Conference proceeding
Mining Indecisiveness in Customer Behaviors
2015 IEEE INTERNATIONAL CONFERENCE ON DATA MINING (ICDM), v 2016-, pp 281-290
01 Jan 2015
Abstract
In the retail market, the consumers' indecisiveness refers to the inability to make quick and assertive decisions when they choose among competing product options. Indeed, indecisiveness has been investigated in a number of fields, such as economics and psychology. However, these studies are usually based on the subjective customer survey data with some manually defined questions. Instead, in this paper, we provide a focused study on automatically mining indecisiveness in massive customer behaviors in online stores. Specifically, we first give a general definition to measure the observed indecisiveness in each behavior session. From these observed indecisiveness, we can learn the latent factors/reasons by a probabilistic factor-based model. These two factors are the indecisive indexes of the customers and the product bundles, respectively. Next, we demonstrate that this indecisiveness mining process could be useful in several potential applications, such as the competitive product detection and personalized product bundles recommendation. Finally, we perform extensive experiments on a large-scale behavioral logs of online customers in a distributed environment. The results reveal that our measurement of indecisiveness agrees with the common sense assessment, and the discoveries are useful in predicting customer behaviors and providing better recommendation services for both customers and online retailers.
Metrics
Details
- Title
- Mining Indecisiveness in Customer Behaviors
- Creators
- Qi Liu - University of Science and Technology of ChinaXianyu Zeng - University of Science and Technology of ChinaChuanren Liu - Drexel UniversityHengshu Zhu - BaiduEnhong Chen - University of Science and Technology of ChinaHui Xiong - Rutgers, The State University of New JerseyXing Xie - MicrosoftEric M Chen - Medicine (Graduate)
- Publication Details
- 2015 IEEE INTERNATIONAL CONFERENCE ON DATA MINING (ICDM), v 2016-, pp 281-290
- Series
- IEEE International Conference on Data Mining
- Publisher
- IEEE
- Number of pages
- 10
- Resource Type
- Conference proceeding
- Language
- English
- Academic Unit
- Medicine (Graduate)
- Web of Science ID
- WOS:000380541000029
- Scopus ID
- 2-s2.0-84963577304
- Other Identifier
- 991019174629004721
InCites Highlights
Data related to this publication, from InCites Benchmarking & Analytics tool:
- Collaboration types
- Industry collaboration
- Domestic collaboration
- International collaboration
- Web of Science research areas
- Computer Science, Artificial Intelligence
- Computer Science, Information Systems