Influence of Target Reader Background and Text Features on Text Readability in Bangla: A Computational Approach

Proceedings of COLING 2014, the 25th International Conference on Computational Linguistics: Technical Papers, pages 345–354, Dublin, Ireland, August 23-29 2014. Influence of Target Reader Background and Text Features on Text Readability in Bangla: A Computational Approach Manjira Sinha Department of Computer Science and Engineering Indian Institute of Technology Kharagpur West Bengal, India manjira@cse.iitkgp.e rnet.in Tirthankar Dasgupta Department of Computer Science and Engineering Indian Institute of Technology Kharagpur West Bengal, India tirtha@cse.iitkgp.e rnet.in Anupam Basu Department of Computer Science and Engineering Indian Institute of Technology Kharagpur West Bengal, India anupam@cse.iitkgp.e rnet.in Abstract In this paper, we have studied the effect of two important factors influencing text readability in Bangla: the target reader and text properties. Accordingly, at first we have built a novel Bangla readability dataset of 135 documents annotated by 50 readers from two different backgrounds. We have identified 20 different features that can affect the readability of Bangla texts; the features were divided in two groups, namely, „classic‟ and „non-classic‟. Preliminary correlation analysis reveals that text features have varying influence on the text hardness stated by the two groups. We have employed support vector machine (SVM) and support vector regression (SVR) techniques to model the reading difficulties of Bangla texts. In addition to developing different models targeted towards different type of readers, separate combinations of features were tested to evaluate their comparative contributions. Our study establishes that the perception of text difficulty varies largely with the background of the reader. To the best of our knowledge, no such work on text readability has been recorded earlier in Bangla. 1 Introduction Readability of a text generally refers to how well a reader is able to comprehend the content of a text, through reading (Dale and Chall, 1948). Readability is a complex cognitive phenomenon where, the cognitive load of a text for a reader depends on both the characteristics of a text like, lexical choice, syntactic complexity, semantic complexity, discourse level complexity and on the background of the user. Several experiments have already established that readability of texts are quite language dependent and existing readability measures in English cannot directly be used to compute readability of other languages like, Bangla and Hindi (Sinha et al., 2012). Yet, compared to the numerous readability measures in English and other European languages(Benjamin, 2012), few initiatives have been taken to compute text readability in a Eastern Indo-Aryan language like Bangla or any other Indian languages which are structurally very different from many of their Indo-European cousins such as English, which is of West-Germanic descent (Sinha et al., 2012). One important factor that affects the readability of a text is the background of the respective reader. According to Dale (Dale, 1949), “The interpretation of the expressed thought is related more to the reader‟s informational background and motivations than to the internal evidences of the expressional facility of the author”. Reader‟s background is a complex derivative of one‟s educational and socio-economic state. As per one of the pioneering works in readability by Dale and Chall (1949), the …

Content truncated to 500 words