Transitivity and Foregrounding in News Articles: Experiments in Information Retrieval and Automatic Summarising

Transitivity and Foregrounding in News Articles: experiments in information retrieval and automatic summarising Roderick Kay and Ruth Aylett Information Technology Institute University of Salford Manchester M5 4WT United Kingdom {rnk,R.Aylett}@iti.salford.ac.uk Abstract This paper describes an on-going study which applies the concept of transitivity to news discourse for text processing tasks. The complex notion of transitivity is de- fined and the relationship between transi- tivity and information foregrounding is ex- plained. A sample corpus of news articles has been coded for transitivity. The corpus is being used in two text processing exper- iments. 1 Introduction 2 Definition of Transitivity Transitivity is usually considered to be a property of an entire clause (Hopper and Thompson, 1980). It is, broadly, the notion that an activity is transferred from an agent to a patient. It is therefore inherently linked with a clause containing two participants in which an action is highly effective. The concept of transitivity has been defined in terms of the following parameters: A. Participants h 1 B. Kinesis h 1 C. Aspect h 1 D. Punctuality h 1 E. Volitionality h 1 F. Affirmation h 1 G. Mode h 1 H. Agency h 1 I. Affectedness of O h 1 J. Individuation of O h 1 h=high, l=low Each component of transitivity contributes to the overall effectiveness or 'intensity' with which an ac- tion is transferred from one participant to another. A. There must be at least two participants for an action to be transferred. B. Transferable actions can be contrasted with non- transferable states, e.g. he pushed her; he thought about her. C. An action is either wholly or partially completed according to whether it is telic or atelic, e.g. I played the piano; I am playing the piano. D. Punctual actions have no transitional phase be- tween start and end point, having a greater effect The basic hypothesis of this study is that the degree of transitivity associated with a clause indicates the level of importance of a clause in a narrative text. For this assumption to form the basis of a practical implementation, transitivity must be objectively de- fined and the definition must be able to be processed automatically. The notion of transitivity clearly has many impli- cations for text processing, in particular information retrieval and automatic summarising, because it can be used to grade information in a document accord- ing to importance. In an information retrieval con- text, it means that a transitivity index could influ- ence a decision about the relevance of a document to a query. In automatic summarising, it means that less important information could be sieved accord- ing to transitivity, leaving only the most important information to form the basis of a summary. News discourse was chosen because it is narra- tive based and therefore broadly applicable to the notion of transitivity. There has also been exten- sive research in the structural characteristics of this text type (Duszak, 1995) (Kay and Aylett, 1994) (Bell, 1991). However, the study poses a challenge in the …

Content truncated to 500 words