Artificial Intelligence Project 1 Neural Networks

16
Artificial Intelligence Artificial Intelligence Project 1 Project 1 Neural Networks Neural Networks Biointelligence Lab School of Computer Sci. & Eng. Seoul National University

description

Artificial Intelligence Project 1 Neural Networks. Biointelligence Lab School of Computer Sci. & Eng. Seoul National University. Outline. Classification Problems Task 1 Estimate several statistics on Diabetes data set Task 2 - PowerPoint PPT Presentation

Transcript of Artificial Intelligence Project 1 Neural Networks

Page 1: Artificial Intelligence Project 1 Neural Networks

Artificial IntelligenceArtificial IntelligenceProject 1Project 1

Neural NetworksNeural Networks

Biointelligence Lab

School of Computer Sci. & Eng.

Seoul National University

Page 2: Artificial Intelligence Project 1 Neural Networks

(C) 2000-2002 SNU CSE BioIntelligence Lab

2

OutlineOutline

Classification Problems Task 1

Estimate several statistics on Diabetes data set

Task 2 Given unknown data set, find the performance as good as you

can get The labels of test data are hidden.

Page 3: Artificial Intelligence Project 1 Neural Networks

(C) 2000-2002 SNU CSE BioIntelligence Lab

3

Network Structure (1)Network Structure (1)

positive

negative

fpos(x) > fneg(x),→ x is postive

Page 4: Artificial Intelligence Project 1 Neural Networks

(C) 2000-2002 SNU CSE BioIntelligence Lab

4

Network Structure (2)Network Structure (2)

f (x) > thres,→ x is postive

Page 5: Artificial Intelligence Project 1 Neural Networks

Medical Diagnosis: DiabetesMedical Diagnosis: Diabetes

Page 6: Artificial Intelligence Project 1 Neural Networks

(C) 2000-2002 SNU CSE BioIntelligence Lab

6

Pima Indian DiabetesPima Indian Diabetes

Data (768) 8 Attributes

Number of times pregnant Plasma glucose concentration in an oral glucose tolerance test Diastolic blood pressure (mm/Hg) Triceps skin fold thickness (mm) 2-hour serum insulin (mu U/ml) Body mass index (kg/m2) Diabetes pedigree function Age (year)

Positive: 500, negative: 268

Page 7: Artificial Intelligence Project 1 Neural Networks

(C) 2000-2002 SNU CSE BioIntelligence Lab

7

Report (1/4)Report (1/4)

Number of Epochs

Page 8: Artificial Intelligence Project 1 Neural Networks

(C) 2000-2002 SNU CSE BioIntelligence Lab

8

Report (2/4)Report (2/4)

Number of Hidden Units At least, 10 runs for each setting

# Hidden

Units

Train Test

Average SD

Best Worst Average SD

Best Worst

Setting 1

Setting 2

Setting 3

Page 9: Artificial Intelligence Project 1 Neural Networks

(C) 2000-2002 SNU CSE BioIntelligence Lab

9

Report (3/4)Report (3/4)

Page 10: Artificial Intelligence Project 1 Neural Networks

(C) 2000-2002 SNU CSE BioIntelligence Lab

10

Report (4/4)Report (4/4)

Normalization method you applied. Other parameters setting

Learning rates Threshold value with which you predict an example as

positive. E.g. if f(x) > thres, you can say it is postive, otherwise negativ

e.

Page 11: Artificial Intelligence Project 1 Neural Networks

(C) 2000-2002 SNU CSE BioIntelligence Lab

11

Challenge (1)Challenge (1)

Unknown Data Data for you: 5822 examples Pos: 348, Neg: 5474

Test data 4000 examples Pos: 238, Neg: 3762 Labels are HIDDEN!

Page 12: Artificial Intelligence Project 1 Neural Networks

(C) 2000-2002 SNU CSE BioIntelligence Lab

12

Challenge (2)Challenge (2)

Data train.data : 5822 x 86 (5822 examples with 86 dim; labe

ls are attached at 86th-column: positive 1, negative 0) test.data: 4000 x 85 (5822 examples with 85 dim) Test labels are not given to you.

Verify your NN at http://knight.snu.ac.kr/aiproj1/ai_nn.asp

Page 13: Artificial Intelligence Project 1 Neural Networks

(C) 2000-2002 SNU CSE BioIntelligence Lab

13

Challenge (3)Challenge (3)

Include followings at your report The best performance you achieved. The spec of your NN when achieving the performance.

Structure of NN Learning epochs Your techniques

Other remarks…

True

PredictPositive Negative

Positive

NegativeConfusion matrix

Page 14: Artificial Intelligence Project 1 Neural Networks

(C) 2000-2002 SNU CSE BioIntelligence Lab

14

ReferencesReferences

Source Codes Free softwares NN libraries (C, C++, JAVA, …) MATLAB Toolbox Weka

Web sites http://www.cs.waikato.ac.nz/~ml/weka/

Page 15: Artificial Intelligence Project 1 Neural Networks

(C) 2000-2002 SNU CSE BioIntelligence Lab

15

Pay Attention!Pay Attention!

Due (April 14, 2004): until pm 11:59 Submission

Results obtained from your experiments Compress the data Via e-mail ([email protected])

Report: printed version. (419 호 오장민 ) Used software and running environments Results for many experiments with various parameter settings Analysis and explanation about the results in your own way

메일 제목에 “ [4a05project1]” 반드시 포함

Page 16: Artificial Intelligence Project 1 Neural Networks

(C) 2000-2002 SNU CSE BioIntelligence Lab

16

Optional ExperimentsOptional Experiments

Various learning rate Number of hidden layers Applying feature selection techniques Output encoding