Summarization of OSS Developers IRC logs

From Master Projects
Jump to: navigation, search

About Summarization of OSS Developers IRC logs


A summary is a condensed version of the original. To summarize one should understand the language. There are many people whose daily work consists mainly, if not exclusively, in production of condensations or the summaries: summarized records of medical patients, to as well as facilitate the transfer of the patients for urgencies when the medical reports/ratios are immediately required; the journalist who gives an account of the results of an investigation; the judge who summarizes the obviousness is presented in front of the court; the general secretary who records the report of a board meeting; the scientist who reviews the recent publications in his field of research. If someone were to ask us what we did yesterday, what happened at the meeting earlier, what was on television last night or what’s on the news of late, we have to be able to once again explain everything quickly but relating all the facts, thus we have to be able to condense the information or in other words summarise. The summarization of the texts is an interesting and important field. Summarizing news articles or some text is easier when compared to summarization of chat logs. News articles are ‘static’ information because there is no interaction. Internet Relay Chats (IRC) or online discussions comprise ‘dynamic’ frequent exchanges between various participants. In this project we concentrate on summarizing the IRC which is essential for both as a resource and an organizational memory of activities of that project.