I have categorized features for book recommender system. these features are presented to users and asked to state whether these features have any significance for them in purchasing books. can you assist me in selection criteria of features?
How much percentage of users should agree on a feature to retain it in recommendation procedure?
There are plenty of feature selection methods available in the literature. I propose to read the some of the following survey articles and decide what is most suitable for your case.
- Chandrashekar, Girish, and Ferat Sahin. "A survey on feature selection methods." Computers & Electrical Engineering 40, no. 1 (2014): 16-28. http://romisatriawahono.net/lecture/rm/survey/machine%20learning/Chandrashekar%20-%20Feature%20Selection%20Methods%20-%202014.pdf
- Wang, Jialei, Peilin Zhao, Steven CH Hoi, and Rong Jin. "Online feature selection and its applications." IEEE Transactions on Knowledge and Data Engineering 26, no. 3 (2014): 698-710. http://ink.library.smu.edu.sg/cgi/viewcontent.cgi?article=3277&context=sis_research
- Tang, Jiliang, Salem Alelyani, and Huan Liu. "Feature selection for classification: A review." Data Classification: Algorithms and Applications (2014): 37. http://eprints.kku.edu.sa/170/1/feature_selection_for_classification.pdf
Dear Shahab,
Book is a rich source of knowledge. It cannot be treated as an ordinary product. Feature selection process for books will have great impact on the recommendation results. It is therefore important to choose such features that could represent a book properly. A book can be represented by a set of features including Book title, author, ISBN, Book table of contents, Publishing date, associated metadata and tags. As recommender systems have been categorized into three broad categories, feature selection process vary from type to type. In CB, vector space model and classifiers can be used to represent books, whereas in CF, Book-rating matrix and historical ratings are used to find nearest neighbors and recommend books.
In order to find more information about book recommendation, please have a look of our research paper on Book recommendaiton.
"A Hybrid Book Recommender System Based on Table of Contents (ToC) and Association Rule Mining".
Regards
Zafar Ali
Dear Zafar
Assalam
Hope it finds you well, I have gone throw your article, it is looking interesting. I also remember once you have asked about my work for citing it into your article, due to some unavoidable circumstances I could not reply then. I was searching my name in your reference, (oop, missing). you must have find "OWA BASED BOOK RECOMMENDATION TECHNIQUE".
however, I will let you know the problem I am facing in deciding the selection criteria of features, Regards and Salam.
Dear Shahab Sohail,
Walikum Salam, thanks for appreciating our work. At that time, I was collecting data for my paper. Due to your busy schedule, you were not able to respond me back. By the way, I would like to ask some questions about your research work as I am working on a new topic where I would need to gather data from different research articles in recommendation domain.
Regards
Zafar Ali
- Badges
- Science topic
- Similar topics
- Books