resources, including devices, information, and knowledge. In addition, it also includes services
that could dynamically collaborate to provide some effective just-in-time services to cope with an emerging
crisis.
The key issue of designing a CPSS is how to describe and integrate the interaction of physical, social,
and hardware in an efficient way. Some methods are presented. To improve computational and networking
contextual complexity in deploying a CPSS, a smart services framework in CPSS, called dynamic social
structure of things, was developed aiming at boosting sociality and narrowing down the contextual complexity
based on situational awareness [14]. In one sense, CPSS can promote the interconnection of distributed CPS.
3) CPST: CPST hyperspace is accordingly established through the mergence of a new dimension of
thinking space into the CPS space [10]. Therefore, the wisdom along with the intelligent interactions of
data, information, and knowledge are through the CPST hyperspace. Currently, the hyperworld is integrated
into CPST hyperspace by coupling data, information, knowledge, and cognition related cyber interactions,
physical perceptions, social correlations, and human thinking. The CPST hyperspace has the following
characteristics.
• Integration. The CPST hyperspace may be a combination of different spaces.
• Interconnection. The CPST hyperspace is connected with other same type or different type of spaces.
• Interaction. The CPST hyperspace can exchange information through cyber spaces to achieve ubiq-
uitous intelligence.
B. Cloud Computing
Cloud computing is a computing style in which dynamically scalable and virtualized resources are provided
as a service over the Internet [9]. Then, the traditional system architecture should be extended to meet the
requirement of software development in cloud computing scheme. Under the cloud computing framework,
users can employ a variety of terminal access services in any positions. The requested resources from the
3
cloud is virtual, rather than a fixed physical entity. Then, the user may enjoy all kinds of super services
through Internet [15]. Due to its automatic management feature, cloud computing is cost-effective while
greatly reducing the cost for data center management.
Cloud service provides some solutions or real-time software services through Internet. Existing cloud
service can be categorized into three major groups: infrastructure-as-a-service (IaaS), platform-as-a-service
(PaaS), and software-as-a-service (SaaS) [16].
C. IoT
Currently, the IoT achieves a great success with the development of wireless technology [17]. Among the
available key technologies in the IoT, the radio frequency identification (RFID) and wireless sensor network
(WSN) play an important role in implementation applications. Through the use of RFID technology, there
are many real-time monitoring applications in the IoT.
Meanwhile, a WSN uses spatially distributed autonomous sensors to monitor physical or environmental
conditions [18] and to cooperatively pass their data through the network to a main location. Today such
network is used widely in many industrial and consumer applications, such as industrial process monitoring
and control, machine health monitoring, and so on [19]. Moreover, with the development of high performance
computing technologies, the IoT becomes an attractive system paradigm to realize universal interactions
through the cloud computing [20].
III. CYBER-PHYSICAL-SOCIAL-THINKING MODELING FOR GEOLOGICAL INFORMATION SERVICE
SYSTEM
A. Features of Geological Data
In the era of big data, data is being collected to serve the purpose of making decisions in a data-driven
way. Then, such big data analysis now drives nearly every aspect of society, including financial services,
manufacturing, and mobile services.
Emergency data is an important resource from geological disaster emergency command system. And
those data in the emergency command and management of the geological disasters include three types,
i.e., emergency event data, emergency service data, and emergency basic data. Here, emergency basic data
include geographic information, document, knowledge, and many others. Emergency event data is the basic
information of the incidents, including video, audio, and other environmental information. Emergency service
data is in the process of emergency management, and it is generated after analyzing and making decision.
Recently, big data is revolutionizing all aspects of society ranging from science to government. Geological
data is a typical case of big data. Geological big data include geological data produced by multi-source and
multi-mode from geological survey as well as data from public service and management support [21].
Actually, geological data has three characters, i.e., rich sources, wide area, and great capacity of data [22].
So organization and management of those data is a key issue while realizing GISS. Geological data is a series
of important results in geological work, including geological documents, electronic data, specimens, and so
on. Then, it is obtained from the geological scientific research, engineering exploration, and production
4
Fig. 1. CPST model of geological information service system.
process. Essentially, the geological data is considered the outcome of the wisdom of experts in geological
areas. Through the use of geological data with an effective scheme during the collecting, integrating, sharing,
and developing processes, the public can obtain an orderly and long-term service.
