User Modeling, Adaptation, and Personalization

Name: User Modeling, Adaptation, and Personalization | 20th International Conference, UMAP 2012, Montreal, Canada, July 16-20, 2012 Proceedings
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

20th International Conference, UMAP 2012, Montreal, Canada, July 16-20, 2012 Proceedings

Judith Masthoff Bamshad Mobasher Michel Desmarais Roger Nkambou(Herausgeber*in)

Springer (Verlag)

Erschienen am 19. Juni 2012

XVIII, 396 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-642-31454-4 (ISBN)

53,49 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Title
Preface
Organization
Table of Contents
Long Papers
Personalized Network Updates: Increasing Social Interactions and Contributions in Social Networks
Introduction
Related Work
The Online Total Wellbeing Diet Portal
Personalized News Feeds
Evaluation
Activity Feed Uptake
Feed Ranking
Impact on User Activities
Feeds and Friending
Conclusions and Future Work
References
Realistic Simulation of Museum Visitors' Movements as a Tool for Assessing Sensor-Based User Models
Introduction
Related Research
Prerequisites
Simulation of Coordinate-Based Visitor Pathways
Generating Walking Squares
Generating Hovering Squares
Smoothing the Trajectory of Squares
Simulating Sensor Noise
Inference and Prediction of Viewed Exhibits from Coordinates
Classification-Based Inference of Walking and Hovering
Score-Based Inference of Exhibits
Model-Based Prediction of Exhibits
Evaluation
Experiments and Results
Conclusions
References
GECKOmmender: Personalised Theme and Tour Recommendations for Museums
Introduction
Related Research
Personalised Exhibit Theme and Tour Recommendations
Core Functionality
Recommendation Presentation
Evaluation
Field Study Procedures
Results
Conclusions
References
Property-Based Interest Propagation in Ontology-Based User Model
Introduction
Background: Properties in OWL
Property-Based Similarity and Relatedness of Domain Elements
User Model
User Model Definition
User Feedback
User Model Update and Interest Propagation
Evaluation
Related Work
Conclusions and Future Work
References
EEG Estimates of Engagement and Cognitive Workload Predict Math Problem Solving Outcomes
Introduction
Prior Work
Methodology
EEG Data Acquisition
Self Report
Data Analysis
Data Processing
SVM Classification
Results
SVM Predictions
Relation of EEG and Self-report Data
Discussion
References
Preference Relation Based Matrix Factorization for Recommender Systems
Introduction
Related Work
Using Preference Relations in the Matrix Factorization Framework
Preliminaries and Problem Definition
Details of Our Approach
Learning the Factorization Model
Recommendation Generation
Complexity of the Approach
Evaluation
Dataset Used
Algorithms Compared
Experimental Results
Discussion on the Results
Conclusion
References
A Framework for Modeling Trustworthiness of Users in Mobile Vehicular Ad-Hoc Networks and Its Validation through Simulated Traffic Flow
Introduction
Multi-faceted Trust Model
Majority Consensus
Messaging in Support of Trust Modeling
Simulation Framework
Simulation Evaluation
Discussion
References
A Comparative Study of Users' Microblogging Behavior on Sina Weibo and Twitter
Introduction
Related Work
Research Methodology and Evaluation Platform
Research Questions
Evaluation Platform
Analysis of User Behavior on Sina Weibo and Twitter
Analysis of Access Behavior
Syntactic Content Analysis
Semantic Content Analysis
Sentiment Analysis
Analysis of Temporal Behavior
Discussion
References
Modeling Multiple Distributions of Student Performances to Improve Predictive Accuracy
Introduction
Motivation
The Base Classifier: Performance Factors Analysis
Approach
Rationale: Modeling Multiple Distributions of Student Performances
Distinguish Samples of Multiple Distributions
Learn Multiple Classification Models
Select a Classification Model for an Unknown Instance
Experiment and Results
Data and Performance Metrics
Results
Contributions
Future Work and Conclusions
References
A Simple But Effective Method to Incorporate Trusted Neighbors in Recommender Systems
Introduction
Related Work
The Merge Method
Merging the Ratings of Trusted Neighbors
Incorporating with Collaborative Filtering
The Insights of the Merge Method
Experimental Validation
Data Acquisition
Experimental Settings
The Performance of the Merge Method
