Finding latent needs from the consideration set using data polishing technic

2018 Fiscal Year Final Research Report

• Project/Area Number: 15K17146
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Commerce
• Research Institution: Senshu University
• Principal Investigator: Nakahara Takanobu 専修大学, 商学部, 准教授 (60553089)
• Research Collaborator: HAMURO YUKINOBU
• Project Period (FY): 2015-04-01 – 2019-03-31
• Keywords: 考慮集合 / 商品選択プロセス / データ研磨 / 類似度グラフ / 潜在ニーズ

Outline of Final Research Achievements

In this research, we focus on consumer's purchasing behavior and predict the consideration set which is product groups of candidates for purchasing products, and clarify how to treat those products as latent needs. The purpose is to construct a consumer behavior model that uses latent needs.
So far, a graph representation of purchasing behavior using product similarity graph and a method for finding potential needs using structural equivalence have been proposed, and a certain degree of prediction accuracy has been obtained. By applying graph polishing, not only direct connection relationships but also indirect connection relationships can be extracted from the connections between other products, so it is effective for capturing products that have become purchase candidates without connection relationships. The recommendation accuracy is improved by using this method.

Free Research Field

マーケティングを対象にしたデータマイニングのビジネス応用

Academic Significance and Societal Importance of the Research Achievements

情報通信技術、AIなどの急速な普及により消費者のライフスタイルは変化し、消費者ニーズの多様化と個性化はますます進んでいる。企業はさまざまな消費者ニーズに応えるべく多様な品揃えや購入手段、そして幅広い商品選択の機会を提供しているが、より効率的に消費者のニーズを捉え適切な商品・サービスを提供していくことが重要である。本研究では、小売店で蓄積されているID付きPOSデータを利用し消費者の潜在的なニーズを明らかにすることを試みており、本研究で提案した手法を利用することで消費者に適切な商品を推薦することが可能である。