Practically speaking, geological data is mainly composed of structured and unstructured data. And the
unstructured data, including reports and files generated by word, PDF, excel, graphics, video, and PPT,
is all stored in a traditional mode. Since it is inefficient with traditional storage mode while conducting
data retrieval, data querying, statistic analysis, and data mining, thus resulting in extremely low data service
capability [23]. Hence, it is urgent that geological big data should be addressed with a new modeling scheme
by integrating some of the technologies presented in the state-of-the-art analysis. And the CPST modeling
scheme developed here may be a competitive choice.
B. System Modeling
As mentioned above, GISS confronts some challenging issues, which requires the GISS should address the
geological big data with a more effective mode during data acquisition, search, analysis, and visualization.
Actually, although the diversity and completeness of source data are strengthened during the geological
data acquisition, it makes data analysis difficult due to the huge amounts of geological data. Then, it is
necessary to make a synthesis of all data through the combination of users requirement, geological context
knowledge, and application background. During this process, it hardly achieves such object without the help
of intelligent thinking and learning capabilities from human. As a result, CPST modeling scheme may be
a good choice and is accordingly employed to GISS on the basis of its intelligence and social computing
paradigm.
The CPST includes the cyber space, physical space, social space, and thinking space. Fig. 1 shows the
GISS and its key components involving the cyber, physical, social, and thinking modules described by CPST.
The functions of every part are given as follows, and the main equipments and technologies are presented
in Fig. 2.
5
Fig. 2. Key equipments and technologies.
1) Cyber Space: In the cyber space, it refers to the generalized information resources, including virtual
and digital abstractions to achieve interconnections among cyber entities [10]. In this space, a cognitive
information framework is designed to provide the geological data, resource, and service in an intelligent
self-adaptive learning way while implementing the cyber interactions. Considering the big data characteristics
in geological data, it should provide some effective data services, including holistic data management for
massive geological data storage, on-demand geological resource management in virtual cloud computing
environment, and online spontaneous management during distributed interactions for geological data.
2) Physical Space: Physical space represents the real world in which geological objects are respectively
perceived and controlled by semantic sensors or cooperative actuators to collect the real-time data to achieve
interactions remote collaboration through the use of the communication channels in context-aware networks.
Here, in order to realize interconnections in the cyber space mentioned above, a geological object should
be transformed into single or multiple cyber entities under virtual working framework.
3) Social Space: Social space is designed as a logical component in CPST modeling scheme. It is
responsible for collecting those social attributes and social relationships from the human beings and other
geological objects or cyber entities. Firstly, in the social space those information are from human activities
and social events, including supervision, organization, and coordination addressed by geological experts and
public users. Secondly, the inherent relationships among the geological objects, including direct and indirect
correlations, are also integrated into the social space in a semantic representation way.
4) Thinking Space: Within CPST, the thinking space plays a special role in addresses a high-level thought
or idea raised during the intellectual activities of geological experts and public users. Specifically, those
thinking information can be obtained through a self-adaptive learning way, e.g., analysis, synthesis, judgment,
reasoning, and some computational intelligence-based learning algorithms, e.g., the popular deep learning
method [24], reinforcement learning method [25], and many other state-of-the-art techniques.
6
Fig. 3. Architecture of geological information service system.
5) System Architecture: Considering the background of geological service and complex features of
geological big data, a system architecture for GISS is designed on the basis of the modeling idea of CPST.
Here, the administration can use web and mobile devices in this system to collect and manage the geological
data. Due to some limitations of network technology, it is not possible to connect all the devices directly to
the next generation network. Then, to make full use of geological data through some technologies of CPS
and IoT, a novel information platform system framework for geological data service is shown in Fig. 3.
This system architecture is composed of three layers, including application layer, network layer, and
perceptual layer. Meanwhile, some specific functions are implemented within this architecture. Here, the
network layer focus on the interconnection and interoperability of equipments, and it is specialized for
data transmission and resource sharing while ensuring that the next generation of network characteristics is
effectively integrated in the transmission processing. The network layer aims at transmitting various real-
time and un-real-time geological data for users through the special-purpose connection network. In the
proposed architecture, the perceptual layer is a front-end of information collection. The application layer is
the business logic layer in the integrated information platform, which provides some business to meet the
geological information service needs of different users.
6) An Application Case: Here, a simple case regarding the geological data analysis in thinking space is
provided under the CAST framework.
