CONCERT: A Concept-Centric Web News Recommendation System

A concept is a key phrase which can represent an entity, event or idea that people are interested in. Concept-centric Web news recommendation is a novel content-based recommendation paradigm which can partially alleviate the cold-start problem and provide better recommendation results in terms of di...

Full description

Saved in:
Bibliographic Details
Published inWeb-Age Information Management pp. 796 - 798
Main Authors Ren, Hongda, Feng, Wei
Format Book Chapter
LanguageEnglish
Published Berlin, Heidelberg Springer Berlin Heidelberg
SeriesLecture Notes in Computer Science
Subjects
Concept Extractor
Cosine Distance
Inverted List
Main Concept
News Text
Online AccessGet full text

Cover

More Information
Summary:A concept is a key phrase which can represent an entity, event or idea that people are interested in. Concept-centric Web news recommendation is a novel content-based recommendation paradigm which can partially alleviate the cold-start problem and provide better recommendation results in terms of diversity than traditional news recommendation systems, as it can capture users’ interest in a natural way and can even recommend a new Web news to a user as long as it is conceptually relevant to a main concept of the Web news the user is browsing. This demonstration paper presents a novel CONcept-Centric nEws Recommendation sysTem called CONCERT. CONCERT consists of two parts: (1) A concept extractor which is based on machine learning algorithms and can extract main concepts from Web news pages, (2) A real-time recommender which recommends conceptually relevant Web news to a user based on the extracted concepts.
Bibliography:The demonstration system of the CONCERT system is available online at: http://dmgroup.cs.tsinghua.edu.cn/CONCERT/main.html
ISBN:9783642385612
3642385613
ISSN:0302-9743
1611-3349
DOI:10.1007/978-3-642-38562-9_82