Type-2 Fuzzy Ontology and Fuzzy Markup Language for Real-World Applications

Type-2 Fuzzy Ontology and Fuzzy Markup Language for Real-World Applications

FUZZ-IEEE 2016 Tutorial: FUZZ-IEEE-03 Type-2 Fuzzy Ontology and Fuzzy Markup Language for Real-World Applications Organized by Chang-Shing Lee, National University of Tainan, Taiwan Giovanni Acampora, Nottingham Trent University, UK Yuandong Tian, Facebook AI Research, USA 24 July, 2016 0 / 71 FUZZ-IEEE 2016 Tutorial: FUZZ-IEEE-03 Part 1: Type-2 Fuzzy Ontology and Applications Chang-Shing Lee, NUTN, Taiwan Part 2: Fuzzy Markup Language Giovanni Acampora, NTU, UK Part 3: Real-World Application on Game of Go Yuandong Tian, Facebook AI Research, USA 1 / 71 FUZZ-IEEE 2016 Tutorial: FUZZ-IEEE-03 Part 1 Type-2 Fuzzy Ontology and Applications Chang-Shing Lee National University of Tainan, Taiwan 2 / 71 Research Team 3 / 71 Co-Sponsors 4 / 71 Type-2 Fuzzy Ontology Applications • FML IEEE 1855-2016 Standard • Type-2 Fuzzy Set • Fuzzy Ontology • Game of Go Application • Personalized Diet Recommendation • Adaptive Learning Application 5 / 71 FML IEEE 1855-2016 6 / 71 Introduction to T2FS (1/5) 7 / 71 Introduction to T2FS (2/5) i m ~ (x ) Vertical Slice u A 2 W 1 x'1 ~ Wx'N UMF(A) ~ UMF(A) u i i MF1(x ) MFN (x ) i u u MFN (x ) 1 n ~ m A (x, u) Embedded T2 FS i x MF1(x ) u ~ Embedded LMF(A) x T1 FS 0 i ~ l x r Some eye Contact (A) Uncertainty About Uncertainty About Left End-Point Right End-Point 8 / 71 Introduction to T2FS (3/5) u 1 ~ ~ UMF (A) UMF (A) Embedded FS ~ LMF (A) ~ ~ FOU (A) FOU (A) X 9 / 71 Introduction to T2FS (4/5) Type-2 FLS Output Processing Crisp Rules Defuzzifier outputs y ≠Y Crisp Type-reducer inputs Fuzzifier Type-reduced x ≠X Set(Type-1) Fuzzy Fuzzy input sets Inference output sets ~ ~ F Ax (or Ax ) x 第二型模糊邏輯系統 10 / 71 Introduction to T2FS (5/5) u u ~ Low Low 1.0 1.0 UMF 0.8 0.8 0.6 0.6 0.4 0.4 LMF 0.2 0.2 0 0 0 ( ) 5 10 15 20 25 30 40 x C 0 5 10 15 20 25 30 40 x( C) mA (xÅ,u) 1.0 ~ Å 0.8 uA (x ,u) 0.6 1.0 0.4 0.8 0.2 5 10 15 20 25 30 40 0 0.6 0.2 x 0.4 0.4 0.6 0.2 0.8 1.0 0 0.2 0.4 0.6 0.8 1.0 u u 11 / 71 12 / 71 13 / 71 14 / 71 15 / 71 16 / 71 Dynamic Assessment and IRT-based Learning Application 17 / 71 18 / 71 19 / 71 20 / 71 21 / 71 Video Demonstration • Taiwan Open 2009 • Human vs. Computer Go @ IEEE WCCI 2012 • Human vs. Computer Go @ FUZZ-IEEE 2011 • Human vs. Computer Go in Taiwan in 2011 22 / 71 Adaptive linguistic assessment Game Results Repository Domain Expert Domain Expert Game Results T2FS Construction Repository Mechanism KB/RB Adaptive Go-Ranking Repository Assessment Ontology PSO Model Human vs. MoGoTW Estimation Mechanism T2FS-based Genetic Learning Adaptive UCT-based Mechanism Go-Ranking Mechanism Bradley-Terry Model Estimation Mechanism T2FS-based Fuzzy Inference Mechanism MoGoTW Players Human-Performance Mapping Mechanism Semantic Analysis Mechanism Personal Profile Repository Players Rank Repository 23 / 71 Go-ranking assessment ontology Adaptive Go-Ranking Assessment Ontology Domain Layer 7x7 13x13 9x9 . 