-
- TSUBOI Kazuaki
- Graduate School of Informatics Systems, The University of Electro-Communications
-
- KOSUKE Shinoda
- Graduate School of Informatics Systems, The University of Electro-Communications
-
- SUWA Hirohiko
- Graduate School of Information Science, Nara Institute of Science and Technology
-
- KURIHARA Satoshi
- Graduate School of Informatics Systems, The University of Electro-Communications
Bibliographic Information
- Other Title
-
- ACO型時系列パターン抽出法を用いたマーケティングデータの解析
- ACOガタ ジケイレツ パターン チュウシュツホウ オ モチイタ マーケティングデータ ノ カイセキ
Search this article
Abstract
<p>It is important to understand various consumers' needs and clarify the target of goods and service in marketing. As information processing technology develops, as more various data of consumer's purchase action could be collected and accumulated. And the technology which extract the consumer's real intention, we call consumer insight, from the collected and accumulated data is noted. In this study, we are developing an ACO-based pattern mining method which extracts frequent purchase pattern from the information of receipt data as a purchase result of consumers'. We consider that consumer behavior is ambiguous and always changes because of in uence of the fashion and season. Then we try to use Ant Colony Optimization (ACO) algorithm which is one of collective intelligence-based approach and the optimization method which modeled the behavior of ant seeking for food in nature. ACO algorithm is famous as a solution of a traveling salesman problem having high robustness and adaptability. Like ACO algorithm, we extract frequent pattern by adding and evaporating the pheromone to the virtual graph. And we analyze the marketing data for one year in the two store in Ebetsu city, Hokkaido by using ACO-based pattern mining approach. Two kinds of experiment is made about each of the store. One is to input data to 31 December 2013, the other is to input data to 30 June 2014. Each result of experiments can be visualized by Cytoscape. the data was offered by Joint Association Study Group of Management Science.</p>
Journal
-
- JSAI Technical Report, SIG-KBS
-
JSAI Technical Report, SIG-KBS 104 (0), 04-, 2015-02-26
The Japanese Society for Artificial Intelligence
- Tweet
Details 詳細情報について
-
- CRID
- 1390570166646533248
-
- NII Article ID
- 40020371874
- 130008064780
-
- NII Book ID
- AN10231834
-
- ISSN
- 24364592
-
- NDL BIB ID
- 026161921
-
- Text Lang
- ja
-
- Data Source
-
- JaLC
- NDL
- CiNii Articles
-
- Abstract License Flag
- Allowed