Currently, there are some ways used to help us to make decisions, such as mapping knowledge domain
and semantic relation analysis. Here, the term mapping knowledge domains was chosen to describe a newly
evolving interdisciplinary area of science during the process of charting, mining, analyzing, sorting, and
displaying knowledge. Scientists, academics, and librarians have historically worked to codify, classify, and
7
organize knowledge, thereby making it useful and accessible [26]. In addition, perhaps even more important,
the new analysis techniques that are being developed to process extremely large databases give promise of
revealing implicit knowledge that is presently known only to domain experts.
Some of these techniques are now being applied, aiming to identify and organize research areas according
to experts, publications, text, and figures; discover interconnections among these; establish the import of
research; reveal the export of research among fields; examine dynamic changes such as speed of growth and
diversification; highlight key factors in information production and dissemination; find and map scientific
and social networks. The new techniques support and complement human judgment.
Geological data is a kind of expression form of geological significance in the long history of the earth. The
traditional geological data includes geological map, structural map, mineral map, geological hazard map,
and rock phase diagram, and many others. Classification of rocks and lithofacies is an important part of it.
Here, an example of a rock classification and lithofacies relationship is provided. And the user interface is
demonstrated in Chinese.
Firstly, the statistics and classification of the existing geological data text are sorted, including the size
and type of data, the document number, the paragraph number, and numerical data type.
Then, semantic relation analysis is conducted to identify and extract subjective information by the natural
language processing (NLP), text analysis, and other machine learning methods [27]. In this example,
the hidden-Markov-model (HMM)-Viterbi-based standard word segmentation is employed to segment the
document, so that the text can be divided into different parts. Then, the TextRank algorithm is applied to
filter out the those words, such as the irrelevant adverbs and adjectives behind the participle, so as to achieve
the effect of keyword extraction.
The result of word segmentation is accordingly analyzed, and the relationship between words is shown
through the use of a visual programming tool, which stores the structured data on the web instead of the
table. Specifically, through use of Neo4j map database tool, geological information processing and retrieval
are implemented. And a good semantic knowledge mapping is thus developed to show the magmatic rock
classification and attribute better. A demonstration result is shown in Fig. 4.
In this figure, the nodes represent entities, and the edges represent the relationships between entities.
And different classes of entities are distinguished by different colors, and each one has a specific semantic
meaning. Here, the semantic meaning denotes classification and attribute relationships. The former includes
the shape classification (i.e., plutonic intrusive rock, hypabyssal intrusive rock, and extrusive rock) and the
acidity classification (i.e., ultrabasic rock, mafic rock, intermediate rock, and acidic rock). The latter includes
the color, mineral component, accessory mineral, structure, alteration, alkalinity, and SiO2 content. From
Fig. 4, the hierarchy result of classification for magmatic rock can be clearly found with a strong visual
effect. Furthermore, it can help the user to find more accurate information, make a more comprehensive
summary, and provide more depth information. Meanwhile, it also can systematically display the keywords
and the related knowledge system. Complex areas of knowledge are shown through data mining, information
processing, knowledge measurement, and graphics rendering. Hence, it provides a practical and valuable
8
Fig. 4. Relational mapping of geological information.
reference for the geological research and decision.
IV. CYBER-PHYSICAL-SOCIAL-THINKING COMPUTING FOR GEOLOGICAL INFORMATION SERVICE
SYSTEM
Computing technologies play an elemental role in facilitating the intelligent circumstance monitoring,
semantic analysis, social behavior modeling, and insight generation. In our developed system, computing
technologies include traditional data computing and the CPST computing. Specifically, the security of the
system is inevitable to be considered in the computing technologies.
A. The Computing for Geological data
1) Management of Distributed Database: The management on distributed database is one of the important
issues in the geological service system. To ensure the high reliability of data, distributed database often
employs backup strategy to achieve fault tolerance. Therefore, when reading data, the client can concurrently
read simultaneously from multiple backup server, so as to improve the speed of data access and provide
higher concurrent traffic.
2) Data Processing and Analysis: Generally, the collected geological information data is abundant, and
those data can be fused to analyze and evaluate geologic events. The issue on how to preprocess data to
improve the data analysis quality is very important. Data processing extracts valuable information from large
amounts of data. There exists many methods to process and analyze data [12]. In the practical applications,
the combination of two or more methods is implemented to achieve satisfactory data processing and analysis
performance. Fig. 5 shows a typical distributed implementation framework for geological data processing and
sharing. In this figure, it can be observed that in the cloud-based distributed computing environment, many
data sources are organized to form different databases through the data sharing and exchange specification.
Furthermore, the advanced data processing, exchanging, and presentation services are provided on the basis
of those databases.
9
Fig. 5. Framework of geological data sharing and processing.
