Research Paper Recommender System Research Articles

Research paper recommendation system based on multiple features from citation network

AbstractWith 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.

AI 기반 맞춤형 논문 추천 시스템 개발 및 적용 사례 연구: K 대학을 중심으로

AI 기반 맞춤형 논문 추천 시스템 개발 및 적용 사례 연구: K 대학을 중심으로

High-Performance Artificial Intelligence Recommendation of Quality Research Papers Using Effective Collaborative Approach

The Artificial Intelligence Recommender System has emerged as a significant research interest. It aims at helping users find things online by offering recommendations that closely fit their interests. Recommenders for research papers have appeared over the last decade to make it easier to find publications associated with the field of researchers’ interests. However, due to several issues, such as copyright constraints, these methodologies assume that the recommended articles’ contents are entirely openly accessible, which is not necessarily the case. This work demonstrates an efficient model, known as RPRSCA: Research Paper Recommendation System Using Effective Collaborative Approach, to address these uncertain systems for the recommendation of quality research papers. We make use of contextual metadata that are publicly available to gather hidden relationships between research papers in order to personalize recommendations by exploiting the advantages of collaborative filtering. The proposed system, RPRSCA, is unique and gives personalized recommendations irrespective of the research subject. Thus, a novel collaborative approach is proposed that provides better performance. Using a publicly available dataset, we found that our proposed method outperformed previous uncertain methods in terms of overall performance and the capacity to return relevant, valuable, and quality publications at the top of the recommendation list. Furthermore, our proposed strategy includes personalized suggestions and customer expertise, in addition to addressing multi-disciplinary concerns.

An anatomization of research paper recommender system: Overview, approaches and challenges

The purpose of this study is to present an exhaustive analysis on research paper recommender systems which have become very popular and gained a lot of research attention. Though the major focus is on developing new recommendation algorithms, other research dimensions are left untouched. Renown recommendation classes include content-based approaches, collaborative filtering, link-based algorithms, co-occurrence based approaches, global relevance and hybrid methods. These approaches mainly differ in background knowledge and modes of user behavior analysis. For instance, content-based filtering uses paper descriptions which are mostly word-based features. Collaborative filtering makes predictions based on peers’ interests. Link-based algorithms utilize academic associations that exist between different entities in academia. Co-occurrence based techniques incorporate event occurrences to locate related papers. Global relevance adopts a ‘one-for-all’ policy for recommending popular articles. Hybrid methods combine the above approaches to design efficient algorithms. We have reviewed articles implementing several versions of these classes, however minor customizations make it difficult to categorize the new methods to one of the base classes. We have defined the concept of ‘background knowledge and operating principle’ for proper classification and to make a clear distinction among the recommendation approaches. We have used a combination of systematic literature review, critical review, and conceptual review to conduct this survey which presents current advancements in the field and discusses popular recommendation approaches. This paper reveals the factors affecting users’ behavior and introduces a taxonomy of knowledge acquisition sources. Moreover, various evaluation methods and important performance criteria are explored. Finally, this paper examines the research trends and reports major loopholes in current research to foster the development of efficacious recommender systems.

SHARE: Designing multiple criteria-based personalized research paper recommendation system

Extraneous growth of scientific information over the Internet makes the searching task non-trivial and as a consequence researchers are facing difficulties in finding relevant papers from the millions of research papers in digital repositories. The research paper recommendation systems have been advocated to address this problem. The existing research paper recommendation systems lack in exploiting prominent information of papers, such as relevancy with the current time, novelty, scientific contribution, writing complexity of the papers, etc. Further, the existing models emphasize only on user’s preference rather than user’s intention that may change with time. Furthermore, the existing models do not consider a sound ranking strategy to unleash the personalization aspect and relevancy of papers. This work aims to address the existing limitations and proposes a systematic hidden attribute-based recommendation engine (SHARE). SHARE utilizes a feature engineering technique to unfold valuable insights of papers through multiple hidden features. These features are used as a context for users as well as multiple criteria for ranking papers. Additionally, SHARE predicts a user’s intention beyond the user’s preference to capture the dynamic notion of a user. Finally, a novel ranking strategy is proposed to retrieve personalized and the most important papers. SHARE is flexible for recommending both old and new users. In order to evaluate the effectiveness of SHARE both user studies and system evaluations were performed. Experimental results substantiate the efficacy of the proposed approach and are comparable to the existing systems.

