Comparative Analyses of Textual Contents and Styles of Five Major Japanese Newspapers

Comparative Analyses of Textual Contents and Styles of Five Major Japanese Newspapers Takafumi Suzuki Erina Kanou Faculty of Sociology, Toyo University / 5-28-20, Hakusan, Bunkyo-ku, Tokyo, Japan Faculty of Sociology, Toyo University / 5-28- 20, Hakusan, Bunkyo-ku, Tokyo, Japan takafumi_s@toyo.jp k_69en@yahoo.co.jp Yui Arakawa Graduate School of Library Information and Media Studies, University of Tsukuba / 1-2 Kasuga, Tsukuba, Ibaraki, Japan s1221577@u.tsukuba.ac.jp Abstract Newspapers remain an important media from which people obtain a wide variety of information. In Japan, there are five major newspapers, having their own opinions and ideologies. Although these are readily recognized, they are infrequently investigated from the viewpoint of their textual characteristics. This study analyzes these differences among the five newspaper editorials. We apply morphological analysis and count the frequency of morphemes within the text data. We then apply principal component analysis and random forests classification experiments to examine their similarities and differences. Throughout these statistical analyses, we use function words and content words as features, which enables us to determine which of the two characteristics -styles or content- more powerfully affects the classification types. This study contributes to text classification studies by deliberately comparing the classification performances provided by different feature sets, function words and content words. In addition, this study will provide an empirical basis for understanding the similarities and differences among the five newspapers. 1. Introduction Newspapers are an important media from which people obtain a wide variety of information, ranging from contemporary political and economic issues to ordinary incidents. Particularly in Japan, where newspapers delivery remains popular, many people read them in their own spaces and have continued to use them as popular information resources, even after the advent and spread of the Web. According to Nihon Shinbun Kyokai (2012), 87.3 percent people in Japan read newspapers, which is second only to television (98.7 percent) among the five surveyed media including newspaper, television, radio, magazines, and Internet. There are five major newspapers in Japan: Asahi, Mainichi, Nikkei, Sankei, and Yomiuri, which have publication offices in Tokyo, Osaka, and other areas, and are distributed to almost all regions of Japan. Though all of these newspapers regard correctness, neutrality and unbiased reporting as important, they have their own opinions and ideologies. According to the Hosyu (conservative)-Kakushin (liberal) image survey by Shinbun Tsushin Chosakai (2009), the five major newspapers scored as follows: Yomiuri 5.6, Sankei 5.3, Nikkei 5.2, Mainichi 5.0, and Asahi 4.4, where larger numbers indicate a newspaper perceived as more conservative and smaller numbers indicate a newspaper perceived as more liberal. Excluding Nikkei, which is a specialized newspaper that focuses on economic issues, the survey results show that people see Yomiuri and Sankei as more conservative and Mainichi and Asahi as more liberal. These differences might affect the textual characteristics of newspapers; however, they have not been investigated in a comprehensive and systematic manner. With the development of natural language processing techniques and the creation of many online text corpora, quantitative text analysis has been expanding in scope. Such methods have begun to be …

Content truncated to 500 words