B. The CPST Computing
1) A Multi-level Computing Framework: The CPS computing has been designed to support human-centric
paradigms, and address data and information to provide contextually relevant knowledge for human beings
[28]. Moreover, the CPST computing could provide a holistic treatment of human thought and thinking to
achieve supportive wisdom. Generally speaking, the CPST computing is a multi-level computing framework.
Cyber-physical computing is an essential mechanism of abstracting physical resources while achieving
CPST in the computing framework [29]. Some methods, e.g., cloud computing-assisted big data analysis,
can achieve this purpose. They are seen as an extension and deep application for physical infrastructure in
GISS. In general, a cloud computing environment is composed of physical servers that contain resources
shared by many users and applications. Within this environment, the big-data-analysis-centric computing
strategy can copy with the application requirements of geological big data.
Within this framework, in addition to cyber-physical computing, the social computing in the cybermatics
is a key part. More recently, it has become a focus in computing application fields. By integrating social
dynamics and mining social knowledge, the social intelligence is thus highlighted in this computing mode
[30]. Then, the valuable geological information and knowledge can be obtained by interacting with social
networks and recommendations from geological experts while carrying out social computing.
It should be pointed out that thinking computing arisen from emotions also plays an important role in
CPST computing. It is emerged to recognize, interpret, and learn human cognitive states. Consequently,
10
Geological data text
Text preprocess
Chinese word segmentation
Deleting stop words
Deleting low-frequency words
A collection of text
(vector representation)
Map1( )
Map2( )
Mapn( )
Classifier1
Classifier2
Classifiern
Boosting-based
weighted ensemble
result
Result evaluation
··· ···
Mapper Reducer
Sequence
File
Geological
expert
Hadoop-based geological data text computing framework
Fig. 6. Geological data text computing framework under Hadoop architecture.
it enhances self-adaptive and self-understanding performance while addressing geological data for GISS.
In thinking computing, there exists many analysis techniques, including deep learning, fuzzy logic, kernel
learning, brain-inspired intelligence methods [24], [25], [31].
2) An Application Case: Here, an application case is analyzed while implementing cyber-physical com-
puting in CPST multi-level computing framework.
Aiming at the large amount of structured and unstructured geological big data, the computational perfor-
mance of traditional computing approaches is not satisfactory. Considering that there are huge volume of
geological data texts and the size of some texts is relatively not big, the Hadoop-based Sequence File method
is employed while merging a large number of small files into one big file stored with the form of key. After
completing the store processing for geological texts, some big data analysis applications can be conducted.
For example, the text classification prediction is expected to be implemented. A specific implementation
process is shown in Fig. 6.
Hadoop is known as a distributed system infrastructure [32]. It uses cluster computing and high-speed
storage technologies to address huge amount of data. The key of Hadoop is MapReduce computing framework
and Hadoop distributed file system (HDFS). Here, the data are stored by using HDFS, and the parallel
computing is implemented by using MapReduce. In Fig. 6, the working process of text classification is
summarized as follows. Firstly, the preprocessing for those collected geological texts is conducted, including
Chinese word segmentation, deleting stop words, and deleting low-frequency words. Secondly, the vector
representation for a collection of text is achieved using vector space model (VSM) model [33]. Thirdly,
within the MapReduce mechanism, the text vectors are classified using K-nearest neighbor (KNN) algorithm
method [34]. Then, all of the classification results are integrated with a Boosting-based weighted ensemble.
Finally, the classification is evaluated by some geological experts in social computing mode.
C. System Security
Security is one of the most concerned problems for system. Due to open network structure and service
sharing scheme in cloud, it imposes very challenging obstacles to the security guarantee. In the system
11
proposed here, through monitoring abnormal behavior of software used in network environment of clients,
the cloud can intercept malicious program and Trojan. Security analysis shows that the third party provider
chosen in the proposed system could guarantee data availability to a certain degree, and thus effectively
enhance data confidentiality. Furthermore, data security is one of the most concerned issues for the user.
Some effective protection measures for data have been developed to ensure the security of the data, such
as multiple copies of data storage, encryption [35]. A cloud security policy focuses on managing users,
protecting data, securing virtual machines, and specifying the infrastructure usage policies. The developed
system in this article has data backup and data restore functions to ensure data security. Meanwhile, the
classification strategy for modules in the system can also make the system manager effectively check access
permissions on system resource, and accordingly display content.
