This webpage is for an old version of the course; content may be out of date!

CSE 291: Trends in Recommender Systems and Human Behavioral Modeling

Autumn 2017, Monday/Wednesday 9:30-10:50, CSE 4140

CSE 291 is a graduate course devoted to current trends for recommender systems and models of human behavior. This course covers material similar to CSE 258, though is more focused on research papers and student-led presentations. It is suggested (but not required) that you have already taken CSE 258 or are taking it concurrently, though the first 1-2 lectures will be spent on revision.

The course meets twice a week on Monday/Wednesday mornings, starting October 2. Meetings are in CSE 4140. Attendance is expected, as grades are primarily based on participation.

Office hours: I'll hold office hours on Tuesdays 9:00-13:00 in CSE 4102.

	piazza page

	assignment spec (50% of grade)

	in-class presentations (50% of grade)	example discussion questions

1	Background

Monday October 2: Matrix Factorization

• Matrix Factorization Techniques for Recommender Systems (Koren et al., 2009) | pdf
• The Bellkor 2008 Solution to the Netflix Prize (Bell et al., 2008) | pdf
• Application: Recommending Product Sizes to Customers (Sembium et al., 2016) | pdf

Wednesday October 4: Item-to-Item Recommendation

• BPR: Bayesian Personalized Ranking from Implicit Feedback (Rendle et al., 2012) | pdf
• Learning Visual Clothing Style with Heterogeneous Dyadic Co-occurrences (Veit et al., 2015) | pdf
• Application: Amazon.com Recommendations: Item-to-item Collaborative Filtering (Linden et al., 2003) | pdf

Suggested reading:

• Factorization Meets the Neighborhood: A Multifaceted Collaborative Filtering Model (Koren, 2008) | pdf
• One-Class Collaborative Filtering (Paquet and Koenigstein, 2013) | pdf
• Factorization Machines (Rendle, 2010) | pdf
• Factorization Meets the Item Embedding: Regularizing Matrix Factorization with Item Co-occurrence (Liang et al., 2016) | pdf
• Collaborative Metric Learning (Hsieh et al., 2017) | pdf

	course outline slides

2	Ranking, Retrieval, & Sequential Recommendation

Monday October 9: Markov-Chains

• Factorizing Personalized Markov Chains for Next-Basket Recommendation (Rendle et al., 2010) | pdf
• Personalized Ranking Metric Embedding for Next New POI Recommendation (Feng et al., 2015) | pdf
• Translation-based Recommendation (He et al., 2017) | pdf

Wednesday October 11: More Markov Chains, and Metric Learning

• Learning to Rank with Trust and Distrust in Recommender Systems (Rafailidis and Crestani, 2017) | pdf
• Modeling User Consumption Sequences (Benson et al., 2016) | pdf
• Application: Playlist Prediction via Metric Embedding (Chen et al., 2012) | pdf

Suggested reading:

• Learning Hierarchical Representation Model for Next-Basket Recommendation (Wang et al., 2015) | pdf
• Fifty Shades of Ratings: How to Benefit from Negative Feedback in Top-N Recommendations Tasks (Frolov and Oseledets, 2016) | pdf
• Do "Also-Viewed" Products Help User Rating Prediction? (Park et al., 2017) pdf
• A Dynamic Recurrent Model for Next Basket Recommendation (Yu et al. 2016) | pdf
• Neural Ranking Models with Weak Supervision (Dehghani et al. 2017) | pdf
• Personalizing Session-based Recommendations with Hierarchical Recurrent Neural Networks (Quadrana et al. 2017)
• Bridging Collaborative Filtering and Semi-Supervised Learning: A Neural Approach for POI recommendation (Yang et al. 2017) | pdf

	assignment slides	discussion qs

3	Temporal Dynamics

Monday, October 16:

• Time Weight Collaborative Filtering (Ding and Li, 2005) | pdf
• Collaborative Filtering with Temporal Dynamics (Koren, 2010) | pdf
• Latent Factor Transition for Dynamic Collaborative Filtering (Zhang et al., 2014) | pdf
• Who, What, When, and Where: Multi-Dimensional Collaborative Recommendations Using Tensor Factorization on Sparse User-Generated Data (Bhargava et al., 2015) | pdf
• TribeFlow: Mining & Predicting User Trajectories (Figueiredo et al., 2016) | pdf