The Effect of the Similarity Threshold ?
Conclusion and Future Work
References
Exploring Gaze Data for Determining User Learning with an Interactive Simulation
Introduction
Related Work
The AISpace CSP Applet
User Study
Data Preparation and Preprocessing
Eye Gaze Features
Classifying Learners Based on Gaze Data
User classification Based on the Whole Interaction Dataset
User classification Based on the Interaction Evolution Dataset
Conclusion and Future Work
References
Studies to Determine User Requirements Regarding In-Home Monitoring Systems
Introduction
Related Research
User Studies
Study Population and Recruitment
Brainstorming Sessions and Statement Assessment
Questionnaires
Methodology
Concept Mapping
Questionnaires
Results
Clustering
Rating
Questionnaires
Discussion
References
Improving Tensor Based Recommenders with Clustering
Introduction
Related Work
Clustered Tag Space and GA
Motivation
Higher-Order Singular Value Decomposition (HOSVD)
Cluster Analysis of Tags
GA - Tuning Tensor Based Recommenders
Experiments
Baseline
Clustered Tag Space
Conclusion and Future Work
References
Models of User Engagement
Introduction
Related Work
Metrics and Interaction Data
Diversity in Engagement
Models of User Engagement
General Models
User-Based Models
Time-Based Models
Relationship between Models
Conclusions and Future Work
References
Improving the Performance of Unit Critiquing
Introduction
Unit Critiquing Based Recommendation
Conventional Critiquing - Tweaking
Experience-Based Critiquing
Algorithm: Nearest Neighbor Compatibility Critiquing
Ensemble-Based Variation
Evaluation
Results
Related Work
Conclusion
References
Enhanced Semantic TV-Show Representation for Personalized Electronic Program Guides
Introduction
Motivating Scenario
TV-Show Representation Using Explicit Semantic Analysis
TV-Show Ranked Retrieval
Experimental Evaluation
Experiment 1: Evaluation of the Ranked Retrieval Task
Experiment 2: Evaluation of the Feature Generation Process
Related Work
Conclusions and Future Work
References
Attention and Selection in Online Choice Tasks
Introduction
Related Work
Experiment Setup
Experimental Results
Metrics
Results
Predictive Models
Attention Transition Models
Parameter Fitting
Model Simulations
Conclusions
References
Investigating Explanations to Justify Choice
Introduction
Description
Procedure
Participants
Results and Analysis
Guidelines and Patterns
Guidelines
Patterns
Related Work
Conclusion
References
The Effect of Suspicious Profiles on People Recommenders
Introduction
Literature Review
Security of Recommender Systems
Experimental Design
Corpus of Suspicious Profiles
Effect of Suspicious Profiles on Recommenders
Minimising the Effect of Suspicious Profiles on Recommenders
Conclusions
References
Users and Noise: The Magic Barrier of Recommender Systems
Introduction
Related Work
The Empirical Risk Minimization Principle
The Traditional Setting of a Recommendation Task
Recommendation as Risk Minimization
The Magic Barrier
A Statistical Model for Users' Inconsistencies
Deriving the Magic Barrier
Estimating the Magic Barrier
Case Study Using a Commercial Movie Recommender
Moviepilot
Data
Experimental Setup
Results
Conclusion
References
Improving Construct Validity Yields Better Models of Systematic Inquiry, Even with Less Information
Introduction
Background and Datasets
Learning Environment and Behaviors of Interest
Labeling Behaviors within the Learning Environment
Data Sets
Feature Selection and Detector Construction
Removed Correlated Features (RCF) Detector Construction
Improved Construct Validity (ICV) Detector Construction
Results: Comparing Predictive Capabilities of Detectors
Comparing Detectors' Overall Performance
Comparing Detectors' Performance Predicting with Less Data
Discussion and Conclusions
References
Inferring Personality of Online Gamers by Fusing Multiple-View Predictions
Introduction
Related Work
Personality in Virtual Worlds
The Big 5 Model
World of Warcraft
Data Collection
Personality Inference
Behavioral Information
Text Analysis
Social Network Analysis
Fusing Predictors
Experimental Results
Conclusion
References
Towards Adaptive Information Visualization: On the Influence of User Characteristics
Introduction
Related Work
User Study to Compare Radar and Bar Graphs
Individual Characteristics Explored in the Study
Participants and Experimental Tasks
Study Design
Procedure