V. CONCLUSION
Currently, the complexity of geological data imposes very challenging obstacles to the traditional manage-
ment service of geological information. With the rapid development of computer technologies and Internet,
many software tools and solutions are developed to design a GISS to improve the application performance.
Through the deep understanding of the applications of IoT in the field of geology, the construction require-
ment of GISS is accordingly analyzed, and the overall architecture and physical deployment of the geological
things are proposed. In this article, RFID, sensors, and other things are employed to achieve the orderly
and intelligent management of geological data. Specifically, based on the management and application of
geological equipments, this article presents a modeling and computing scheme using CPST. Moreover, two
application cases are provided to show the efficiency of using CPST modeling and computing scheme for
GISS. In addition, the data security of the developed system is also analyzed. As an effective solution for
GISS management using CPST, there are several interesting research topics to be explored in the future. For
instance, further GISS can be developed using some advanced open source software.
CONFLICT OF INTERESTS
The authors declare that there is no conflict of interests regarding the publication of this article.
REFERENCES
[1] M. Broy, A. Schmidt, “Challenges in engineering cyber-physical systems,” Computer, vol. 47, no. 2, pp. 70–72, 2014.[2] X. Cheng, Y. Sun, A. Jara, H. Song, Y. Tian, “Big data and knowledge extraction for cyber-physical systems,” International
Journal of Distributed Sensor Networks, vol. 2015, Article ID 231527, 4 pages, 2015.[3] X. Luo, D. Zhang, L. T. Yang, J. Liu, X. Chang, H. Ning, “A kernel machine-based secure data sensing and fusion scheme
in wireless sensor networks for the cyber-physical systems,” Future Generation Computer Systems, vol. 61, pp. 85–96, 2016.[4] G. Chen, W. Li, Q. Kong, S. Liu, C. Lv, F. Tian, “Recent progress of marine geographic information system in China: A
review for 2006-2010,” Journal of Ocean University of China, vol. 11, no. 1, pp. 18–24, 2012.[5] A. Ahmad, A. Paul, M. M. Rathore, H. Chang, “Smart cyber society: Integration of capillary devices with high usability based
on cyber-physical system,” Future Generation Computer Systems, vol. 56, pp. 493–503, 2016.[6] Y. C. Sun, H. B. Song, A. J. Jara, R. F. Bie, “Internet of Things and big data analytics for smart and connected communities,”
IEEE Access, vol. 4, pp. 766–773, 2016.[7] S. Y. Cheng, Z. P. Cai, J. Z. Li, X. L. Fang, “Drawing dominant dataset from big sensory data in wireless sensor networks,”
in Proceedings of the IEEE Conference on Computer Communications, 2015, pp. 531–539.[8] X. Luo, Y. Lv, M. Zhou, W. Wang, W. Zhao, “A Laguerre neural network-based ADP learning scheme with its application to
tracking control in the Internet of Things,” Personal and Ubiquitous Computing, vol. 20, no. 3, pp. 361–372, 2016.[9] D. Frederico, J. F. Carvalho, A. Anderson, V. C. Garcia, “A systematic review on cloud computing,” Journal of Supercomputing,
vol. 68, no. 3, pp. 1321–1346, 2014.
12
[10] H. S. Ning, H. Liu, J. H. Ma, L. T. Yang, R. H. Huang, “Cyber-physical-social-thinking hyperspace based science andtechnology,” Future Generation Computer Systems, vol. 56, pp. 504–522, 2016.
[11] P. Derler, E. A. Lee, A. S. Vincentelli, “Modeling cyber-physical systems,” Proceedings of the IEEE, vol. 100, no. 1, pp.13–28, 2012.
[12] X. Hu, T. H. S. Chu, H. C. B. Chan, et al. “Vita: a crowd sensing-oriented mobile cyber-physical system,” IEEE Transactionson Emerging Topics in Computing, vol. 1. no. 1, pp. 148–165, 2013.
[13] F. Y. Wang. “The emergence of intelligent enterprises: from CPS to CPSS,” IEEE Intelligent Systems, vol. 25, no. 4, pp. 85–88,2010.
[14] D. Hussein, S. Park, S. N. Han, N. Crespi, “Dynamic social structure of things: A contextual approach in CPSS,” IEEE InternetComputing, vol. 19, no. 3, pp. 12–20, 2015.
[15] I. A. T. Hashem, I. Yaqoob, N. B. Anuar, S. Mokhtar, A. Gani, S. Ullah Khan, “The rise of “big data” on cloud computing:Review and open research issues,” Information Systems, vol. 47, pp. 98–115, 2015.
