Research paper recommender system based on public contextual metadata
- PDF / 1,366,835 Bytes
- 14 Pages / 439.37 x 666.142 pts Page_size
- 85 Downloads / 205 Views
Research paper recommender system based on public contextual metadata Khalid Haruna1 · Maizatul Akmar Ismail2 · Atika Qazi3 · Habeebah Adamu Kakudi1 · Mohammed Hassan4 · Sanah Abdullahi Muaz4 · Haruna Chiroma5 Received: 11 September 2019 © Akadémiai Kiadó, Budapest, Hungary 2020
Abstract Due to the exponential increase in research papers on a daily basis, finding and accessing related academic documents over the Internet is monotonous. One of the leading approaches was the use of recommendation systems to proactively recommend scholarly papers to individual researchers. The primary drawback to these methods, however, is that their success depends on user profile information and is therefore unable to provide useful suggestions to the new user. In addition, both the public and the non-public used descriptive metadata are used. The scope of the recommendation is therefore limited to a number of documents which are either publicly available or which are granted copyright permits. In alleviating the above problems, we proposed an alternative approach using public contextual metadata for an independent framework that customizes scholarly papers, regardless of the research field and user expertise. Experimental tests have shown significant improvements over other baseline methods. Keywords Research paper recommendation framework · Paper-citation relations · Priori user profile · Public contextual metadata
* Khalid Haruna [email protected] * Maizatul Akmar Ismail [email protected] * Haruna Chiroma [email protected] 1
Department of Computer Science, Faculty of Computer Science and Information Technology, Bayero University, Kano, Nigeria
2
Department of Information Systems, Faculty of Computer Science and Information Technology, University of Malaya, Kuala Lumpur, Malaysia
3
Centre for Lifelong Learning, Universiti Brunei Darussalam, Gadong BE1410, Brunei
4
Department of Software Engineering, Faculty of Computer Science and Information Technology, Bayero University, Kano, Nigeria
5
Future Technology Research Center, National Yunlin University of Science and Technology, Douliu, Yunlin, Taiwan
13
Vol.:(0123456789)
Scientometrics
Introduction The use of research paper recommender systems has become the leading solutions for constructive referral of relevant knowledge (Haruna and Ismail 2018; Sakib et al. 2020). The task is to properly provide relevant papers to the correct researchers (Skillen et al. 2012). Several forms of research that integrate the contextual knowledge from the academic material to personalize relevant information can be traced (Asabere et al. 2015; Beel et al. 2016; Xia et al. 2016). However, the critical concern to those approaches is that their performance depends on the richness of user profiles, which may lead to inaccurate recommendations to users with none or less details (New user problem). Furthermore, the strategies presumed the material of each of the recommending papers to be publicly available, which is not always valid due to copyright restrictions. In addres
Data Loading...
