Analysis of Large Text Datasets, Project Page

Overview

This project aims at the development of new and improvement of the existing computer methods for the analysis of large text datasets. Special emphasis is put on the analysis of Slovenian text. The developed methods will enable automatic document categorization of Slovenian (and potentially also some other natural languages) text, adaptation of the existing methods for text-learning to Slovenian texts, analysis of text datasets based on the new, extended document representation and better Web browsing by using a personal browsing assistant based on the new text analysis methods. The development of different applications will be enabled, including automatic updating of some existing document categorizations that are currently updated manually.

More about our Word-Mining efforts

• Mladenic, D., (2002) Modeling information in textual data combining labeled and unlabeled data ESF Exploratory Workshop on Pattern Detection and Discovery in Data Mining, Lecture Notes in Computer Science, Springer, 2002 (ESF Exploratory Workshop)
• Mladenic, D., (2002) Learning word normalization using word suffix and context from unlabeled data Proceedings of the Nineteenth International Conference on Machine Learning (ICML-2002)
• Mladenic, D., (2001) Using unlabeled data and sequential modeling for classifying words Text Mining Workshop at the 2001 IEEE International Conference on Data Mining (ICDM-2001)

Project Members

• Marko Grobelnik
• Tom Mitchell
• Dunja Mladenic

This project is strongly related to the Yahoo Planet project, Personal WebWatcher project and the PhD thesis project: Machine Learning on non-homogeneous, distributed text data