A Survey of Researches on Personalized Bundle Recommendation Techniques
| Autor: | Long Li, Xuguang Bao, Liang Chang, Miao Li, Zhoubo Xu |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Matching (statistics)
Information retrieval Computer science business.industry Deep learning 02 engineering and technology Recommender system Preference User experience design 020204 information systems Bundle 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Relevance (information retrieval) Artificial intelligence business |
| Zdroj: | Machine Learning for Cyber Security ISBN: 9783030624590 ML4CS (2) |
| DOI: | 10.1007/978-3-030-62460-6_26 |
| Popis: | The recommender system is widely used in various fields such as movies, music, and products. It has been an effective method to handle the preference matching problem by retrieving the most relevant information and services from a large amount of data. Most researches on recommender systems focus on improving the relevance of individual recommendation item. Nevertheless, in recommendation scenarios, users are often exposed to multiple items and may be interested in a collection of items. The platform can offer multiple items as a bundle to user in this case. That is called Bundle Recommendation problem, which refers to predict a user’s preference on a bundle rather than an individual item. The larger scale data of the bundle compared to individual item results in information overload, thus, it is crucial to develop personalized bundle recommendation technology to help users quickly match a high-quality bundle which can improve the user experience and benefit both consumers and product providers. In this paper, we first review the main model methods of bundle recommender systems in recent years, including integer programming methods, association analysis methods, traditional recommendation methods, and recommendation methods based on deep learning. And then, we analyze the relationships of personalized bundle recommender systems compared with the traditional recommender systems. Finally, we discuss the difficulties and future development trend of personalized bundle recommender systems. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|