Article Preview
Top1. Introduction
Recommender systems have been comprehensively examined and deployed expansively in various domains and applications such as electronic commerce, mobile health and collaborative learning. Numerous recommender algorithms and techniques which involve Collaborative Filtering (CF), Content-Based Filtering (CBF), Hybrid and Context-Aware are extensively discussed in literature (Bobadilla, Ortega, Hernando, & Gutiérrez, 2013). Recommender systems deliver a satisfactory method to expedite both teaching and learning tasks by detecting suitable resources from an overwhelming variety of choices in a particular domain. Recommender systems for academic environments such as mobile social learning and mobile event guides try to address the challenges of finding relevant resources and people for learning by attempting to filter contents for different learning settings (Xia, Asabere, Ahmed, Li, & Kong, 2013a; Xia, Asabere, Rodrigues, Basso, & Wang, 2013b; Asabere, Xia, Wang, Rodrigues, Basso, & Ma, 2014; Wang, Bai, Xia, Bekele, Su, & Tolba, 2017; Yu, Liu, Yang, Chen, Jiang, Tolba, & Xia, 2018).
The academic conference community has expanded globally. This expansion has introduced big scholarly data, which refers to millions of massive data in the academic and research environment (Xia, Wang, Bekele, & Liu, 2017a). The presence of big scholarly data has unfortunately paved the way for the information overload problem at academic conferences and this situation has become an evident challenge (Xia et al., 2017a). There is therefore the need for researchers and academicians to retrieve relevant and precise information concerning conference session presenters/presentations at an academic conference with a greater degree of efficacy (Xia et al., 2013b; Asabere et al., 2014; Wang et al., 2017; Yu et al., 2018).
Presently, academicians and researchers find it challenging to associate with precise people (e.g. people with similar educational goals and research interests) and find the right content (e.g. pedagogy, context and specific learning purpose) (Xia et al., 2013b; Asabere et al., 2014; Wang et al., 2017; Yu et al., 2018). Recent years have witnessed the rapid and exponential growth of social behavioral data. This growth is due to the remarkable achievements of several outlets on social websites in different forms and purposes. Furthermore, the evident growth of social websites has laid the foundation for mobile social computing/intelligence research, which has a goal of discovering, investigating and modelling how humans behave socially (Xia et al., 2013b; Asabere et al., 2014).
The high degree and massive growth of academic conferences makes it challenging for researchers to survey or find all relevant conference sessions in their specific fields (Xia et al., 2013b; Asabere et al., 2014; Wang et al., 2017; Xia et al., 2017a; Brusilovsky, Oh, López, Parra, & Jeng, 2017; Yu et al., 2018). Academic Venue Recommender Systems are therefore developed to meet the demands of attendees who are seeking relevant presentation sessions to attend in an academic conference. Conferences generate a greater sense of social consciousness, familiarization and interactions of attendees in comparison to journals (Xia et al., 2013b; Asabere et al., 2014; Wang et al., 2017; Brusilovsky et al., 2017; Yu et al., 2018). Additionally, conference sessions and workshops are time-constrained, and the attendees are usually a lot in number. A particular conference attendee’s research interest may qualify him/her to attend multiple conference sessions in an academic conference. However, it is not possible for such attendees to attend all available conference sessions that meets their research interests (Xia et al., 2013b; Asabere et al., 2014; Wang et al., 2017; Brusilovsky et al., 2017; Yu et al., 2018).