A Study on the Impact of Sequence Setting in Text Mining on Character Relationship Analysis: Through an Analysis of the Children’s Translation of Jane Eyre

The purpose of this study is to analyze the impact of sequence setting, defined as the context window size, on the character relationship network in the children’s translation of Jane Eyre. The results show that, unlike the English original, when the sequence was set to 50, 40, and 30 words, the character forming the most relationships in the children’s translation was not “Jane” but “Rochester.” In contrast, in the English original, “Jane” remained the central character forming the most relationships regardless of sequence size. Furthermore, in terms of co-occurrence frequency, the English original consistently showed the highest value for the pair “Diana Rivers–Mary Rivers,” while the children’s translation displayed diverse patterns such as “Rochester–Grace Poole,” “Rochester–Blanche Ingram,” “Jane–Rochester,” and “Eliza–Mrs. Reed.” These differences suggest that the children’s translation employed strategies aimed at easing complex psychological conflicts, emphasizing emotional stability and familiarity, and focusing on moral or educational messages. The variations observed in sequence settings thus reflect translation strategies designed to facilitate children’s comprehension.