本研究利用R語言的文本探勘技術，分析《張秀亞全集》中收錄的三部西方天主教文學作品：《恨與愛》、《回憶錄》、《聖女之歌》，探查此三部譯品中的共現現象、張秀亞譯文的文字特徵，並藉此建立數位人文式的翻譯文學研究模式，歸結出此研究方法的可能性與局限性。利用R語言的斷詞分析與詞頻統計功能，可證驗三譯品的故事結構皆圍繞著高頻詞「天主」進行延伸，符合天主教文學的共現性，而各作品所抽取出的高頻詞也可再利用關係參數找出特色特徵。其次，從譯本語料庫可看到張秀亞運用獨特詞彙進行翻譯的特性，辭彙數量龐大且不流俗，與其散文或詩詞創作的美文輝映相襯，使這三譯作成為具張氏風格的譯品。由數位人文方法進行的天主教文學翻譯研究，優點在於可析離出譯本中的重要語彙、關鍵詞，我們可以從結果中得到一些想法，判斷譯者如何使用語言來呈現原作的風格。

This research uses R language text-mining to analyze three western Catholic literary works selected from Chang Hsiu-Ya’s Collected Works, which includes Vipers’ Tangle, Autobiography of St. Thérèse of Lisieux, and The Song of Bernadette. The purpose of this study was to investigate the commonality of the catholic literature and to outline Chang’s translation style with an experimental digital humanistic method. A secondary purpose of this study was to establish a digital humanistic translation research model to better determine the possibilities and limitations. The findings indicate that “God” is the core idea and the core keyword of the selected works, and text-mining method should be used to explore the similarities in different types of catholic literature. Even though this body of research has the undeniable merit of offering valuable insights into text mining, it has some limitations. We acknowledged that the research is exploratory and there are problems with the interpretation and practice mode. Besides, we all know that Chang was used to using classy and lyric words as her writing policy, the results confirm that Chang’s writing style made the translation beautiful and readable; however, the styles of the translated works might look all the same.