[16] A. Jula, E. Sundararajan, Z. Othman, “Cloud computing service composition: a systematic literature review,” Expert Systemswith Applications, vol. 41, no. 8, pp. 3809–3824, 2014.
[17] G. Kortuem, F. Kawsar, V. Sundramoorthy, et al. “Smart objects as building blocks for the Internet of Things,” IEEE InternetComputing, vol. 14, no. 1, pp. 44–51, 2009.
[18] Y. Zhan, L. Liu, L. Wang, Y. Shen, “Wireless sensor networks for the Internet of Things,” International Journal of DistributedSensor Networks, vol. 2013, Article ID 717125, 2 pages, 2013.
[19] S. Y. Cheng, Z. P. Cai, J. Z. Li, “Curve query processing in wireless sensor networks,” IEEE Transactions on VehicularTechnology, vol. 64, no. 11, pp. 5198–5209, 2015.
[20] A. Botta, W. De Donato, V. Persico, A. Pescap, “Integration of cloud computing and Internet of Things: A survey,” FutureGeneration Computer Systems, vol. 56, pp. 684–700, 2016.
[21] G. S. Yan, Q. W. Xue, K. Y. Xiao, J. P. Chen, J. L. Miao, H. L. Yu, “An analysis of major problems in geological survey bigdata,” Geological Bulletin of China, vol. 34, no. 7, pp. 1273–1279, 2015.
[22] A. R. H. Swan, M Sandilands. “Introduction to geological data analysis,” International Journal of Rock Mechanics & MiningScience & Geomechanics Abstracts, vol. 32, no. 8, pp. 387, 1995.
[23] C. L. Li, J. Q. Li, H. C. Zhang, A. H. Gong, D. Q. Wei, “Big data application architecture and key technologies of intelligentgeological survey,” Geological Bulletin of China, vol. 34, no. 7, pp. 1288–1299, 2015.
[24] D. Silver, A. Huang, C. J. Maddison, et al, “Mastering the game of Go with deep neural networks and tree search,” Nature,vol. 529, no. 7587, pp. 484–489, 2016.
[25] V. Mnih, K. Kavukcuoglu, D. Silver, et al, “Human-level control through deep reinforcement learning,” Nature, vol. 518, no.7540, pp. 529–533, 2015.
[26] R. M. Shiffrin, K. Borner, “Mapping knowledge domains,” Proceedings of The National Academy of Sciences of The UnitedStates of America, vol. 101, pp. 5183–5185, 2004.
[27] Y. C. Sun, C. Lu, R. F. Bie, J. S. Zhang, “Semantic relation computing theory and its application,” Journal of Network andComputer Applications, vol. 59, pp. 219–229, 2016.
[28] A. Sheth, P. Anantharam, C. Henson, “Physical-cyber-social computing: an early 21st century approach,” IEEE IntelligentSystems, vol. 28, no. 1, pp. 78–82, 2013.
[29] A. Ara. “A secure service provisioning framework for cyber physical cloud computing systems,” International Journal ofParallel Emergent & Distributed Systems, vol. 6, no. 1, pp. 10–19, 2015.
[30] M. Tavakolifard, K. C. Almeroth, “Social computing: An intersection of recommender systems, trust/reputation systems, andsocial networks,” IEEE Network, vol. 26, no. 4, pp. 53–58, 2012.
[31] X. Luo, J. Liu, D. Zhang, X. Chang, “A large-scale web QoS prediction scheme for the industrial Internet of Things basedon a kernel machine learning algorithm,” Computer Networks, vol. 101, pp. 81–89, 2016.
[32] L. Wang, J. Tao, R. Ranjan, H. Marten, A. Streit, J. Chen, D. Chen, “G-Hadoop: MapReduce across distributed data centersfor data-intensive computing,” Future Generation Computer Systems, vol. 29, no. 3, pp. 739–750, 2013.
[33] D. Kim, “Group-theoretical vector space model,” International Journal of Computer Mathematics, vol. 92, no. 8, pp. 1536–1550, 2015.
[34] J. J. Valero-Mas, J. Calvo-Zaragoza, J. R. Rico-Juan, “On the suitability of Prototype Selection methods for kNN classificationwith distributed data,” Neurocomputing, vol. 203, pp. 150–160, 2016.
[35] R. M. Jogdand, “Enabling public verifiability and availability for secure data storage in cloud computing,” Evolving Systems,vol. 6, no. 1, pp. 55–65, 2015.
13