Wednesday, October 18:

• Temporal Effects on Hashtag Reuse in Twitter: A Cognitive-Inspired Hashtag Recommendation Approach (Kowald et al., 2017) | pdf
• On the Temporal Dynamics of Opinion Spamming – Case Studies on Yelp (Kc and Mukherjee, 2016) | pdf
• Personalized Itinerary Recommendation with Queuing Time Awareness (Lim et al., 2017) | pdf
• Optimizing the Recency-Relevancy Trade-off in Online News Recommendations (Chakraborty et al., 2017) | pdf
• Application: User Session Identification Based on Strong Regularities in Inter-activity Time (Halfaker et al., 2015) | pdf

Suggested reading:

• Recurrent Poisson Factorization for Temporal Recommendation (Hosseini et al. 2017)
• A Temporally Heterogeneous Survival Framework with Application to Social Behavior Dynamics (Yu et al. 2017) | pdf

	sign-up sheet	slide deck	discussion qs

4	Deep and Content-Aware

Monday, October 23:

• Recurrent recommender networks (Wu et al., 2017) | pdf
• Neural Collaborative Filtering (He et al., 2017) | pdf
• Sequential User-based Recurrent Neural Network Recommendations (Donkers et al., 2017) | pdf
• Collaborative Variational Autoencoder for Recommender Systems (Li and She, 2017) | pdf
• TransNets: Learning to Transform for Recommendation (Catherine and Cohen, 2017) | pdf

Wednesday, October 25:

• Neural Factorization Machines for Sparse Predictive Analytics (He and Chua, 2017) | pdf
• What Your Images Reveal: Exploiting Visual Contents for Point-of-Interest Recommendation (Wang et al., 2017) | pdf
• Deep Neural Networks for YouTube Recommendations (Covington et al., 2016) | pdf
• 3D Convolutional Networks for Session-based Recommendation with Content Features (Tuan and Phuong, 2017) | pdf
• Deep Learning based Large Scale Visual Recognition and Search for E-Commerce (Shankar et al., 2017) | pdf

Suggested reading:

• On Sampling Strategies for Neural Network-based Collaborative Filtering (Chen et al. 2017) | pdf
• Attentive Collaborative Filtering: Multimedia Recommendation with Item- and Component-Level Attention (Chen et al. 2017) | pdf
• Embedding Factorization Models for Jointly Recommending Items and User Generated Lists (Cao et al. 2017) | pdf
• Collaborative Knowledge Base Embedding for Recommender Systems (Zhang et al. 2017) | pdf

	sign-up sheet	slide deck	discussion qs

5	Text and Question-Answering Systems

Monday, October 30:

• User Review Sites as a Resource for Large-Scale Sociolinguistic Studies (Hovy et al., 2015) | pdf
• Detecting Evolution of Concepts based on Cause-Effect Relationships in Online Reviews (Zhang et al., 2016) | pdf
• Neural Rating Regression with Abstractive Tips Generation for Recommendation (Li et al., 2017) | pdf
• Extracting and Ranking Travel Tips from User-Generated Reviews (Guy et al., 2017) | pdf
• Exploring Latent Semantic Factors to Find Useful Product Reviews (Mukherjee et al., 2017) | pdf
• Ask the GRU: Multi-task Learning for Deep Text Recommendations (Bansal et al., 2016) | pdf

Wednesday, November 1:

• Novelty based Ranking of Human Answers for Community Questions (Omari et al., 2016) | pdf
• Understanding How People Use Natural Language to Ask for Recommendations (Kang et al., 2017) | pdf
• Summarizing Answers in Non-Factoid Community Question-Answering (Song et al., 2017) | pdf
• Characterizing and Predicting Enterprise Email Reply Behavior (Yang et al., 2017) | pdf
• Application: Smart Reply: Automated Response Suggestion for email (Kannan et al., 2016) | pdf
• Application: Efficient Natural Language Response Suggestion for Smart Reply (Henderson et al., 2017) | pdf

Suggested reading:

