UPM Institutional Repository

HYPNER a framework for hybrid personalized news recommender systems


Citation

Darvishy, Asghar and Ibrahim, Hamidah and Sidi, Fatimah and Mustapha, Aida (2014) HYPNER a framework for hybrid personalized news recommender systems. In: Malaysian National Conference of Databases 2014 (MaNCoD 2014), 17 Sept. 2014, Universiti Putra Malaysia, Serdang, Selangor. (pp. 33-39). (Unpublished)

Abstract

The new medium for press release is the online news article publishing over the Internet. Being free from traditional printing limitation, the number of news article made online grow exponentially as new articles are released every subsequent hour. The current problem in the existing news recommender system including the hybrid recommenders is that accuracy is still considerable low. In this paper, we propose HYbrid Personalized NEws Recommender (HYPNER), which is a framework based on ordered clustering and set theory that is able to select news set and generate recommendations that fit the user profiles. We describe how does the framework learn user behavior and interest to predict recommended news set and then discuss it's differences with previous works.


Download File

[img] PDF
39292.pdf
Restricted to Repository staff only

Download (509kB)
Official URL or Download Paper: http://mancod2014.blogspot.my/p/proceedings.html

Additional Metadata

Item Type: Conference or Workshop Item (Paper)
Divisions: Faculty of Computer Science and Information Technology
Keywords: News recommendation system; Personalization; Reader preferences; News entity; User profiling
Depositing User: Nursyafinaz Mohd Noh
Date Deposited: 08 Jul 2015 07:49
Last Modified: 29 Jul 2016 07:58
URI: http://psasir.upm.edu.my/id/eprint/39292
Statistic Details: View Download Statistic

Actions (login required)

View Item View Item