Abstract:
In this thesis work we consider the problem of analyzing the development of a document collection over time without requiring meaningful citation data. Given a collection of time stamped documents, we formulate and explore the following two questions. First, what are the main topics and how do these topics develop over time? Second, what are the documents and who are the authors that are most influential in this process?. We propose methods addressing these questions by taking solely text of the document as input. Because proposed methods use only the text of the documents as input, the methods are applicable to a much wider range of document collections (email, blogs, etc.), most of which lack meaningful citation data. We evaluate our methods on two kinds of data sets one is the documents from the proceedings of the Neural Information Processing Systems (NIPS) conference and the other is collection of news articles. The results show that the methods are effective and that addressing the questions based on the text alone . In fact, the text-based methods sometimes even identify influential papers that are missed by citation analysis.
