機械学習を活用したテキストマイニング(2) : 仮説の発見と検証

機関リポジトリ Web Site オープンアクセス

書誌事項

タイトル別名
  • Text Mining Using Machine Learning(2) : Discovery and Verification of Hypotheses
  • キカイ ガクシュウ オ カツヨウ シタ テキストマイニング(2)カセツ ノ ハッケン ト ケンショウ

この論文をさがす

説明

In recent years, AI has gained social attention as there are reports of leap in image recognition contest ILSVRC, the appearance of AlphaGo, and applied cases in various industrial fields. Words related to AI such as machine learning and deep learning has become widely recognized. As an application possibility to management research and management practice, it can be said that it is becoming a sufficiently usable technology due to improvement of natural language processing technology in recent years, especially improvement of distributed representation technology. In this research, we propose a method to analyze and visualize the characteristics of goods and services with text mining which using machine learning technology included in AI technologies, especially the method on discovery and verification of hypotheses. The text mining method proposed in this research is not based on current weighing text analysis which aggregate values of words, but based on distributed representation of words calculated by machine learning. By using this method, it is possible to analyze based on the consumer’s experience and meaning made up of the consuming process. The method proposed by this paper is also useful for business researchers, but even for practitioners, when planning and model change of new products or services, comparison with other company’s product services becomes easier than ever, more detailed analysis It is thought that it is beneficial because it can be considered to be able to proceed with practical work based on it. Text mining based on distributed representation is still an incomplete technology, not a fixed method, but it is an area where further development is expected in the future.

収録刊行物

関連プロジェクト

もっと見る

キーワード

詳細情報 詳細情報について

問題の指摘

ページトップへ