Facilitating interdisciplinary knowledge transfer with research paper recommender systems

With the advent of digital libraries, researchers can now access a plethora of publications and journals from any part of the world. These research papers play an instrumental role in acquainting researchers with the latest technological advancements taking place all over the world. Students who wish to know about the latest technologies also peruse these research papers. But currently, there is a dearth of satisfactory approaches for getting relevant recommendations. In the era of digital libraries, the importance of research paper recommendations is increasing day by day. However, there is a paucity of recommendation systems that allows us to leverage these information sources effectively. This places some limitations on an application that has great potential. So, a research paper recommendation system that allows us to generate accurate predictions is exigent. As abstracts are representative of the whole research paper, a research paper recommendation system that generates relevant keywords out of the abstract using BERT and provides recommendations would be both quick as well as accurate. BERT embeddings for keyword extraction from the abstracts and FastText for classification of the words and sentences are used for better evaluation and recommendations of the research papers. This method will fetch better results in comparison to the traditional approaches used for keyword extraction as well as research paper recommendation.

Research-related publications and articles have flooded the internet, and researchers are in the quest of getting better tools and technologies to improve the recommendation of relevant research papers. Ever since the introduction of research paper recommender systems, more than 400 research paper recommendation related articles have been so far published. These articles describe the numerous tools, methodologies, and technologies used in recommending research papers, further highlighting issues that need the attention of the research community. Few operational research paper recommender systems have been developed though. The main objective of this review paper is to summaries the state-of-the-art research paper recommender systems classification categories. Findings and concepts on data access and manipulations in the field of research paper recommendation will be highlighted, summarized, and disseminated. This chapter will be centered on reviewing articles in the field of research paper recommender systems published from the early 1990s until 2017.

With the collaborative filtering techniques becoming more and more mature, recommender systems are widely used nowadays, especially in electronic commerce and social networks. However, the utilization of recommender system in academic research itself has not received enough attention. A research paper recommender system would greatly help researchers to find the most desirable papers in their fields of endeavor. Due to the textual nature of papers, content information could be integrated into existed recommendation methods. In this paper, we proposed that by using topic model techniques to make topic analysis on research papers, we could introduce a thematic similarity measurement into a modified version of item-based recommendation approach. This novel recommendation method could considerable alleviate the cold start problem in research paper recommendation. Our experiment result shows that our approach could recommend highly relevant research papers.

Research paper recommender systems are widely used by academics to discover and explore the most relevant publications on a topic. While existing recommendation interfaces present researchers with a ranked list of publications based on a global relevance score, they fail to visualize the full range of non-textual features uniquely present in academic publications: citations, figures, charts, or images, and mathematical formulae or expressions. Especially for STEM literature, examining such non-textual features efficiently can provide utility to researchers interested in answering specialized research questions or information needs. If research paper search and recommender systems are to consider the similarity of such features as one facet of a content-based similarity assessment for academic literature, new methods for visualizing these non-textual features are needed. In this paper, we review the state-of-the-art in visualizing feature-based similarity in documents. We subsequently propose a set of user-customizable visualization approaches tailored to STEM literature and the research paper recommendation context. Results from a study with 10 expert users show that the interactive visualization interface we propose for the exploration of non-textual features in publications can effectively address specialized information retrieval tasks, which cannot be addressed by existing research paper search or recommendation interfaces.

In this chapter, the authors give an overview of the main data mining techniques that are utilized in the context of research paper recommender systems. These techniques refer to mathematical models and tools that are utilized in discovering patterns in data. Data mining is a term used to describe a collection of techniques that infer recommendation rules and build models from research paper datasets. The authors briefly describe how research paper recommender systems' data is processed, analyzed, and then, finally, interpreted using these techniques. They review different distance measures, sampling techniques, and dimensionality reduction methods employed in computing research paper recommendations. They also review the various clustering, classification, and association rule-mining methods employed to mine for hidden information. Finally, they highlight the major data mining issues that are affecting research paper recommender systems.

