Dialogic reading, which involves interactive exchanges between a parent and a child during picture book reading, has been shown to effectively promote children’s language development. While many support systems for picture book reading have been developed to reduce the burden on parents, existing systems are not yet capable of handling dialogic reading, which requires dynamic parent-child interaction. To develop conversational agents capable of dialogic reading, we constructed a multimodal corpus of parent-child picture-book reading dialogues. The corpus comprises recordings from 36 Japanese parent-child pairs taken during actual picture book reading sessions. In this study, we annotated the corpus with dialogue acts relevant to parent-child communication and categorized the types of quizzes and questions used in the sessions, analyzing the linguistic aspects of parent-child interaction during dialogic reading. After dividing the dialogues into two groups based on the proportion of the child’s utterances, our analyses revealed that dialogue systems should adapt their interaction strategies according to individual child characteristics.

Information on subcategorization and selectional restrictions is important for natural language processing tasks such as deep parsing, rule-based machine translation and automatic summarization. In this paper we present a method of adding detailed entries to a bilingual dictionary, based on information in an existing valency dictionary. The method is based on two assumptions: words with similar meaning have similar subcategorization frames and selectional restrictions; and words with the same translations have similar meanings. Based on these assumptions, new valency entries are constructed from words in a plain bilingual dictionary, using entries with similar source-language meaning and the same target-language translations. We evaluate the effects of various measures of similarity in increasing accuracy.