Tutorials

Tutorials Venue : The Guinness Storehouse Conference Centre
Download the PDFs of the conference venues at a glance sheet and the schedule for Workshops & Tutorials.

Tutorials Date : 18th April 2011

Information: There will be ECIR and Guinness staff on hand all day to send you, upon arival, to the registration desk on the second floor  of the storehouse.

Risk Management in Information Retrieval Designing Effective Search and Discovery Experiences
8:15 - 20:00 Registration   8:15 - 20:00 Registration
9:00 - 10:30 Session A   9:00 - 10:30 Session A
10:30 - 11:00 Coffee   10:30 - 11:00 Coffee
11:00 - 13:00 Session B   11:00 - 13:00 Session B / Close
13:00 - 14:00 Lunch   13:00 - 14:00 Lunch
 
Online Advertising: An Information Scientist's Perspective Web Search: The Role of the Users
14:00 - 15:30 Session A   14:00 - 15:30 Session A
15:30 - 16:00 Coffee   15:30 - 16:00 Coffee
16:00 - 18:00 Session B / Close   16:00 - 18:00 Session B / Close
18:00 - 19:30 Storehouse - Access to the Visitor Experience   18:00 - 19:30 Storehouse - Access to the Visitor Experience
19:30 - 23:00 ECIR 2011 Welcome Reception in the Guinness Storehouse Gravity Bar   19:30 - 23:00 ECIR 2011 Welcome Reception in the Guinness Storehouse Gravity Bar



Risk Management in Information Retrieval

Risk modelling and management are an emerging concept in Information Retrieval (IR) theory and modelling. A new type of theoretical research that regards the retrieval results as a whole has significantly departed from the classic information retrieval methodologies originated from the Probability Ranking Principle, the Robertson-Spärck Jones model (the resulting BM25 formula), and the statistical language models. The recent progress in this research direction has been made by taking an analogy with the financial risk management. It has been demonstrated that the ideas about portfolio retrieval and retrieval risks provide a sounded theoretical framework and more powerful mathematical tool for us to understand and analyze retrieval processes such as query expansion, document ranking, and evaluation.

Objectives: The tutorial is aimed at providing a comprehensive introduction of the emerging portfolio retrieval and risk modelling and management techniques in information retrieval systems. We will provide a unified account of various information retrieval models from this new angle. While the basic theories (such as portfolio theory of IR and mean-variance analysis) and risk models of information retrieval are covered, the tutorial is also focused on the resulting practical algorithms of query expansion, relevance ranking, IR metric optimization as well as their performance evaluations. Practical IR applications such as Web search engines, multimedia retrieval, and collaborative filtering will also be covered. A discussion of new opportunities for future research and applications will also be given.

Duration: A half-day (3 hours plus breaks)

Speakers:



Online advertising: an information scientist's perspective

Over the past 15 years online advertising, a billion industry worldwide in 2008, has been pivotal to the success of the World Wide Web. This tutorial will review the main business models of online advertising including: the pay-per-impression model (CPM); and the pay-per-click model (CPC); a relative new comer, the pay-per-action model (CPA), where an action could be a product purchase, a site visit, a customer lead, or an email signup; and dynamic CPM (dCPM) which optimizes a campaign towards the sites and site sections that perform best for the advertiser. In addition, this tutorial will discuss in detail the technologies that have transformed advertising from the low-tech, human intensive, "Mad Men" way of doing work (that were common place for much of the 20th century and the early days of online advertising) to the highly optimized, mathematical, computer-centric processes that form the backbone of many current online advertising systems. These new approaches to advertising have many parallels with, and build on experience from the financial markets of Wall Street. For example, online advertising slots on a publisher can be sold in a futures market (selling publisher real estate in advance with various guarantees) or in the spot market (selling publisher real estate in a real-time manner). Both marketplaces leverage techniques from the fields of machine learning (e.g., logistic regression, online learning), statistics (e.g., binomial maximum likelihood), information retrieval (vector space model, BM25, ranking systems) and economics (auction mechanisms, game theory). Challenges such as click fraud (the spam of online advertising), deception, privacy and other open issues will also be discussed. Some new developments in online advertising such as social advertising, behavioral targeting will be reviewed also.

Objectives: This half-day tutorial (3 hours) focuses on helping researchers and developers attending ECIR develop new skills as well providing an account of the latest developments in this fast evolving and diverse discipline of online advertising. In doing so it will provide a clear and detailed overview of the technologies and business models that are transforming the field of (online) advertising along the following themes: sponsored search, display advertising, business models; market size and scope; forward and spot markets; machine learning and statistical technologies; economic models; new directions such as social advertising, realtime bidded exchanges (RTB) and behavioral targeting; challenges and open issues that face online advertising.

