Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

/pdf/the-theory-of-industrial-organization-19bejph0ej.pdf

The Theory of Industrial Organization

One distinctive feature of any adaptive system is the user model that represents essential information about each user This chapter complements other chapters of this book in reviewing user models and user modeling approaches applied in adaptive Web systems The presentation is structured along three dimensions: what is being modeled, how it is modeled, and how the models are maintained After a broad overview of the nature of the information presented in these various user models, the chapter focuses on two groups of approaches to user model representation and maintenance: the overlay approach to user model representation and the uncertainty-based approach to user modeling

User models for adaptive hypermedia and adaptive educational systems

With the exponentially growing amount of information available on the Internet, the task of retrieving documents of interest has become increasingly difficult. Search engines usually return more than 1,500 results per query, yet out of the top twenty results, only one half turn out to be relevant to the user. One reason for this is that Web queries are in general very short and give an incomplete specification of individual users' information needs. This paper explores ways of incorporating users' interests into the search process to improve the results. The user profiles are structured as a concept hierarchy of 4,400 nodes. These are populated by 'watching over a user's shoulder' while he is surfing. No explicit feedback is necessary. The profiles are shown to converge and to reflect the actual interests quite well. One possible deployment of the profiles is investigated: re-ranking and filtering search results. Increases in performance are moderate but noticeable and show that fully automatic creation of large hierarchical user profiles is possible.

/pdf/ontology-based-personalized-search-1p31p0v07s.pdf

Ontology based personalized search

Recommender systems for automatically suggested items of interest to users have become increasingly essential in fields where mass personalization is highly valued. The popular core techniques of such systems are collaborative filtering, content-based filtering and combinations of these. In this paper, we discuss hybrid approaches, using collaborative and also content data to address cold-start - that is, giving recommendations to novel users who have no preference on any items, or recommending items that no user of the community has seen yet. While there have been lots of studies on solving the item-side problems, solution for user-side problems has not been seen public. So we develop a hybrid model based on the analysis of two probabilistic aspect models using pure collaborative filtering to combine with users' information. The experiments with MovieLen data indicate substantial and consistent improvements of this model in overcoming the cold-start user-side problem.

Addressing cold-start problem in recommendation systems

We study a new task, proactive information retrieval by combining implicit relevance feedback and collaborative filtering. We have constructed a controlled experimental setting, a prototype application, in which the users try to find interesting scientific articles by browsing their titles. Implicit feedback is inferred from eye movement signals, with discriminative hidden Markov models estimated from existing data in which explicit relevance feedback is available. Collaborative filtering is carried out using the User Rating Profile model, a state-of-the-art probabilistic latent variable model, computed using Markov Chain Monte Carlo techniques. For new document titles the prediction accuracy with eye movements, collaborative filtering, and their combination was significantly better than by chance. The best prediction accuracy still leaves room for improvement but shows that proactive information retrieval and combination of many sources of relevance feedback is feasible.

/pdf/combining-eye-movements-and-collaborative-filtering-for-2d9tduszi3.pdf

Combining eye movements and collaborative filtering for proactive information retrieval

Digitalization of content and exponential growth of the Internet and electronic commerce are changing the media industry. The availability of structured content enables new ways to produce and deliver information. The paper explains the role of semantic metadata in developing content for an adaptive news service in the SmartPush-project. In SmartPush, news content is categorized using semi-automatic tools and pre-defined vocabularies. Metadata enhanced content is then matched against user profiles to provide customers with a personalized news service. After providing the personalized news to the customer, the SmartPush system adapts the personalization based on user feedback. The paper discusses the requirements of personalized content services and challenges in an approach based on structured metadata. We describe how supporting ontologies for the content were developed and maintained and what kinds of tools were developed to support the structured metadata creation. We also present some results of the pilot phase of the project and introduce some of the issues observed during the system implementation and in the performed field trial.

/pdf/the-role-of-structured-content-in-a-personalized-news-4tvvvzxeeu.pdf

The role of structured content in a personalized news service

https://research.cs.aalto.fi/pml/online-papers/ylipaavalniemi09.pdf

Dependencies between stimuli and spatially independent fMRI sources: towards brain correlates of natural stimuli.

The growing amount of new information has created a need for information filtering. Filtering is usually done on the textual content of the documents. Recent developments in the field of metadata suggest that filtering could be done according to only metadata (description of the actual document content). In information filtering documents are matched against user interest profiles. This is based on some measure for similarity or distance between a representation of documents and user profiles. Both the representation and the distance measure should make comparisons meaningful. A common problem is caused by closely related concepts that are considered independent in the representation model. Furthermore, the matching should not be considered symmetric, since documents may cover some area of interest very well and should be matched against parts of the user profile instead of the whole profile. In this paper we suggest a hierachical representation for describing documents and user profiles that attempts to model the related concepts. Our model includes an asymmetric distance measure that can also detect documents that cover some subtopic of an interest profile.

Metadata Based Matching of Documents and User Profiles

We propose a two-step approach for the analysis of func- tional magnetic resonance images, in the context of natural stimuli. In the first step, elements of functional brain activity emerge, based on spatial independence assumptions. The second step exploits temporal covariation between the elements and given features of the natural stimuli to identify functional networks. The networks can have complex activation patterns related to common task goals.

/pdf/functional-elements-and-networks-in-fmri-abjlruda5y.pdf

Eerika Savia

Papers

Combining eye movements and collaborative filtering for proactive information retrieval

The role of structured content in a personalized news service

Dependencies between stimuli and spatially independent fMRI sources: towards brain correlates of natural stimuli.

Metadata Based Matching of Documents and User Profiles

Functional Elements and Networks in fMRI