Tutorials

Invited Tutorial: Mining Social Networks for Recommendation

by Martin Ester (Simon Fraser University, Canada)

With the emergence of online social networks, academia and industry have explored ways to exploit the information in social networks to improve the quality of recommendations and to support new recommendation tasks. The underlying motivation is to capture the effects that govern the evolution of social networks, i.e. social influence, selection, correlational influence and transitivity, to enhance the typically very sparse rating matrix. Recommender systems exploiting a social network promise to outperform traditional recommenders in particular for cold-start (new) users who have not yet provided enough information about their preferences. After introducing the motivation and some of the practical applications, we discuss social networks and the main factors affecting their evolution. We then review state-of-the-art methods for item recommendation in social networks, both memory-based approaches and model-based approaches, in particular matrix factorization. We discuss friend recommendation, an important recommendation task that is unique to the context of social networks. We conclude the tutorial with a discussion of future research directions such as privacy-preserving recommendations and social recommendation in distributed/peer-to-peer networks.

Outline

1. Introduction
2. Social networks and the effects that govern their evolution
3. Memory-based approaches for item recommendation in social networks
4. Model-based approaches for item recommendation in social networks
5. Friend recommendation
6. Future directions

This tutorial targets researchers who want to get up to speed in this emerging research area as well as practitioners who are interested in developing their own applications. The tutorial assumes familiarity with the common methods of recommender systems. Some background in data mining and social network analysis will be helpful, but is not required.

Date

Oct 12, 2013 (08:30 – 10:15)

Room

LT-13

Slides

• Download slides (PDF)

ACM Digital Library

• Recommendation in social networks by Martin Ester

Learning to Rank

by Alexandros Karatzoglou (Telefonica Research, Spain), Linas Baltrunas (Telefonica Research, Spain) & Yue Shi (Delft University of Technology, Netherlands)

Recommender system aim at providing a personalized list of items ranked according to the preferences of the user, as such ranking methods are at the core of many recommendation algorithms. The topic of this tutorial focuses on the cutting-edge algorithmic development in the area of recommender systems. This tutorial will provide an in depth picture of the progress of ranking models in the field, summarizing the strengths and weaknesses of existing methods, and discussing open issues that could be promising for future research in the community. A qualitative and quantitative comparison between different models will be provided while we will also highlight recent developments in the areas of Reinforcement Learning.

Outline

1. Goals
2. Background
3. Convectional ranking models in RecSys
4. Learning to rank for recommender systems
5. Categorization and main contributions
6. Interactive recommendation (reinforcement learning/Bandits)
7. Open issues

The tutorial is intended for researchers and practitioners in the area of recommender systems, particularly those who are interested in novel algorithms. Only basic knowledge about recommender systems is required.

Date

Oct 12, 2013 (14:00 – 15:45)

Room

LT-13

ACM Digital Library

• Learning to rank for recommender systems by Alexandros Karatzoglou, Linas Baltrunas, Yue Shi

Beyond Friendship:­ The Art, Science and Applications of Recommending People to People in Social Networks

by Luiz Augusto Pizzato (University of Sydney, Australia) & Anmol Bhasin (LinkedIn, USA)

While Recommender Systems are powerful drivers of engagement and transactional utility in social networks, People recommenders are a fairly involved and diverse subdomain. Consider that movies are recommended to be watched, news is recommended to be read, people however, are recommended for a plethora of reasons – such as recommendation of people to befriend, follow, partner, targets for an advertisement or service, recruiting, partnering romantically and to join thematic interest groups.

This tutorial aims to first describe the problem domain, touch upon classical approaches like link analysis and collaborative filtering and then take a rapid deep dive into the unique aspects of this problem space like Reciprocity, Intent understanding of recommender and the recomendee, Contextual people recommendations in communication flows and Social Referrals – a paradigm for delivery of recommendations using the Social Graph. These aspects will be discussed in the context of published original work developed by the authors and their collaborators and in many cases deployed in massive-scale real world applications on professional networks such as LinkedIn.

Outline

1. Introduction
   • The basics of Social Recommenders
   • People recommender systems
2. Special Topics in People Recommenders
   • Why reciprocal (people) recommenders are different to traditional (product) recommendations
   • Multi-Objective Optimization
   • Intent Understanding
   • Feature Engineering
   • Social Referral
   • Pathfinding
3. Concluding remarks

The pre-requisite for this tutorial is some familiarity with foundational Recommender Systems, Data Mining, Machine Learning and Social Network Analysis literature.

Date

Oct 13, 2013 (08:30 – 10:15)

Room

LT-13

Slides

• Slides by Anmol Bhasin (Slideshare)
• Slides by Luiz Augusto Pizzato (Slideshare)

ACM Digital Library

• Beyond friendship: the art, science and applications of recommending people to people in social networks by Luiz Augusto Pizzato, Anmol Bhasin

Preference Handling

by Alexis Tsoukiàs (Université Paris Dauphine, France) & Paolo Viappiani (Université Pierre et Marie Curie, France)

The tutorial aims at introducing the general field that deals with techniques for representing, learning and reasoning with preferences. While the community of Recommender Systems has recently done extensive work in techniques for providing user recommendations in different settings, preferences have been formally studied since several decades (the community that studies preferences from a formal algorithmic point of view is now often called algorithmic decision theory). Preferences are a basic concept for several research fields such as economics, decision theory, game theory, artificial intelligence, classification, data bases, etc.; preferences constitute a key concept for the development of recommender systems since they represent the elementary information upon which recommendations are constructed. The tutorial will introduce both the formalisms more widely used to model and represent preferences as well as the procedures aimed at learning preferences and at producing a recommendation for an end-user (decision maker).

Outline

1. General introduction, Historical Perspective
2. Axiomatic, Behavioral and Prescriptive approaches
3. Binary Relations, Languages for preferences
4. Preference Modeling, representation theorems, foundations of utility theory
5. Decision-making under certainty and uncertainty
6. Measuring Preferences, Compact representations
7. Preference Learning, utility elicitation
8. Preference Aggregation
9. Constructing a recommendation
10. Preferences in Argumentation
11. Applications and Discussion

The tutorial is mainly addressed to researchers in the area of recommendation systems. Practitioners willing to understand how to use correctly preferences as well as the algorithmic challenges may also be interested in attending. Basic notions of discrete mathematics and probability theory are the only prerequisite.

Date

Oct 13, 2013 (14:00 – 15:45)

Room

LT-13

ACM Digital Library

• Tutorial on preference handling by Alexis Tsoukiàs, Paolo Viappiani