A Collaborative Approach Toward Scientific Paper Recommendation Using Citation Context

Abstract

Researchers face difficulties in finding relevant papers to their research interest as the number of scientific publication is rapidly increasing on the web. Scientific paper recommenders have emerged as a leading solution to help researchers by automatically suggesting relevant and useful publications. Several approaches have been proposed on improving recommender systems. However, most existing approaches depend on priori user profiles, and thus they cannot recommend papers to new user. Furthermore, the existing approaches utilize non-public contextual information, and thus it cannot adequately find similarities between papers due to copyright restrictions. Also, the existing approaches consider only single level paper-citation relation to identify similarities between papers. Considering the above challenges, this paper presents a collaborative filtering based recommendation approach for scientific papers that does not depend on priori user profiles and which utilizes only public contextual information. Using citation context, we utilized 2-level paper-citation relations to find hidden associations between papers. The rational underlying this approach is that, two papers are co-occurred with same cited paper(s) and two papers are co-occurring with same citing paper(s) are significantly similar to some extent. To evaluate the performance of the proposed approach, publicly available datasets are used to conduct extensive experiments. The experimental results demonstrate that the proposed approach has significantly outperforms the baseline approaches in terms of precision, recall, F1, mean average precision, and mean reciprocal rank, which are commonly used information retrieval metrics. The novelty of this study is that, with the proposed approach, researchers are able to find relevant and useful publications over the internet regardless of their previous research experiences and research area.

A Hybrid Personalized Scientific Paper Recommendation Approach Integrating Public Contextual Metadata

“…In addition, unlike the work presented in [42], that combined papers' metadata with single level collaborative filtering, in this work, the paper's metadata was employed using a 2-level paper-citation relation with the help of collaborative filtering to find similarities between the POI and each of the candidate papers. Furthermore, unlike the works presented in [17], [33] and [35], that used only collaborative similarity, whereas in this work, both content and collaborative similarity was employed by creating a hybrid approach to make better recommendation.…”

“…To overcome the sparsity problem of conventional collaborative filtering, Dai et al [34] proposed an alternative method that combined low-rank sparse matrix with a fine-grained paper and an author affinity matrix. Sakib et al [35] proposed a collaborative scientific paper recommender framework in which 2-level papercitation relationships were mined separately using citation context to find related neighbors. Wang et al [36] proposed an alternative hybrid collaborative filtering approach by considering both the paper content and network topology.…”

“…As shown in Table 1, it is clear that the CF approach, as a successful and popular approach, was utilized in scholarly recommendation and it can provide useful recommendations. However, these approaches [15][16][17][31][32][33][34][35][36] only measure the collaborative similarities to find similar neighbors.…”

“…Sakib et al [35] proposed a Scientific Paper Recommendation (SPR) method which implemented a collaborative filtering approach to expose the secret relationships between a POI and its citations and reference papers. In order to compute similarities between the POI and each of the candidate papers, it made use of 2-level papercitation relations to mine secret associations of these two paper-citation relations.…”

Investigating Maps of Science Using Contextual Proximity of Citations Based on Deep Contextualized Word Representation

“…Keeping in view the importance of the need, understanding the semantics of relationship among research articles have been studied for more than a decade utilising different techniques. Recent research in this area is conducted by Sakib et al [33], who proposed a collaborative approach using citation context and provided a 2-level citation relationship. They proposed a recommender system that takes a paper of interest (POI) and recommends a number of candidate papers that might be relevant to the POI.…”