Measures
Data Analysis and Results
Completion Time - Single Scenario
Completion Time - Double Scenario
User Preference and Ease of Use
Discussion - Envisioning Adaptive Interventions
Conclusion and Future Work
References
WTF? Detecting Students Who Are Conducting Inquiry Without Thinking Fastidiously
Introduction
Data Set
WTF Detector
Obtaining Ground Truth Labels of WTF Behavior Using Text Replays
Data Features
Detector Development
Results
Discussion and Conclusions
References
Short Papers
Adapting Performance Feedback to a Learner's Conscientiousness
Introduction
Study Design
Design
Participants
Hypotheses
Study Results
Comparing Slants Where the Grades Are Passing
Comparing Slants Where the Grades Are Failing
Topics
Discussion and Future Work
References
A Multi-faceted User Model for Twitter
Introduction
Intentional and Extensional Tag-Based User Models
Profile Facets
Preliminary Investigation
Conclusions
References
Evaluating Rating Scales Personality
Introduction
The Rating Scales: An Analysis
The Experiments
Conclusion
References
Automating the Modeling of Learners' Erroneous Behaviors in Model-Tracing Tutors
Introduction
Knowledge Representation
Modeling Erroneous Knowledge
Repairs
Conclusion
References
Using Touch as a Predictor of Effort: What the iPad Can Tell Us about User Affective State
Introduction
mathTouch System
Description
Implementation
Testing Environment
Findings
Discussion and Future Work
References
Domain Ranking for Cross Domain Collaborative Filtering
Introduction
Background
Cross Domain Recommendation Using SNS data
Discussion and Conclusions
References
User Modelling Ecosystems: A User-Centred Approach
Introduction
Related Work
The Personis User Modelling Ecosystem
User-Centred Study into UM Design Requirements
Conclusions
References
Adaptive Score Reports
Introduction
The Case for Adaptive Score Reports
An Interactive Score Report for Students
Adaptive Features
Future Work
References
Doctoral Consortium
Improving Matrix Factorization Techniques of Student Test Data with Partial Order Constraints
Introduction
Q-Matrix and Matrix Factorization
Constraints on the Factorization
Factorization Algorithm
Problem Specification
Hypothesis
POKS Constraints on the Matrices V, W, and H
Validation
Conclusion
References
Evaluating an Implementation of an Adaptive Game-Based Learning Architecture
Research Topic: Adaptivity in Game-Based Learning
Proposed Contributions
A Sound Educational Game
Adaptivity
Trial
Status and Remaining Work
Research Questions
References
Towards a Generic Model for User Assistance
Introduction
User Assistance
Our Approach of Assistance to User
Conclusion and Perspectives
References
Resolving Data Sparsity and Cold Start in Recommender Systems
Research Problems
Progress to Date
Incorporating Trust with Collaborative Filtering
Recommender Systems in Virtual Reality
Future Research
References
Data Mining for Adding Adaptive Interventions to Exploratory and Open-Ended Environments
Background and Motivation
Goals and Objectives
Approach and Expected Contributions
References
Formalising Human Mental Workload as Non-monotonic Concept for Adaptive and Personalised Web-Design
Introduction
Human Mental Workload
Argumentation Theory
Proposal
Evaluation Strategy
Key contributions
References
Detecting, Acquiring and Exploiting Contextual Information in Personalized Services
Introduction
Goals to Achieve
Work Done So Far
Contributions
References
Multi-source Provenance-aware User Interest Profiling on the Social Semantic Web
Introduction
Research Contributions
Current and Future Work
Conclusions and Future Research Questions
References
User Feedback and Preferences Mining
Introduction, Motivation, Contribution and Related Work
Models of User's Preference and Methods How to Learn Them
Conclusions and Future Work
References
Ubiquitous Fuzzy User Modeling for Multi-application Environments by Mining Socially Enhanced Online Traces
Introduction
User Model Construction by Mining Socially Enhanced Online Traces
Unified Fuzzy Hypergraph User Model Structure
Conclusion
References
Facilitating Code Example Search on the Web through Expertise Personalization
Research Problem and Solution Outline
Proposed Solutions
User Profiles Based on Programmer Expertise (RQ1)
Applying the User Profiles (RQ2 and RQ3)
Progress to Date
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken