KDD 2019 (25th ACM SIGKDD Conference on Knowledge Discovery and Data Mining) was held from 2019/08/04 to 2019/08/08 at Anchorage, Alaska, USA. You may reference more information about this conference at KDD 2019 homepage.
KDD 2019 has many recommender system (RS) articles focus on Ratings and Reviews Information, both of which received more importance in developing recommender system, to get better Stability and dynamic performance. There are still some articles discussing the robustness of RS.
More RS are starting to use deep models, such as Deep Q-learning or DNN of the class to get better accuracy that increases over time. There are also articles focus on cold starts, but not so many.
Most of the RS articles focus on the problem of recommendations in a specific context, such as road or shopping recommendations, and consider a variety of realistic scenarios, with the Rating scenario being considered more often, and the gap between the Top Rating and List-wise Rating approaches being discussed in a number of related articles, as shown in my reviewing below.
All in all, the discussion is relatively focused on RS only, and there is some commonality in the issues addressed. You may read the article or browse the Review below for more information.
The Review PowerPoint follows the structure:
- Each article will have a front page to describe the information in the article so that the reader can easily find the corresponding original text.
- Immediately following the article's information description page, there will be a paragraph summarizing the Abstract to provide a general idea of what the article is working on.
- Presentation of the methodological background of the article.
- Introduce the specific model structure of the article methodology.
- To present the results of the tests and experiments carried out by the article to understand the performance of the model proposed in the article.
Publications folder contains most available publications PDF document, which would save your time for finding such papers. However, some papers are not here since the limitation of copyright and download limitations, you may read them online.
Here provides the full original PowerPoint document and generated PDF document. Feel free to use and edit them, happy researching.
If there are questions or problems, please report an issue.