In recent years, a large body of literature has accumulated around the topic of research paper recommender systems. However, since most studies have focused on the variable of accuracy, they have overlooked the serendipity of recommendations, which is an important determinant of user satisfaction. Serendipity is concerned with the relevance and unexpectedness of recommendations, and so serendipitous items are considered those which positively surprise users. The purpose of this article was to examine two key research questions: firstly, whether a user’s Tweets can assist in generating more serendipitous recommendations; and secondly, whether the diversification of a list of recommended items further improves serendipity. To investigate these issues, an online experiment was conducted in the domain of computer science with 22 subjects. As an evaluation metric, we use the serendipity score (SRDP), in which the unexpectedness of recommendations is inferred by using a primitive recommendation strategy. The results indicate that a user’s Tweets do not improve serendipity, but they can reflect recent research interests and are typically heterogeneous. Contrastingly, diversification was found to lead to a greater number of serendipitous research paper recommendations.

We compare various kernel-based link analysis measures on graph nodes to evaluate their utility as a research paper recommendation system. The compared measures include the Kandola et al.'s von Neumann kernel, its extension that takes communities into account, and Smola and Kondor's regularized Laplacian. Chebotarev and Shamis' matrix forest-based algorithm, Kleinberg's HITS authority ranking, and classic co-citation coupling are also evaluated. The experimental result shows that kernel-based methods outperform HITS and co-citation coupling, with the community-based von Neumann kernel achieving the highest score.

With tremendous growth in the volume of published scholarly work, it becomes quite difficult for researchers to find appropriate documents relevant to their research topic. Many research paper recommendation approaches have been proposed and implemented which include collaborative filtering, content-based, metadata, link-based and multi-level citation network. In this research, a novel Research paper Recommendation system is proposed by integrating Multiple Features (RRMF). RRMF constructs a multi-level citation network and collaboration network of authors for feature integration. The structure and semantic based relationships are identified from the citation network whereas key authors are extracted from collaboration network for the study. For experimentation and analysis, AMiner v12 DBLP-Citation Network is used that covers 4,894,081 academic papers and 45,564,149 citation relationships. The information retrieval metrices including Mean Average Precision, Mean Reciprocal Rank and Normalized Discounted Cumulative Gain are used for evaluating the performance of proposed system. The research results of proposed approach RRMF are compared with baseline Multilevel Simultaneous Citation Network (MSCN) and Google Scholar. Consequently, comparison of RRMF showed 87% better recommendations than the traditional MSCN and Google Scholar.

In the past few years, we have developed a research paper recommender system for our reference management software Docear. In this paper, we introduce the architecture of the recommender system and four datasets. The architecture comprises of multiple components, e.g. for crawling PDFs, generating user models, and calculating content‐based recommendations. It supports researchers and developers in building their own research paper recommender systems, and is, to the best of our knowledge, the most comprehensive architecture that has been released in this field. The four datasets contain metadata of 9.4 million academic articles, including 1.8 million articles publicly available on the Web; the articles' citation network; anonymized information on 8,059 Docear users; information about the users' 52,202 mind‐maps and personal libraries; and details on the 308,146 recommendations that the recommender system delivered. The datasets are a unique source of information to enable, for instance, research on collaborative filtering, content‐based filtering, and the use of reference‐management and mind‐mapping software.

Rapidly growing scholarly data has been coined Big Scholarly Data (BSD), which includes hundreds of millions of authors, papers, citations, and other scholarly information. The effective utilization of BSD may expedite various research-related activities, which include research management, collaborator discovery, expert finding and recommender systems. Research paper recommender systems using smaller datasets have been studied with inconclusive results in the past. To facilitate research to tackle the BSD challenge, we built an analytic platform and developed a research paper recommender system. The recommender system may help researchers find research papers closely matching their interests. The system is not only capable of recommending proper papers to individuals based on his/her profile, but also able to recommend papers for a research field using the aggregated profiles of researchers in the research field.

Review for "Facilitating Interdisciplinary Knowledge Transfer with Research Paper Recommender Systems"

Decision letter for "Facilitating Interdisciplinary Knowledge Transfer with Research Paper Recommender Systems"

Review for "Facilitating Interdisciplinary Knowledge Transfer with Research Paper Recommender Systems"

Author response for "Facilitating Interdisciplinary Knowledge Transfer with Research Paper Recommender Systems"

Review for "Facilitating Interdisciplinary Knowledge Transfer with Research Paper Recommender Systems"