應用文件分群與文字探勘技術於機器學習領域趨勢分析以SSCI資料庫為例

機器學習領域期刊文獻的研究與發表，一直是電腦科學未來應用與新科技誕生的基礎，本研究利用SSCI資料庫中與機器學習應用相關研究文獻，使用文字探勘技術，擷取具文章鑑別力之特徵詞彙，進行詞彙叢聚分析，將每份文章出現各詞彙叢聚的頻率做為自組織映射網路的輸入變數，利用網路神經元自動群集的功能，將機器學習應用的分成10大領域，最後配合文章發表年份進行趨勢分析，找出各研究領域的歷史脈絡，並進一步預測未來可能趨勢。

關鍵字

文件分群；文字探勘；自組織映射網路

並列摘要

This paper introduces the new concept for data mining manipulation. The research utilizes a document clustering technology to gain the homogeneous glossaries in each article at SSCI database, and forwarding toward onto the literature cluster assay. To select the term frequency indexes which generated by the glossaries aggregation as the parameters of the Self-Organization Map (SOM) network, proceeding the network neuron automatic clustering function, it is to strengthen the discovering ability through the historical tracking and gathering the results from various research domain, and forecasting the future possible research tendency.

並列關鍵字

Document Clustering ； Text Mining ； Self-Organization Map

參考文獻

Aas, K.,Eikvil, L.(1999).Text categorization: a survey.,::Norwegian Computing Center.

Google Scholar

Baker, L. D.,McCallum, A. K.(1998).Distributional clustering of words for text classification.Proceedings of the 21st International ACM SIGIR Conference on Research and Development in Information Retrieval.(Proceedings of the 21st International ACM SIGIR Conference on Research and Development in Information Retrieval).

Google Scholar

Bassiou, N.,Kotropoulos, C.,Pitas, J.(2001).Hierarchical word clustering for relevance judgment in information retrieval.Third International Conference on Enterprise Information Systems.(Third International Conference on Enterprise Information Systems).

Google Scholar

Bekkerman, R.,El-Yaniv, R.,Winter, Y.,Tishby, N.(2001).On feature distributional clustering for text categorization.Proceedings of the 24th International ACM SIGIR Conference on Research and Development in Information Retrieval.(Proceedings of the 24th International ACM SIGIR Conference on Research and Development in Information Retrieval).