Search

Citation recommendation helps researchers perform reference searching more efficiently. Traditional methods often focus separately on diversification and personalization, each with unique advantages and limitations. In this study, we propose a new citation recommendation paradigm, personalized global citation recommendation with diversification awareness (PGCR-DA), which integrates the two approaches to generate more relevant candidate citations. Our work involved two major tasks. The first task involves generating a pool of diversified candidate citations for each target paper, by using the Random Walk with Restart on a constructed heterogeneous graph to identify the first relevant citation. The remaining diversified candidates are returned by using the Maximal Marginal Relevance model, where diversified citations are obtained based on a two-dimensional, i.e., the semantic space and publication date of the paper, diversification strategy. The second task focuses on personalization, where the ranking list obtained in the first task is reranked by modeling fine-grained and dynamic user preferences, informed by the analysis of both the textual and entity space from the users' previous publications. Preliminary experiments on the AAN and DBLP datasets verify our hypothesis that diversification and personalization can be effectively integrated through our approach. The results further demonstrate that PGCR-DA outperforms the competitive global citation recommendation methods with respect to a series of metrics. [ABSTRACT FROM AUTHOR]

With the field of technology has witnessed rapid advancements, attracting an ever-growing community of researchers dedicated to developing theories and techniques. This paper proposes an innovative ICRM (Intelligent Citation Recommendation Mechanism), designed to automate the process of suggesting the appropriate number of citations for individual brackets within a document. The proposed ICRM comprises three phases: Coarse-grained Weighted Bag of Word (WCBW), Fine-grained SciBERT (FSB) and Citation Adjustment phases. Firstly, the WCBW phase employs TF-IDF to extract keywords from both target and candidate documents, forming vectors that capture word significance along with metadata like authorship, keywords, and titles. It aims to identify relevant papers from a database, serving as initial candidates for each bracket. Secondly, the FSB phase employs the SciBERT model to assess the similarity between candidate documents and the local context around brackets, enhancing the precision of recommendations. It refines this selection by analyzing candidate-document relationships within the proximity of the brackets. Lastly, the Citation Adjustment phase tackles overlapping citations and ensures that recommended citation numbers align with user-defined criteria, resolving issues of imbalance. The simulation results demonstrate that the proposed ICRM outperforms existing models significantly in terms of precision, recall and F1-score. [ABSTRACT FROM AUTHOR]

The growing number of publications in the field of artificial intelligence highlights the need for researchers to enhance their efficiency in searching for relevant articles. Most paper recommendation models either rely on simplistic citation relationships among papers or focus on content-based approaches, both of which overlook interactions within academic networks. To address the aforementioned problem, knowledge graph embedding (KGE) methods have been used for citation recommendations because recent research proves that graph representations can effectively improve recommendation model accuracy. However, academic networks are dynamic, leading to changes in the representations of users and items over time. The majority of KGE-based citation recommendations are primarily designed for static graphs, thus failing to capture the evolution of dynamic knowledge graph (DKG) structures. To address these challenges, we introduced the evolving knowledge graph embedding (EKGE) method. In this methodology, evolving knowledge graphs are input into time-series models to learn the patterns of structural evolution. The model has the capability to generate embeddings for each entity at various time points, thereby overcoming limitation of static models that require retraining to acquire embeddings at each specific time point. To enhance the efficiency of feature extraction, we employed a multiple attention strategy. This helped the model find recommendation lists that are closely related to a user's needs, leading to improved recommendation accuracy. Various experiments conducted on a citation recommendation dataset revealed that the EKGE model exhibits a 1.13% increase in prediction accuracy compared to other KGE methods. Moreover, the model's accuracy can be further increased by an additional 0.84% through the incorporation of an attention mechanism. [ABSTRACT FROM AUTHOR]

With the exponential increase in the number of published articles, recommending them on the basis of the citation context (also called local or citation-aware citation recommendation) has attracted many researchers in the last few years. Recently, some papers have been devoted to reviewing previous works about scientific paper recommendation. As far as can be discerned, none of the previous review papers has carried out an in-depth study to explain citation context and compare previous studies. This paper presents a comparative analysis of recent studies about context-aware citation recommendation. Moreover, four gaps related to citation context extraction, citation context classification, temporal and structural aspects of a citation context, and benchmarking datasets are identified. This comparative study can assist researchers interested in further exploring these four gaps. [ABSTRACT FROM AUTHOR]

