Trends in Text Mining for Wordless Picture Books : A Focus on Keyword Network Analysis and Topic Modeling

This study explores research trends in wordless picture books using text mining techniques. For this purpose, abstracts of 181 articles were collected, and then data preprocessing and analysis were performed. In the data analysis process, keyword network analysis and topic modeling analysis were performed. The keyword analysis showed that the most frequent word was ''young children'', followed by ''storytelling'' and ''wordless picture book activities''. The network analysis showed that the strength of the connections between ''wordless picture book activities → young children'' and ''using wordless picture books → young children'' was high. This suggests that in studies of wordless picture books, the use of picture book activities or wordless picture books is mainly targeted at young children. Topic modeling resulted in four themes: multicultural approach and visual literacy, playfulness and artistic development, integration of language and art, and methodology of integration and repetition. This study is significant in that it identifies major research themes in the field of wordless picture books and suggests future research directions.