Combining Citation Network Information and Text Similarity for Research Article Recommender Systems

Researchers often need to gather a comprehensive set of papers relevant to a focused topic, but this is often difficult and time-consuming using existing search methods. For example, keyword searching suffers from difficulties with synonyms and multiple meanings. While some automated research-paper recommender systems exist, these typically depend on either a researcher’s entire library or just a single paper, resulting in either a quite broad or a quite narrow search. With these issues in mind, we built a new research-paper recommender system that utilizes both citation information and textual similarity of abstracts to provide a highly focused set of relevant results. The input to this system is a set of one or more related papers, and our system searches for papers that are closely related to the entire set. This framework helps researchers gather a set of papers that are closely related to a particular topic of interest, and allows control over which cross-section of the literature is located. We show the effectiveness of this recommender system by using it to recreate the references of review papers. We also show its utility as a general similarity metric between scientific articles by performing unsupervised clustering on sets of scientific articles. We release an implementation, ExCiteSearch (bitbucket.org/mmmontemore/excitesearch), to allow researchers to apply this framework to locate relevant scientific articles.

Hidden features identification for designing an efficient research article recommendation system

The digital repository of research articles is increasing at a rapid rate and hence searching the right paper becoming a tedious task for researchers. A research paper recommendation system is advocated to help researchers in this context. In the process of designing such a system, proper representation of articles, more specifically, feature identification and extraction are two essential tasks. The existing approaches mainly consider direct features which are readily available from research articles. However, there are certain features which are not readily available from a paper, but may greatly influence the performance of recommendation systems. This paper proposes four indirect features: keyword diversification, text complexity, citation analysis over time, and scientific quality measurement to represent a research article. The keyword diversification measures the uniqueness of the keywords of a paper which helps variation in recommendation. The text complexity measurement helps to provide a paper by matching the user’s understandability level. The citation analysis over time decides the relevancy of a paper. The scientific quality measurement helps to measure the scientific values of papers. Formal definitions of the proposed indirect features, schemes to extract the feature values given a research article, and metrics to measure them quantitatively are discussed in this paper. To substantiate the efficacy of the proposed features, a number of experiments have been carried out. The experimental results reveal that the proposed indirect features uniquely define a research article than the direct features. Given a research paper, extraction of feature vector is computationally fast and thus feasible to filter a large corpus of papers in real time. More significantly, indirect features are matchable with user’s profile features, thus satisfying an important criterion in collaborative filtering.

Research paper recommender system based on public contextual metadata

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.

Influence of tweets and diversification on serendipitous 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.

Combining metadata and co-citations for recommending related papers

Identification of relevant documents is performed to keep track of the state-of-the-art methods and relies on research paper recommender systems. The proposed approaches for these systems can be classified into categories like content-based, collaborative filtering-based, and bibliographic information-based approaches. The content-based approaches exploit the full text of articles and provide more promising results than other approaches. However, most content is not freely available because of subscription requirements. Therefore, the scope of content-based approaches is limited. In such scenarios, the best possible alternative could be the exploitation of other openly available resources. Therefore, this research explores the possible use of metadata and bibliographic information to find related articles. The approach incorporates metadata with co-citations to find and rank related articles against a query paper. The similarity score of metadata fields is calculated and combined with co-citations. The proposed approach is evaluated on a newly constructed dataset of 5116 articles. The benchmark ranking against each co-cited document set is established by applying Jensen--Shannon divergence JSD and results are evaluated with the state-of-the-art content-based approach in terms of normalized discounted cumulative gain NDCG . The state-of-the-art content-based approach achieved an NDCG score of 0.86 while the traditional co-citation-based approach scored 0.72. The presented method achieved NDCG scores of 0.73, 0.77, and 0.78 by incorporating the title, co-citation and title, and abstract, respectively, whereas the highest NDCG score of 0.77 was achieved by combining co-citations with metadata. However, better results are achieved by incorporating the title and abstract with NDCG score of 0.81. Therefore, it can be concluded that the proposed approach could be a better alternative in cases where content is unavailable.

An Efficient Mining for Recommendation System for Academics

At present time huge numbers of research articles are available on World Wide Web in any domain. The research scholar explores a research papers to get the appropriate information and it takes time and effort of the researcher. In this scenario, there is the need for a researcher to search a research based on its research article. In the present paper a method of Knowledge ablation from a collection of research articles, is presented to evolve a system research paper recommendation system (RPRS), which would generate the recommendations for research article based on researcher choice. The RPRS accumulate the knowledge ablated from the pertinent research articles in the form of semantic tree. It accumulates all the literal sub parts with their reckoning in nodes. These parts are arranged based on their types in such a way that the leaf nodes stores the words with its prospect, the higher layer gives details about dictum with its reckoning, next to it an abstract. A Bayesian network is applied to construct a verisimilitude model which would quotation the pertinent tidings from the knowledge tree to construct the recommendation and word would be scored through TF-IDF value.

