Topic Identification in Discourse

Topic Identification in Discourse Kuang-hua Chen Department of Computer Science and Information Engineering National Taiwan University Taipei, Taiwan, R.O.C. khchen@nlg.csie.ntu.edu.tw Abstract This paper proposes a corpus-based language model for topic identification. We analyze the association of noun-noun and noun-verb pairs in LOB Corpus. The word association norms are based on three factors: 1) word importance, 2) pair co-occurrence, and 3) distance. They are trained on the paragraph and sentence levels for noun-noun and noun- verb pairs, respectively. Under the topic coherence postulation, the nouns that have the strongest connectivities with the other nouns and verbs in the discourse form the preferred topic set. The collocational semantics then is used to identify the topics from paragraphs and to discuss the topic shift phenomenon among paragraphs. 1 Introduction Although only speakers and writers instead of texts have topics (Brown and Yule, 1983, p. 68), natural language researchers always want to identify a topic or a set of possible topics from a discourse for further applications, such as anaphora resolution, information retrieval and so on. This paper adopts a corpus-based approach to process discourse information. We postulate that: (1) Topic is coherent and has strong relation- ships with the events in the discourse. Now, consider the following example quoted from the Lancaster-Oslo/Bergen (LOB) Corpus (Johansson, 1986). The topics in this example are "problem" and "dislocation". The two words are more strongly related to the verbs ("explain", "fell", "placing" and "suppose") and nouns ("theories", "explanations", "roll", "codex", "disorder", "order", "disturbance" and "upheaval"). There is a whole group of theories which attempt to explain the problems of the Fourth Gospel by explanations based on assumed textual dislocations. The present state of the Gospel is the result of an accident-prone history. The original was written on a roll, or codex, which fell into disorder or was accidentally damaged. An editor, who was not the author, made what he could of the chaos by placing the fragments, or sheets, or pages, in order. Most of those who expound a theory of textual dislocation take it for granted that the Gospel was written entirely by one author before the disturbance took place but a few leave it open to suppose that the original book had been revised even before the upheaval. We also postulate that (2) Noun-verb is a predicate-argument relation- ship on the sentence level and noun-noun relationship is associated on discourse level. The postulation (2) could be also observed from the above example. These relationships may be represented implicitly by collocational semantics. Collocation has been applied successfully to many possible applications (Church et al. , 1989), e.g, lexicography (Church and Hanks, 1990), information retrieval (Salton, 1986a), text input (Yamashina and Obashi, 1988), etc. This paper will touch on its feasibility in topic identification. This paper is organized as follows. Section 2 presents a corpus-based language model and discuss how to train this model. Section 3 touches on topic identification in discourse. Section 4 shows a series of experiments based on the proposed model and discusses the …

Content truncated to 500 words