19x19 Category Layer God . NCKU Class Layer IEEE WCCI Temple FUZZ-IEEE . NUTN 2012 Where 2013 . Amateur Professional Who Certificated Rank Player Player 6D . Gender Age MoGoTW Male 45 2013/7/9 . 2013/7/8 . When . How White Time Setting Machine Spec 45mins/Side Rule HP ProLiant DL785 Komi Black Round 7.5 Chinese 2 Round1 Game 13 Game14 . SN ,GR . Game11 13 13 SN14,GR14 . Game1K-1 SN11,GR11 Game12 SN1K-1,GR1K-1 Game1K SN12,GR12 RoundN SN1K,GR1K ~ ~ ~ . SN WinningRate ~ Komi 121145 60 . GameWeight 7.5 19 . RankMethod RankActual 6.38D 6D What SN: Simulation Number ~ ~ 24 / 71 GR: Game Result Low Medium H~igh Fuzzy inference structure Note: (1) M denotes number of fired rules Output Lay e r (2) xÅ = {GW Å,WRÅ, SN Å, KomiÅ} Rank (xÅ) (3) GW is GameWeight AVG (4) WR is WinningRate Type-Reduction Layer Rankl (xÅ) Rankr (xÅ) KM KM Consequent Layer 1 1 M M [Rank , Rank ] [Rank l , Rank r ] . l r . KM KM Rule Layer 1 1 M M . [ f (xÅ), f (xÅ)] . [ f (xÅ), f (xÅ)] MIN MIN Antecedent Layer ~ [m ~ (Kom iÅ), m Komi (Kom iÅ)] [m ~ ( Å), m ~ ( Å)] Komi Medium Medium . GW GW Å ~ Å GWLow [m ~ (Komi ), m Komi (Komi )] GWLow KomiLow Low . ~ ~ [m (GW Å),m GW (GW Å)] GWMedium Medium ~ [m ~ (GW Å), m GW (GW Å)] GWHigh High ~ [m ~ (KomiÅ),m Komi (KomiÅ)] KomiHigh High Input Layer S N Å GW Å WRÅ KomiÅ 25 / 71 Personalized Diet Recommendation 26 / 71 Diet assessment / recommendation ontology Adaptive Diet Assessment Ontology Domain Layer Japan UK Taiwan . USA Category Layer NUTN Class Layer FuCheng . ChiKu Campus RongYu Campus Campus . OASE CASDL VCI Lab. Where Lab. Lab. Under- Assistant Graduate Advisor Who Graduate . 11/1/2009 . 11/14/2009 . 11/30/2009 When ~Diet Goal How 2000(2000)kcal ~Fats & Nuts Vegetables Whole Grains & 6(6) servings ~ Starches Meats & 3.5(3.5)servings ~Low-Fat Milk ~ 1.5(1.5) servings 12(12) servings ~ Fruits ~ Proteins 3.5(3.5) servings 6(6)servings Actual Caloric Intake 2500~ (2500)kcal What Breakfast Dinner Lunch ~Corn Soup . 1(1)portion ~ Dumpling ~Corn Soup Caramel 1.5(1.5) portions 1(1)portion Seafood Spaghetti Pudding with Tomato Sauce ~ Black Tea ~ 1(1) portion Pork Bun Soy Milk ~ 1(1) portion ~ ~ 1(1) portion . 1(1) portion 1(1) portion Whole Grains Fats & Meats & Proteins Sugar Vegetables & Starches Nuts ~ ~ ~ Fruits Low-Fat Milk ~ ~ 10(9.65) 72(72)g 0.5(0.5) serving ~ ~ 14.5(14.5) 9(9.3) servings 0(0) serving 1(0.6) serving servings servings ~Carbohydrate ~ Fat 1197(1196.8)kcal ~ Protein 874(874.35)kcal FGB 407(407.4)kcal ~ PCF 1(0.66) PCC ~ PCR ~ PCP 35(35.28)% ~ 48(48.29)% ~ 124(123.93)% 16(16.44)% PCR: Percentage of Caloric Ratio FGB: Food Group Balance . DHL DHL: Dietary Healthy Level DHL~ Method ~ DO DO: Desired Output 3.4(3.42) 4(4) PCC: Percentage of Calories from Carbohydrate PCP: Percentage of Calories from Protein Rclass_Semantic PCF: Percentage of Calories from Fat ~ ~ ~ Recommended 