Participants will learn about: The primary business models that make online advertising; The online advertising markets, their revenues and sizes in key markets such as Europe, USA and Asia; Forward and spot markets; Information retrieval models; Sponsored search; Display advertising; Contextual Advertising; Linear programming, quadratic programming, Markowitz model; Auction models, game theory and how it can be used to analyze online auctions; Metrics and evaluation practices (AB, Fractional Factorial design); Collaborative filtering; Bayesian modeling; Learning to rank; Predict ad quality/click-thru-rates; Automatically targeting ads both in sponsored search and in contextual advertising; Understand the specific issues that face online advertising such as privacy, deception and fraud (in particular, click fraud, the spam of online advert; How advertising is being leveraged in Web 2.0 applications such as social networks, and video/photo-sharing; New directions: behavioral targeting, social advertising, exchanges, data exchanges; Open research issues facing the field of online advertising

Duration: A half-day (3 hours plus breaks)

Speakers:

  • Dr. James G. Shanahan (Independent Consultant, San Francisco)


Web Search: The Role of the Users

Web retrieval methods have evolved through three major steps in the last decade or so. They started from standard document-centric IR in the early days of the Web, then made a major step forward by leveraging the structure of the Web, using link analysis techniques in both crawling and ranking challenges. A more recent, no less important but maybe more discrete step forward, has been to enter the user in this equation in two ways:

  • Implicitly, through the analysis of usage data captured by query logs, and session and click information in general; the goal here being to improve ranking as well as to measure user's happiness and engagement.
  • Explicitly, by offering novel interactive features; the goal here being to better answer users' needs.

Hence, the tutorial will cover the user-related challenges associated with the implicit and explicit role of users in Web retrieval. More specifically, we will review and discuss challenges associated with two types of activities, namely:

  • Usage data analysis and metrics - It is critical to monitor how users take advantage and interact with Web retrieval systems, as this implicit relevant feedback aggregated at a large scale, can spproximate quite accurately the level of success of a given feature. Here we have to consider not only clicks statistics but also the time spent in a page, the number of actions per session, etc.
  • User interaction - Given the intrinsic problems posed by the Web, the key challenge for the user is to conceive a good query to be submitted to the search system, one that leads to a manageable and relevant answer. The retrieval system must complete search requests fast and give back relevant results, even for poorly formulated queries, as is the common case in the Web. Web retrieval engines thus interact with the user at two key stages:
  • Expressing a query: Human beings have needs or tasks to accomplish, which are frequently not easy to express as ``queries''. Queries, even when expressed in a more natural manner, are just a reflection of human needs and are thus, by definition, imperfect. This phenomenon could be compared to Plato's cave metaphor, where shadows are mistaken for reality.
  • Interpreting results: Even if the user is able to perfectly express a query, the answer might be split over thousands or millions of Web pages or not exist at all. In this context, numerous questions need to be addressed. Examples include: How do we handle a large answer? How do we rank results? How do we select the documents that really are of interest to the user? Even in the case of a single document candidate, the document itself could be large. How do we browse such documents efficiently?

Objectives: The goal of this tutorial is thus to teach the key principles of user interactions in modern Web search engines, make attendees aware of what's happening behind the scene and hopefully foster future research.

Duration: A half-day (3 hours plus breaks)

Speakers:

  • Ricardo Baeza-Yates, VP of Yahoo! Research for Europe and Latin America, Leading the labs at Barcelona, Spain and Santiago, Chile.
  • Yoelle Maarek, Senior Director of Yahoo! Research in Israel


Designing Effective Search and Discovery Experiences

This half-day tutorial provides a practical introduction to Human Centred Design for information retrieval, access and discovery. We present a concise overview of the fundamental concepts, principles and models of human information-seeking behaviour and show how to apply these in the design of information search and discovery user experiences. A key element of the tutorial is the opportunity to practice these skills in a group exercise.

Objectives: Our aim is to deliver a learning experience grounded in good scholarship, integrating the latest and most significant research findings with insights derived from practical experience of designing and evaluating discovery applications; delivered in a manner that focuses on transferable, practical skills that can be learnt and practiced within a 3.5 hour session. In this tutorial participants will learn:

  • the fundamental concepts and principles of Design for Discovery
  • how to differentiate between various types of search behaviour: known-item, exploratory, etc.
  • models of human information-seeking behaviour (e.g. Broder, Norman, Marchionini, Saracevic, Bates, etc.) , and how to apply interaction design principles based on those models
  • an understanding of the key variables of user type, goal and mode of interaction, and how to apply these variables when designing for different user contexts
  • the role of design patterns, and how to apply UI design patterns from patterns.endeca.com and those of other pattern libraries in designing search user interfaces
  • an awareness of the key design resources available within the HCIR community and how to apply these to practical design challenges

Duration: Half day

Speakers:

  • Dr. Tony Russell-Rose & Dr. Mark Burrell

Supported By

  • Dublin City University
  • University of Sheffield
  • The British Computer Society