社群媒體文字探勘應用於郵輪旅客重遊意願之研究

郵輪旅客重遊意願資訊之取得為郵輪公司極為重要的課題。本研究透過社群媒體資料蒐集旅客體驗後之留言，並以文字探勘技術辨識關鍵字詞與分類，計算郵輪旅客重遊意願的可能性。另外，本研究以情緒評分方法將各貼文字詞評分量化，透過迴歸分析探討貼文字詞對旅客重遊意願是否顯著。研究結果顯示，貼文內容中包含目的地、時間、體驗與感官四項分類之字詞，對於旅客重遊意願具有正向顯著的影響，而關聯分析結果可發現旅客貼文字詞的關聯規則，依據不同貼文字詞分類組合，計算出貼文字詞分類組合之重遊意願可能性。

關鍵字

郵輪；社群媒體；文字探勘；重遊意願

並列摘要

The cruise traveler's revisit intention information is an important issue for cruise companies. This study uses the social media to collect the message after the passenger experience, and the text mining to identify key words and classification, calculate the possibility of cruise passengers wish to revisit. In addition, this study uses the sentimental analysis to quantify the scores of each post, and analyzes whether the intention of the post on the visitors is significant by regression analysis. The results show that the content of the post contains the words of destination, time, experience and sensory has a positive effect on the revisit intention of the tourists, and the association analysis results can find the association rules of the words which traveler uses, and according to the different posts word classification combination, calculate the posting word classification combination of the possibility of the revisit intention.