Selecting Proper Lexical Paraphrase for Children

Selecting Proper Lexical Paraphrase for Children Tomoyuki Kajiwara Hiroshi Matsumoto Kazuhide Yamamoto Nagaoka University of Technology {kajiwara, matsumoto, yamamoto}@jnlp.org Abstract We propose a method for acquiring plain lexical paraphrase using a Japanese dictionary in order to achieve lexical simplification for children. The proposed method extracts plain words that are the most similar to the headword from the dictionary definition. The definition statements describe the headword using plain words; therefore, paraphrasing by replacing the headword with the most similar word in the dictionary definition is expected to be an accurate means of lexical simplification. However, it is difficult to determine which word is the most appropriate for the paraphrase. The method proposed in this paper measures the similarity of each word in the definition statements against the headword and selects the one with the closest semantic match for the paraphrase. This method compares favorably with the method that acquires the target word from the end of the definition statements. Keywords: Lexical Simplification, Lexical Paraphrase. 1. Introduction In the current information age, a various readers have easy access to diverse text data. To achieve information transmission and gathering effectively, we must address the gap in readers’ linguistic skills. The gap of linguistic skills results from differences in age, such as between children and adults, as well as from differences in expert knowledge. In the effort to bridge this gap, and also to facilitate better communication with foreign language speakers[8] and people with disabilities, technology can play an important role. To investigate how technology can be applied toward bridging the gap in readers’ linguistic skills, we simplify the text of newspaper articles containing words that pose difficulties in communication, especially for elementary school students. Children are still developing their language skills, and as such, they have smaller vocabularies than adults. In this paper, we perform text simplification for children by paraphrasing selected newspaper articles using only words found in Basic Vocabulary to Learn (BVL)(3). BVL is a collection of words selected based on a lexical analysis of elementary school textbooks. It contains 5,404 words that can help children write expressively. We define words not included in BVL as Difficult Words (DWs) and those in BVL paraphrased from DW as Simple Words (SWs). Paraphrasing newspaper articles using words that children can understand makes a great contribution to reading assistance for young students. Proceedings of the Twenty-Fifth Conference on Computational Linguistics and Speech Processing (ROCLING 2013) 59 2. Related Works Although there are some methods [10] proposed for automatically acquiring paraphrasable expressions from Web pages, the quality of the results are still unsatisfactory. Hence typical methods use thesauri or dictionaries. Thesaurus is a language resource that contains semantically classified vocabulary words. Methods that utilize a thesaurus have an advantage in that they can measure the semantic relatedness between words (i.e., the distance between meanings). Japanese dictionaries are another language resource that provides the definition of a given lemma. Methods that utilize a dictionary have an advantage in that they are able to acquire simplified text. …

Content truncated to 500 words