The process of retrieving suitable papers which are related to an interesting topic while doing research normally takes a lot of time and effort. Citation recommendation is frequently used to solve this problem by automatically suggesting a list of candidate papers that should match with the user's references or topics of interest. Applying the recent research results of deep learning in multiple disciplines, the performance of citation recommendation systems has been significantly improved with the facilitation of powerful deep neural data analysis and representation learning techniques. However, most of the recent Neural Citation Network (NCN) model-based techniques still encounter limitations related to the capability of integrating auxiliary information to assist the citation contextual learning process. Thus, in this paper, we propose a novel context-aware NCN-based model with the extra textual data integration and Bidirectional Encoder Representations from Transformers (BERT) model to improve the performance of the citation recommendation task. To do this, an extensive deep neural auto-encoding mechanism with a self-attention-based mechanism is utilized in our proposed model to flexibly learn both associated textual and citation contextual data in the given dataset. These enriched citation contextual information representations are then utilized to improve the performance of the citation recommendation task as the end-to-end neural learning process. Extensive experiments in the standard arXiv dataset show the effectiveness and good performance of the proposed model. [ABSTRACT FROM AUTHOR]

Personalized PageRank is a variant of PageRank, widely developed for citation recommendation. However, the personalized PageRank that works with a vast amount and rich scholarly data still results in information overload. Sometimes, junior scholars still need help to arrange queries quickly because of limited domain knowledge. Senior researchers need reference papers regarding a similar topic they intend to search for and related topics as a new insight. In this research, scientific citation recommendation aims to find the most influential papers with similar and related topics. Related topic papers in serendipitous perspectives are reference papers that are novel, diversified and unexpected to a user. The unexpectedness of recommended papers can be papers with different topics to queries but still relevant. To accomplish these challenges, we propose a framework of scientific citation recommendation with serendipitous perspectives. The framework includes feature extraction of an academic citation network, selection of multi-topic communities, and ranking papers in the selected multi-topic communities by modified PageRank. Papers in the chosen communities tend to link to similar and related papers. Modified PageRank is an extension of personalized PageRank, which works on multi-topic communities and manuscript queries. The experiments reveal that the proposed models outperform some models of personalized PageRank and some models of Content-Based Filtering. The multi-topic communities-based models work more effectively than the baselines if they run in a large dataset since the topic communities become more cohesive.

With the growth in scientific papers, citation recommendation which enables researchers to find useful references efficiently and further to promote academic communication and cooperation has become increasingly important. However, little research has been done to explore how to recognize the semantically relevant references according to research scenarios and the context of the paper citation. Motivated by the research gap, the present study attempts to adopt SciBERT to represent text and expand its semantics through the fusion of WordNet knowledge. Further, core themes from references are automatically extracted by TextRank to solve the problem of incomplete content extraction. In this case, the model named SciBERT + DPCNN is constructed for semantic representation and citation recommendation of scientific papers. Afterwards, multiple experiments are designed and implemented in three parts to verify the effectiveness of the model. The first result is that the outcomes of SciBERT + DPCNN obtain the highest among all baseline models. Additionally, when the model performs in 1 WordNet fusion at the end of the sentence, the best outcomes are 84.72%, 84.80%, 84.72%, and 84.71% in terms of accuracy, precision, recall, and F1-score, respectively. Ultimately, for the classification results of the reference structure, the long text 'title + abstract + TextRank full text (except the title and abstract)' outperforms most short text 'title + abstract' without WordNet fusion. However, when WordNet is fused for the classification, the short text is mostly more accurate than the long text. [ABSTRACT FROM AUTHOR]

