This volume contains the papers accepted for presentation at the 37th Annual Meeting of the Association for Computational Linguistics, held at the University of Maryland from 20th to 26th June 1999.

As co-chairs, we had two particular objectives for the program this year. First, we wanted to acknowledge the increasing worldwide interest in computational linguistics, and mirror the move towards true internationalism that is underway within the ACL itself. Second, we wanted to reflect the broader connections that computational linguistics has with other research communities by increasing the scope of the material covered. You will see the results of our pursual of these objectives throughout the program and its organisation: the 80 papers collected here for presentation were drawn from over 320 received, with the authors representing 29 countries and the reviewers drawn from 25 countries.

The larger-than-usual size of the conference this year is at least in part due to our incorporation of community-proposed theme sessions. In an iteration prior to the call for papers itself, we issued a call for theme session proposals. This was our attempt to break with the somewhat traditional set of paper topic categories that one tends to think of at ACL conferences. As a way of highlighting real areas of topical interest in the community, we think the experiment worked extremely well, despite the inevitable hiccups that accompany the introduction of any new idea. But you can be the judge of this: the ACL executive committee and next year's program chairs will be interested to hear what you think of this innovation.

Proceedings of the 37th annual meeting of the Association for Computational Linguistics on Computational Linguistics

A corpus (eg, patents or news texts) is an important knowledge resource that contains various topics, such as specific technologies or social events. Topic detection models of corpus, eg, Latent Dirichlet Allocation and KeyGraph, provide an important basis for exploring the status quo and trends in science, technology, or social events. However, these models suffer from low retrieval performance as they only consider text own explicit semantics in a single‐domain corpus. In addition, many incremental models, such as online‐LDA, are based on time slices. In this paper, a new topic detection model is proposed to improve the topic detection performance of a single‐domain corpus, which is inspired by a human memory cognitive process (THC). First, to improve the accuracy, distributions over words and inter‐word relations across a corpus are utilized as background knowledge, which is a type of implicit semantics, and we can find a more semantic‐sensitive part of texts. Second, to realize online topic detection without time slices, we introduce a probability gain‐based dynamic probabilistic model to detect latent topics by learning a model based on the dynamic human memory cognitive process. These two steps constitute the framework of our model. The experimental results for four public datasets (Reuters‐R8, Reuters‐R52, WebKB, and Cade12) reveal that our model is approximately ten percent higher than other baselines (eg, KeyGraph and LDA) on the Adjusted Rand Index (ARI).

Topic detection model in a single‐domain corpus inspired by the human memory cognitive process

Automatic text summarization is an important and useful research area in natural language processing and information retrieval. Most of current approaches for text summarization do not make full use of human reading process. This paper proposes a multi-document scanning mechanism by simulating human reading process. The mechanism simulates human memory of words, association between words and three cognitive processes invoked when reading. Changes of human memory of topic words in reading process are used to denote sentences' significance, based on which sentences are then ordered and extracted to form a summary. Experiments on DUC2007 test data show that our proposing method is efficient and outperforms two baseline methods.

http://iip.ict.ac.cn/sites/default/files/publication/Cognitive-based%20Multi-Document%20Summarization%20Approach.pdf

Wei Li

Papers

Cognitive-Based Multi-document Summarization Approach