• Representativeness-aware Aspect Analysis for Brand Monitoring in Social Media (Liao et al. 2017) | pdf
• A Semantic-Aware Profile Updating Model for Text Recommendation (Zagheli et al. 2017) | pdf
• End-to-end Learning for Short Text Expansion (Tang et al. 2017) | pdf

	sign-up sheet	slide deck	discussion qs

6	Project Proposals

Monday Nov. 6/Wednesday Nov. 8

• Everyone will briefly describe their project proposals in order to discuss them with the class.

	sign-up sheet	slide deck

7	Novel Tasks & Domains

Monday, November 13:

• Domain-Aware Grade Prediction and Top-n Course Recommendation (Elbadrawy and Karypis, 2016) | pdf
• Gaze prediction for recommender systems (Zhao et al., 2016) | pdf
• A Novel Recommender System for Helping Marathoners to Achieve a New Personal-Best (Smyth and Cunningham, 2017) | pdf
• Chemical Reactant Recommendation Using a Network of Organic Chemistry (Savage et al., 2017) | pdf
• Groove Radio: A Bayesian Hierarchical Model for Personalized Playlist Generation (Lavee et al., 2017) | pdf

Wednesday, November 15:

• Exploiting Food Choice Biases for Healthier Recipe Recommendation (Elsweiler et al., 2017) | pdf
• Personalized Key Frame Recommendation (Chen et al., 2017) | pdf
• Meta-Graph Based Recommendation Fusion over Heterogeneous Information Networks (Zhao et al. 2017) | pdf
• Dynamic Attention Deep Model for Article Recommendation by Learning Human Editors Demonstration (Wang et al. 2017) | pdf
• Multi-Modality Disease Modeling via Collective Deep Matrix Factorization (Wang et al. 2017) | pdf

	sign-up sheet	slide deck	discussion qs

8	Socially & Geographically Aware Models

Monday, November 20:

• A Probabilistic Model for Using Social Networks in Personalized Item Recommendation (Chaney et al., 2015) | pdf
• Online popularity and topical interests through the lens of instagram (Ferrara et al., 2014) | pdf
• The Effect of Recommendations on Network Structure (Su et al., 2016) | pdf
• Recommendations in Signed Social Networks (Tang et al., 2016) | pdf
• STAR: Semiring trust inference for trust-aware social recommenders (Gao et al., 2016) | pdf
• Fairness-Aware Group Recommendation with Pareto-Efficiency (Xiao et al., 2017) | pdf

Wednesday, November 22:

• Additive Co-Clustering with Social Influence for Recommendation (Du et al., 2017) | pdf
• A General Model for Out-of-town Region Recommendation (Pham et al., 2017) | pdf
• Growing Wikipedia Across Languages via Recommendation (Wulczyn et al., 2016) | pdf
• Social Collaborative Viewpoint Regression with Explainable Recommendations (Ren et al. 2017) | pdf
• Exploiting Socio-Economic Models for Lodging Recommendation in the Sharing Economy (Vazquez et al. 2017) | pdf
• A Location-Sentiment-Aware Recommender System for Both Home-Town and Out-of-Town Mobile Users (Wang et al. 2017) | pdf

	sign-up sheet	slide deck	discussion qs

9	Human-subject experiments, evaluation, and deployment

Monday, November 27:

• When do Recommender Systems Work the Best?: The Moderating Effects of Product Attributes and Consumer Reviews on Recommender Performance (Lee and Hosanagar, 2016) | pdf
• Leading the Herd Astray: An Experimental Study of Self-fulfilling Prophecies in an Artificial Cultural Market (Salganik and Watts, 2008) | pdf
• Using Navigation to Improve Recommendations in Real-time (Wu et al., 2013) | pdf
• An Empirical Analysis of Algorithmic Pricing on Amazon Marketplace (Chen et al., 2016) | pdf
• Modeling the Assimilation-Contrast Effects in Online Product Rating Systems Debiasing and Recommendations (Zhang et al. 2017) | pdf

Wednesday, November 29:

• Post Processing Recommender Systems for Diversity (Antikacioglu and Ravi, 2017)
• The Role of Social Networks in Information Dissemination (Bakshy et al., 2012)
• Personalized Key Frame Recommendation (Chen et al., 2017)
• Online Popularity and Topical Interests Through the Lens of Instagram (Ferrara et al., 2014)

	sign-up sheet	slide deck	discussion qs

10	Project Presentations

	sign-up sheet	slide deck