Analysis of Titles and Readers for Title Generation Centered on the Readers

Analysis of Titles and Readers – For Title Generation Centered on the Readers Yasuko Senda†‡ and Yasusi Sinohara† †Communication & Information Research Laboratory Central Research Institute of Electric Power Industry, Japan ‡Department of Computational Intelligence and Systems Science Tokyo Institute of Technology, Japan {senda, sinohara}@criepi.denken.or.jp Abstract The title of a document has two roles, to give a compact summary and to lead the reader to read the document. Conventional title gener- ation focuses on finding key expressions from the author’s wording in the document to give a compact summary and pays little attention to the reader’s interest. To make the title play its second role properly, it is indispensable to clarify the content (“what to say”) and word- ing (“how to say”) of titles that are effective to attract the target reader’s interest. In this arti- cle, we first identify typical content and wording of titles aimed at general readers in a compar- ative study between titles of technical papers and headlines rewritten for newspapers. Next, we describe the results of a questionnaire sur- vey on the effects of the content and wording of titles on the reader’s interest. The survey of general and knowledgeable readers shows both common and different tendencies in interest. 1 Introduction The title is expected to play two roles. One is to give the reader a very compact summary of the document, and the other is to attract the target reader’s interest and lead the reader to read the document. It is preferable that a title plays both roles, because the reader may be disappointed with a gap between the title and the document if the title plays the former role poorly, and the reader may not read the document if the title plays the latter role poorly. Therefore, it is very important what title is attached to a document. Several techniques have been reported on generating titles (Jin and Hauptmann, 2000)(Berger and Mittal, 2000), and they focus on the former role, that is, to give a compact summary. The main approach is to find a few keywords from the document by calculating the importance of each word in the document. This approach, incidentally, is similar to most text summarization techniques. The selected keywords or title strongly reflect the author’s wordings. In other words, this approach is an “author-centered approach”. In some cases, the title generated by this approach might play the latter role poorly and fail to get the reader’s interest. To make generated titles play their latter role properly, it is not sufficient to look into only the author’s document. It is important to also pay more attention to the reader. It is neces- sary in the “reader-centered approach” to clar- ify the features of the reader’s attention, that is, the relationship between the reader’s atten- tion and the content and wording of the title. Based on this knowledge, it will be possible to extract information from the document that is more attractive to the reader than the author’s key expressions and to …

Content truncated to 500 words