We propose a new unsupervised learning technique for extracting information about authors and topics from large text collections. We model documents as if they were generated by a two-stage stochastic process. An author is represented by a probability distribution over topics, and each topic is represented as a probability distribution over words. The probability distribution over topics in a multi-author paper is a mixture of the distributions associated with the authors. The topic-word and author-topic distributions are learned from data in an unsupervised manner using a Markov chain Monte Carlo algorithm. We apply the methodology to three large text corpora: 150,000 abstracts from the CiteSeer digital library, 1,740 papers from the Neural Information Processing Systems (NIPS) Conferences, and 121,000 emails from the Enron corporation. We discuss in detail the interpretation of the results discovered by the system including specific topic and author models, ranking of authors by topic and topics by author, parsing of abstracts by topics and authors, and detection of unusual papers by specific authors. Experiments based on perplexity scores for test documents and precision-recall for document retrieval are used to illustrate systematic differences between the proposed author-topic model and a number of alternatives. Extensions to the model, allowing (for example) generalizations of the notion of an author, are also briefly discussed.

The primary purpose of news articles is to convey information about who, what, when and where. But learning and summarizing these relationships for collections of thousands to millions of articles is difficult. While statistical topic models have been highly successful at topically summarizing huge collections of text documents, they do not explicitly address the textual interactions between who/where, i.e. named entities (persons, organizations, locations) and what, i.e. the topics. We present new graphical models that directly learn the relationship between topics discussed in news articles and entities mentioned in each article. We show how these entity-topic models, through a better understanding of the entity-topic relationships, are better at making predictions about entities.

Abstract. Statistical language models can learn relationships between topics discussed in a document collection and persons, organizations and places mentioned in each document. We present a novel combination of statistical topic models and named-entity recognizers to jointly analyze entities mentioned (persons, organizations and places) and topics discussed in a collection of 330,000 New York Times news articles. We demonstrate an analytic framework which automatically extracts from a large collection: topics; topic trends; and topics that relate entities.

Human-defined concepts are fundamental building-blocks in constructing knowledge bases such as ontologies. Statistical learning techniques provide an alternative automated approach to concept definition, driven by data rather than prior knowledge. In this paper we propose a probabilistic modeling framework that combines both human-defined concepts and data-driven topics in a principled manner. The methodology we propose is based on applications of statistical topic models (also known as latent Dirichlet allocation models). We demonstrate the utility of this general framework in two ways. We first illustrate how the methodology can be used to automatically tag Web pages with concepts from a known set of concepts without any need for labeled documents. We then perform a series of experiments that quantify how combining human-defined semantic knowledge with data-driven techniques leads to better language models than can be obtained with either alone.