Download Cross System Personalization:Enabling personalization across multiple systems PDF

TitleCross System Personalization:Enabling personalization across multiple systems
File Size1.6 MB
Total Pages130
Table of Contents
	Problem Statement
State of the Art and Related Work
	User Modeling
		Representation Formats and Standards for User Profiles
		Personalization Engines and User Modeling servers
	Machine Learning and Statistical Techniques
		Dimensionality Reduction
		Linear Methods for Dimensionality Reduction
		Non-Linear Methods for Dimensionality reduction
	Collaborative Filtering
		Types of Collaborative Filtering Algorithms
		Relevant Collaborative Filtering Algorithms
		Evaluation in Collaborative Filtering
		Privacy in Collaborative Filtering
		Trust in Collaborative Filtering
	Final comments on the Literature Survey
Conceptual Model and Methods
	A Semantic Approach to Cross System Personalization
		The Unified User Context Model
		The Context Passport Metaphor
		The Cross System Communication Protocol
		Discussion and Conclusion
	A Learning Approach to Cross System Personalization
		Challenges in Automatic Cross System Personalization
	Learning Methods for enabling Cross System Personalization
		Manifold Alignment
		Cross System Personalization as a matrix completion problem
		Sparse Factor Analysis
		Distributed Probabilistic Latent Semantic Analysis
		Discussion and Conclusion
	Spam detection in Collaborative Filtering
		What Is Spam In Collaborative Filtering ?
		Characteristics Of Shilling Profiles
		Optimal Shilling Strategy
		Using PCA for Spam Detection
		Soft clustering using PLSA
	Robustness in Collaborative Filtering
		SVD and Its Variations
		Robust Matrix Factorization
		Discussion and Conclusion
	Evaluation Plan
	Evaluation of Learning methods for CSP
		Experimental Setup
	Evaluation Results for CSP
		Manifold Alignment
		Sparse Factor Analysis
		Distributed PLSA
	Evaluation of Shilling detection
		Experimental Setup
		PLSA based spam detection
		PCA based spam detection
	Evaluation of Robustness in Collaborative Filtering
		Experimental Setup
		Metrics Used
		Experimental results
Conclusions and Future Work
	Future Work
		List of Figures
		List of Tables
		List of Algorithms
		List of Publications

Similer Documents