Recently, there have been significant advances in several areas of language technology, including clustering, text categorization, and summarization. However, efforts to combine technology from these areas in a practical system for information access have been limited. In this paper, we present Columbia's Newsblaster system for online news summarization. Many of the tools developed at Columbia over the years are combined together to produce a system that crawls the web for news articles, clusters them on specific topics and produces multidocument summaries for each cluster.

This paper describes a domain-independent, machine-learning based approach to temporally anchoring and ordering events in news. The approach achieves 84.6% accuracy in temporally anchoring events and 75.4% accuracy in partially ordering them.

This paper describes a multidocument summarizer built upon research into the detection of new information. The summarizer uses several new strategies to select interesting and informative sentences, including an innovative measure of importance derived from the analysis of a large corpus. The system also computes concept frequencies rather than word frequencies as an additional measure of importance. It merges these strategies with a number of familiar summarization heuristics to rank sentences. The initial version of the summarizer performed successfully in the evaluation reported at the Document Understanding Conference last year, although the system addressed only the content of the summary and not the presentation. We also discuss here the procedures we are developing to improve the presentation and readability of the summaries.

We demonstrate the value of using context in a new-information detection system that achieved the highest precision scores at the Text Retrieval Conference's Novelty Track in 2004. In order to determine whether information within a sentence has been seen in material read previously, our system integrates information about the context of the sentence with novel words and named entities within the sentence, and uses a specialized learning algorithm to tune the system parameters.

This paper describes experiment's in the automat'ic construction of lexicons that would be useflfl in searching large document collect'ions tot text frag~ ments tinct address a specific inibrmation need, such as an answer to a quest'ion.