內容大鋼
本書利用模糊技術建立了智能服裝推薦系統，使消費者在電商平台選擇服裝時，通過輸入體型數據、風格關鍵詞和偏好圖像等信息，生成虛擬人體模型。本書通過服裝專家知識建立了知識庫，以便為消費者推薦服裝，消費者不僅可以實時看到虛擬試穿效果，而且如果對推薦結果不滿意，可將不滿意部位的信息反饋給系統，系統利用反饋演算法智能學習消費者反饋信息，自適應地更新知識庫。推薦模塊能根據更新后的知識庫進行二次推薦，直到消費者滿意為止，成功推薦的結果會存入成功知識庫作為今後推薦的依據。
本書可作為服裝企業、電商平台研發推薦系統時的指導用書，也可作為服裝行業從業者的參考用書。
作者介紹
張俊傑//袁樺//張贇|責編:苗苗
張俊傑，武漢紡織大學電腦與人工智慧學院特別教授，博士，碩士研究生導師。2005年12月獲華中科技大學軟體工程碩士學位，2017年5月獲法國里爾第一大學自動化博士學位。
主要研究方向為服裝數字化、人工智慧與人文社科交叉學科。已發表SCI論文10余篇，主持多項教育部協同創新項目、湖北省教育廳科研項目、中國紡織工業聯合會科研項目，獲批國家發明專利4項，獲批軟體著作權10余項；獲全國多媒體課件大賽一等獎1項，湖北省教學成果獎一等獎、三等獎各1項，中國紡織工業聯合會高等教育教學成果獎多項，獲武漢市社會科學優秀成果獎二等獎、陝西高等學校科學技術研究優秀成果獎二等獎和歐洲創意與創新展覽會科技獎金獎各1項；指導學生獲湖北省「互聯網＋」大賽金獎1項。
目錄
General Introduction
Chapter 1 E-commerce for Fashion Products
1.1 E-commerce and Recommendation Systems
1.2 Fashion Products and Fashion Markets
1.2.1 The marketing of fashion
1.2.2 Targeting a market
1.2.3 Customer profiles
1.2.4 Seasonal and occasion markets
1.2.5 Fashion products
1.3 Fashion Consumer Behavior
1.3.1 Cultural factor
1.3.2 Social factor
1.3.3 Individual factor
1.3.4 Psychological factor
1.4 Main Ideas
1.4.1 Successful Cases Database Module
1.4.2 Market Forecasting Module
1.4.3 Knowledge-based Recommendation Module
1.4.4 Knowledge Updating Module
1.5 Originality
Chapter 2 Fashion Data Acquisition from Social Networks and Design of Sensory Experiments
2.1 Social Networks
2.2 Sensory Evaluation
2.2.1 Definition and industrial application
2.2.2 Basic notions about sensory evaluation
2.3 Design of Experiments
2.3.1 Preliminary knowledge
2.3.2 Experiments
2.4 Conclusions
Chapter 3 Concerned Computational Tools
3.1 Knowledge and Its Features
3.2 Fuzzy Sets and Fuzzy Operations
3.2.1 Fuzzy sets theory
3.2.2 Standard operations of fuzzy sets
3.2.3 Properties of fuzzy sets
3.2.4 Operations on fuzzy relations
3.2.5 Triangular fuzzy number
3.2.6 Similarity degree of fuzzy sets
3.3 FCEM and FuzzyAHP
3.3.1 FCEM
3.3.2 Fuzzy AHP
3.4 Complex Networks
3.4.1 Scale-free networks
3.4.2 Small-world networks
3.5 Markov Chain
3.6 Case-based Reasoning
3.7 Organization of the Computational Techniques
Chapter 4 Successful Cases Database Module and Market Forecasting Module
4.1 Mathematical Formalization
4.2 Consumer Profile
4.2.1 Size rules definition
4.2.2 Fuzzy description of height
4.2.3 Fuzzy description of "fat/thin
4.2.4 Application of the FAHP algorithm
4.2.5 Fuzzy comprehensive evaluation method (FCEM)
4.2.6 A real case of consumer profile
4.3 Successful Cases Database Module
4.3.1 General principle
4.3.2 Formalization, similarity and reasoning
4.3.3 A real case of successful cases database module
4.4 Market Forecasting Module
4.4.1 General principle
4.4.2 Market data analysis with Markov Chain
4.4.3 Complex network
4.4.4 A real case of market forecasting module
4.5 Conclusions
Chapter 5 Knowledge-based Recommendation Module and Knowledge Updating Module
5.1 Knowledge-based Recommendation Module
5.1.1 General principle
5.1.2 Data aggregation from various trainees
5.1.3 Generation of the knowledge base (KB)
5.1.4 Recommendation process through a real case
5.2 Knowledge Updating Module
5.2.1 General principle
5.2.2 Virtual try-on and evaluation
5.2.3 Knowledge base updating 1
5.2.4 Knowledge base updating 2
5.3 Validation of the Recommendation Modules
5.4 Conclusions
Chapter 6 General Conclusion and Perspectives
6.1 General Conclusion
6.2 Perspectives
References
Appendix Questionnaires for Sensory Evaluations Experiments and Historical Shopping Data
[