A Original or Exact Research Article Approbation Procedure

This essay establishes CIIT, the foremost hybrid research paper recommender system along with a powerful substitute to presently used academic search engines. CIIT develops the look of the typically implemented keyword-based search through uniting it by citation analysis, author analysis, implicit ratings, source analysis, explicit ratings and also, pioneering and yet idle methods similar to the ‘Distance Similarity Index’ (DSI) as well as the ‘In-text Impact Factor’ (ItIF). Rather than searching just keywords, a user may give whole documents, as well as reference lists as input as well as make implicit and explicit ratings to develop recommendations. With quotation, author along with source analysis, similar and associated documents are effortlessly determinable. Every of these methods are managed with a accessible GUI.

A collaborative approach for research paper recommender system.

Research paper recommenders emerged over the last decade to ease finding publications relating to researchers’ area of interest. The challenge was not just to provide researchers with very rich publications at any time, any place and in any form but to also offer the right publication to the right researcher in the right way. Several approaches exist in handling paper recommender systems. However, these approaches assumed the availability of the whole contents of the recommending papers to be freely accessible, which is not always true due to factors such as copyright restrictions. This paper presents a collaborative approach for research paper recommender system. By leveraging the advantages of collaborative filtering approach, we utilize the publicly available contextual metadata to infer the hidden associations that exist between research papers in order to personalize recommendations. The novelty of our proposed approach is that it provides personalized recommendations regardless of the research field and regardless of the user’s expertise. Using a publicly available dataset, our proposed approach has recorded a significant improvement over other baseline methods in measuring both the overall performance and the ability to return relevant and useful publications at the top of the recommendation list.

Retrieval by recommendation: using LOD technologies to improve digital library search

This paper investigates how Linked Open Data (LOD) can be used for recommendations and information retrieval within digital libraries. While numerous studies on both research paper recommender systems and Linked Data-enabled recommender systems have been conducted, no previous attempt has been undertaken to explore opportunities of LOD in the context of search and discovery interfaces. We identify central advantages of Linked Open Data with regard to scientific search and propose two novel recommendation strategies, namely flexible similarity detection and constraint-based recommendations. These strategies take advantage of key characteristics of data that adheres to LOD principles. The viability of Linked Data recommendations was extensively evaluated within the scope of a web-based user experiment in the domain of economics. Findings indicate that the proposed methods are well suited to enhance established search functionalities and are thus offering novel ways of resource access. In addition to that, RDF triples from LOD repositories can complement local bibliographic records that are sparse or of poor quality.

RARD: The Related-Article Recommendation Dataset

Recommender-system datasets are used for recommender-system offline evaluations, training machine-learning algorithms, and exploring user behavior. While there are many datasets for recommender systems in the domains of movies, books, and music, there are rather few datasets from research-paper recommender systems. In this paper, we introduce RARD, the Related-Article Recommendation Dataset, from the digital library Sowiport and the recommendation-as-a-service provider Mr. DLib. The dataset contains information about 57.4 million recommendations that were displayed to the users of Sowiport. Information includes details on which recommendation approaches were used (e.g. content-based filtering, stereotype, most popular), what types of features were used in content based filtering (simple terms vs. keyphrases), where the features were extracted from (title or abstract), and the time when recommendations were delivered and clicked. In addition, the dataset contains an implicit item-item rating matrix that was created based on the recommendation click logs. RARD enables researchers to train machine learning algorithms for research-paper recommendations, perform offline evaluations, and do research on data from Mr. DLib's recommender system, without implementing a recommender system themselves. In the field of scientific recommender systems, our dataset is unique. To the best of our knowledge, there is no dataset with more (implicit) ratings available, and that many variations of recommendation algorithms. The dataset is available at http://data.mr-dlib.org, and published under the Creative Commons Attribution 3.0 Unported (CC-BY) license.

A Model- Based Research Material Recommendation System For Individual Users

As there is an enormous amount of online research material available, finding pertinent information for specific purposes has become a tedious chore. So there is a requirement of the research paper recommendation system to facilitate research scholars in finding their interested and relevant research papers. There are many paper recommendation systems available, most of them are depending on paper assemblage, references, user profile, mind maps. This information is generally not easily available. The majority of the prevailing recommender system is based on collaborative filtering that rely on other user’s proclivity. On the other hand, content-based methods use information regarding an item itself to make a recommendation. In this paper, we present a research paper recommendation method that is based on single paper. Our method uses content-based recommendation approach that employs information extraction and text categorization. . It performs the profile learning by using naive Bayesian text classifier and generates recommendation on the basis of an individual’s preference.