27 / 71 VeryLow Low Medium High~ VeryHigh~ . Semantics Layer Personalized diet recommendation Step 6.2 Taiwan T2 FS-based Learning Mechanism Step 5 Step 6 … Step 6.1 … Training Data T2 FS-based Genetic Learning Mechanism Personalized KB Step 1 … Step 4 Step 4.1 Domain Experts Nutritional Balanced Step 6.3 Mechanism T2 FS-based Fuzzy Inference Food Item Database Mechanism Step 2 Step 4.2 Ontology Experts … Caloric Balanced Step 6.4 Mechanism Linguistic Knowledge Meal Record Database Discovery Mechanism Step 3 Adaptive Personal … Diet Assessment and Step 7.1 Step 8 Step 4.3 Recommendation Ontology Type-2 Six-Food-Group Subjects Balanced Mechanism … Dietary Health Balanced Computation Mechanism Level Repository T2GFML Repository Step 7.2 28 / 71 T2FS fuzzy variables u u 1 1 PCC (%) PCP (%) 0 20 40 60 80 100 0 20 40 60 80 100 (a) (b) u u 1 1 PCF (%) FGB -1 0 1 2 3 4 5 6 0 20 40 60 80 100 (c) (d) u u 1 1 PCR (%) DHL 0 40 80 120 160 200 0 2 4 6 8 10 12 (e) (f) 29 / 71 ~ ~ ~ ~ ~ ~ ~ ~ ~~ ~ ~ Low~ Low ~ Medium~ ~ ~ ~ uu Low VeryLow~ LowMediumMedium MediumMediumMedium HighHigh HighHighVeryHighHigh 1 Low High 11 PCC(%) PCRPCP(%) FGB PCF(%) DHL 10102010 20204020 3030 6030 4040 8040 5050 100506060 120607070 140708080 160908090 10010018090 100200 -1 10 2 1 3 2 4 3 5 4 6 5 7 6 8 9 10 11 12 u uu LowLow VeryLowMediumMediumLow Low MediumMedium HighHigh HighHighVeryHigh 1 Medium High u11 Low Low Medium High 1 FGB PCFPCC(%)(%) PCPDHL(%) 10 20 30 40 50 60 70 80 90 100 -1 101010 202 1 303 2 404 3 505 4 606 5 707 6 80808 90990 10010100 11 PCR12(%) 20 40 60 80 100 120 140 160 180 200 T2FS fuzzy variables ~ ~ ~ Low Medium ~ ~ ~ u High u Low Medium High 1 1 PCC(%) PCP(%) 10 20 30 40 50 60 70 80 90 100 10 20 30 40 50 60 70 80 90 100 (a) (b) u Low~ Medium High u Low Medium High ~ ~ ~ ~ ~ u1 Low Medium High 1 u Low Medium High 1 1 PCC(%) PCP(%) 10 20 30 40 50 60 70 80 90 100PCF(%) 10 20 30 40 50 60 70 80 FGB90 100 10 20 30 40 50 60 70 80 90 100 -1 0 1 2 3 4 5 6 u Low Medium (c) High u (d) 1 ~ ~ ~ ~ Low~ ~ Medium ~ High~ u Low Medium High u VeryLow Low Medium High VeryHigh 1 1 PCF(%) FGB 10 20 30 40 50 60 70 80 90 100 ( ) -1 0 1 2 3 4 5 6 PCR % DHL 20 40 60 80 100 120 140 160 180 200 1 2 3 4 5 6 7 8 9 10 11 12 (e) u VeryLow Low Medium (f) High VeryHigh u 1 Low Medium High 1 DHL PCR(%) 1 2 3 4 5 6 7 8 9 10 11 12 20 40 60 80 100 120 140 160 180 200 30 / 71 Adaptive Learning Application 31 / 71 Video Demonstration • Knowledge Web for World-Wide Students Learning (KWSLearn) – Website: https://sites.google.com/site/kwslearn/ – Cooperated Organization: • Boyo Social Welfare Foundation, Taiwan • Tainan City Government, Taiwan • National University of Tainan (NUTN), Taiwan • Center for Research of Knowledge Application & Web Service (KWS), NUTN, Taiwan • Ontology Application & Software Engineering (OASE) Lab.

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    120 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us