Ville H. Tuulos

Combining Topic Models and Social Networks for Chat Data Mining

Informal chat-room conversations have intrinsically different properties from regular static document collections. Noise, concise expressions and dynamic, changing and interleaving nature of discussions make chat data ill-suited for analysis with an off-the-shelf text mining method. On the other hand, interactive human communication has some implicit features which may be used to enhance the results. In our research we infer social network structures from the chat data by using a few basic heuristics. We then present some preliminary results showing that the inferred social graph may be used to enhance topic identification of a chat room when combined with a state-of-the-art topic and classification models. For validation purposes we then compare the performance effects of using this social information in a topic classification task.

A Scalable Topic-Based Open Source Search Engine

Site-based or topic-specific search engines work with mixed success because of the general difficulty of the information retrieval task, and the lack of good link information to allow authorities to be identified. We are advocating an open source approach to the problem due to its scope and need for software components. We have adopted a topic-based search engine because it represents the next generation of capability. This paper outlines our scalable system for site-based or topic-specific search, and demonstrates the developing system on a small 250,000 document collection of EU and UN web pages.

Multi-faceted information retrieval system for large scale email archives

We profile a system for search and analysis of large-scale email archives. The system builds around four facets: Content-based search engine, statistical topic model, automatically inferred social networks and time-series analysis. The facets correspond to the types of information available in email data. The presented system allows chaining or combining the facets flexibly. Results of one facet may be used as input to another, yielding remarkable combinatorial power. In information retrieval point of view, the system provides support for exploration, approximate textual searches and data visualization. We present some experimental results based on a large real-world email corpus.

Story Mashup: design and evaluation of novel interactive storytelling game for mobile and web users

This paper studies the design rationale and evaluation of an urban storytelling game called Story Mashup. In the game ubiquitous computing infrastructure is utilized to facilitate real-time interaction between mobile and web users. Textual stories written in the web by certain people are illustrated by other people taking matching photos with camera phones. Complete stories are then displayed on a large public display and on the web. To carry out a thorough empirical evaluation of the game design in a real world setting, the game was played in New York in September 2006 with 180 players and by people in the internet around the world. The results show that the adopted iterative design process succeeded in achieving the goals set for usability, user experience and game stimulation.

Utilizing Rich Bluetooth Environments for Identity Prediction and Exploring Social Networks as Techniques for Ubiquitous Computing

Personal identification and using that information is in the heart of many ubiquitous systems. We present two complementary techniques, namely personal identification without directly observing the subject, and using that information for understanding the social relations between the subjects. We show that with certain presumptions it is possible to predict ones identity with reasonable certainty only by observing ones Bluetooth neighborhood without the need to directly observe the subject. We also show how this information can be used for exploring the social relations between the subjects

Multi-Faceted Information Retrieval System for Large Scale Email Archives

We profile a system for search and analysis of large-scale email archives. The system builds around four facets: content-based search engine, statistical topic model, automatically inferred social networks, and time-series analysis. The facets correspond to the types of information available in email data. The presented system allows chaining or combining the facets flexibly. Results of one facet may be used as input to another yielding remarkable combinatorial power. In information retrieval point of view, the system provides support for exploration, approximate textual searches and data visualization. We present some experimental results based on a large real-world email corpus.

On text-based estimation of document relevance

This work is part of a proactive information retrieval project that aims at estimating relevance from implicit user feedback. The noisy feedback signal needs to be complemented with all available information, and textual content is one of the natural sources. Here we take the first steps by investigating whether this source is at all useful in the challenging setting of estimating the relevance of a new document based on only few samples with known relevance. It turns out that even sophisticated unsupervised methods like multinomial PCA (or latent Dirichlet allocation) cannot help much. By contrast, feature extraction supervised by relevant auxiliary data may help.

GS textplorer -: adaptive framework for information retrieval

TheWEBSOM is a method developed originally at Helsinki University of Technology for analyzing and visualizing large document collections [1, 4]. In the WEBSOM method, the self-organizing map algorithm [3] is used to automatically organize collections of documents on a map to enable easy exploration and search of the collection. Map regions that are close to each other contain similar documents. The main objectives in developing the WEBSOM method has been to o er a method for exploring text collections that is di erent from the queryresult approach, enabling the user to get an overall view to the document collection. Moreover, there are no principled limits on the type of text material that the method can handle.