Towards reproducibility in recommender-systems research

Numerous recommendation approaches are in use today. However, comparing their effectiveness is a challenging task because evaluation results are rarely reproducible. In this article, we examine the challenge of reproducibility in recommender-system research. We conduct experiments using Plista's news recommender system, and Docear's research-paper recommender system. The experiments show that there are large discrepancies in the effectiveness of identical recommendation approaches in only slightly different scenarios, as well as large discrepancies for slightly different approaches in identical scenarios. For example, in one news-recommendation scenario, the performance of a content-based filtering approach was twice as high as the second-best approach, while in another scenario the same content-based filtering approach was the worst performing approach. We found several determinants that may contribute to the large discrepancies observed in recommendation effectiveness. Determinants we examined include user characteristics (gender and age), datasets, weighting schemes, the time at which recommendations were shown, and user-model size. Some of the determinants have interdependencies. For instance, the optimal size of an algorithms' user model depended on users' age. Since minor variations in approaches and scenarios can lead to significant changes in a recommendation approach's performance, ensuring reproducibility of experimental results is difficult. We discuss these findings and conclude that to ensure reproducibility, the recommender-system community needs to (1) survey other research fields and learn from them, (2) find a common understanding of reproducibility, (3) identify and understand the determinants that affect reproducibility, (4) conduct more comprehensive experiments, (5) modernize publication practices, (6) foster the development and use of recommendation frameworks, and (7) establish best-practice guidelines for recommender-systems research.

Research-paper recommender systems: a literature survey

In the last 16 years, more than 200 research articles were published about research-paper recommender systems. We reviewed these articles and present some descriptive statistics in this paper, as well as a discussion about the major advancements and shortcomings and an overview of the most common recommendation concepts and approaches. We found that more than half of the recommendation approaches applied content-based filtering (55 %). Collaborative filtering was applied by only 18 % of the reviewed approaches, and graph-based recommendations by 16 %. Other recommendation concepts included stereotyping, item-centric recommendations, and hybrid recommendations. The content-based filtering approaches mainly utilized papers that the users had authored, tagged, browsed, or downloaded. TF-IDF was the most frequently applied weighting scheme. In addition to simple terms, n-grams, topics, and citations were utilized to model users' information needs. Our review revealed some shortcomings of the current research. First, it remains unclear which recommendation concepts and approaches are the most promising. For instance, researchers reported different results on the performance of content-based and collaborative filtering. Sometimes content-based filtering performed better than collaborative filtering and sometimes it performed worse. We identified three potential reasons for the ambiguity of the results. (A) Several evaluations had limitations. They were based on strongly pruned datasets, few participants in user studies, or did not use appropriate baselines. (B) Some authors provided little information about their algorithms, which makes it difficult to re-implement the approaches. Consequently, researchers use different implementations of the same recommendations approaches, which might lead to variations in the results. (C) We speculated that minor variations in datasets, algorithms, or user populations inevitably lead to strong variations in the performance of the approaches. Hence, finding the most promising approaches is a challenge. As a second limitation, we noted that many authors neglected to take into account factors other than accuracy, for example overall user satisfaction. In addition, most approaches (81 %) neglected the user-modeling process and did not infer information automatically but let users provide keywords, text snippets, or a single paper as input. Information on runtime was provided for 10 % of the approaches. Finally, few research papers had an impact on research-paper recommender systems in practice. We also identified a lack of authority and long-term research interest in the field: 73 % of the authors published no more than one paper on research-paper recommender systems, and there was little cooperation among different co-author groups. We concluded that several actions could improve the research landscape: developing a common evaluation framework, agreement on the information to include in research papers, a stronger focus on non-accuracy aspects and user modeling, a platform for researchers to exchange information, and an open-source framework that bundles the available recommendation approaches.

The Architecture and Datasets of Docear's Research Paper Recommender System

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.

Application of Content-Based Approach in Research Paper Recommendation System for a Digital Library

Recommender systems are software applications that provide or suggest items to intended users. These systems use filtering techniques to provide recommendations. The major ones of these techniques are collaborative-based filtering technique, content-based technique, and hybrid algorithm. The motivation came as a result of the need to integrate recommendation feature in digital libraries in order to reduce information overload. Content-based technique is adopted because of its suitability in domains or situations where items are more than the users. TF-IDF (Term Frequency Inverse Document Frequency) and cosine similarity were used to determine how relevant or similar a research paper is to a user's query or profile of interest. Research papers and user's query were represented as vectors of weights using Keyword-based Vector Space model. The weights indicate the degree of association between a research paper and a user's query. This paper also presents an algorithm to provide or suggest recommendations based on users' query. The algorithm employs both TF-IDF weighing scheme and cosine similarity measure. Based on the result or output of the system, integrating recommendation feature in digital libraries will help library users to find most relevant research papers to their needs.