The increasing number of patents leads patent applicants and examiners to spend more time and cost on searching and citing prior patents. Deep learning has exhibited outstanding performance in the recommendation of movies, music, products, and paper citation. However, the application of deep learning in patent citation recommendation has not been addressed well. Despite many attempts to apply deep learning models to the patent domain, there is little attention to the patent citation recommendation. Since patent citation is determined according to a complex technological context beyond simply finding semantically similar preceding documents, it is necessary to understand the context in which the citation occurs. Therefore, we propose a dataset named as a PatentNet to capture technological citation context based on textual information, meta data and examiner citation information for about 110,000 patents. Also, this paper proposes a strong benchmark model considering the similarity of patent text as well as technological citation context using cooperative patent classification (CPC) code. The proposed model exploits a two-stage structure of selecting based on textual information and pre-trained CPC embedding values and re-ranking candidates using a trained deep learning model with examiner citation information. The proposed model achieved improved performance with an MRR of 0.2506 on the benchmarking dataset, outperforming the existing methods. The results obtained show that learning about the descriptive citation context, rather than simple text similarity, has an important influence on citation recommendation. The proposed model and dataset can help researchers to understand technological citation context and assist patent examiners or applicants to find prior patents to cite effectively. [ABSTRACT FROM AUTHOR]

We address the problem of recommending relevant items to a user in order to "complete" a partial set of already-known items. We consider the two scenarios of citation and subject label recommendation, which resemble different semantics of item co-occurrence: relatedness for co-citations and diversity for subject labels. We assess the influence of the completeness of an already known partial item set on the recommender's performance. We also investigate data sparsity by imposing a pruning threshold on minimum item occurrence and the influence of using additional metadata. As models, we focus on different autoencoders, which are particularly suited for reconstructing missing items in a set. We extend autoencoders to exploit a multi-modal input of text and structured data. Our experiments on six real-world datasets show that supplying the partial item set as input is usually helpful when item co-occurrence resembles relatedness, while metadata are effective when co-occurrence implies diversity. The simple item co-occurrence model is a strong baseline for citation recommendation but can provide good results also for subject labels. Autoencoders have the capability to exploit additional metadata besides the partial item set as input, and achieve comparable or better performance. For the subject label recommendation task, the title is the most important attribute. Adding more input modalities sometimes even harms the results. In conclusion, it is crucial to consider the semantics of the item co-occurrence for the choice of an appropriate model and carefully decide which metadata to exploit. [ABSTRACT FROM AUTHOR]

Introduction. Citations of scholarly publications are considered an efficient measure of productivity of research and researchers. They are part of scholarly communication, driving the evolving knowledge in all disciplines. Citations form an integral part of literature practices of researchers. The latter are prone to deliberate or unconscious biases. One of the challenges all researchers face is to overcome or at least mitigate identified biases in citation. It may lead to distorting knowledge development in the least possible way. Purpose. As the research of citations is on the rise, more theoretical background is being developed. The editors call the JLE readers attention to the Triangular Citations, with various relationships among publications formed by citations. Results. Empirical studies are not unanimous in defining the features that affect citations most. Most researchers highlight a publication source as the most influential feature, with multi feature approach favoured by the academic community at large. Other features entail features relating to authors, journal, and publication itself. Moreover, there are varieties across the disciplines. The field of research of citation behaviour is only beginning to widen. The focus of such research is made on citation behaviour patterns, individual incentives behind the references and citations aligned with financial or reputation stimuli, and citations patterns linked to citation behaviour. Bias in citation threatens to distort knowledge and may evolve it subtly or obviously in a specific direction. Biases are not easy to deal with. In addition, values and mindsets vary across the countries and academic and scholarly communities that hinders efforts to overcome biases. Сonclusion. The JLE editors sum up the best guidelines on improvements in publications that add to greater citations, with the high quality of articles as the key.

Many researchers study citation recommendation approaches using network representation recently. It learns low-dimensional vector representation of nodes in a citation network, generates a recommendation list using similarity scores within the obtained node representations. Most of the existing approaches learn network representation by preserving structure information ofthe citation network. However, nodes in a citation network often associated with content information, recent approaches tend to learn each node's structure and content representations separately, and apply a simple and empirical combination strategy to produce the final node representation vectors which are suboptimal. To solve the above problems, we propose a Global-Local Neighborhoods based Network Representation model, named GLNNR, to integrate network structure and non- structural information, obtaining better node representation in the network. For global neighborhoods, we first generate parallel sequences containing node identity sequences and the corresponding content sequences, then we propose an attention- based sequence to sequence component to obtain node embeddings (the learned hidden representations of encoder) under global neighborhoods. For local neighborhoods, inspired by matrix factorization, a component can be designed to fuse structural and non-structural information in the lower-order neighborhood, here we first build a co-occurrence matrix (adjacent matrix) of the network, and then use multilayer perceptron to learn node representations under the lower-order neighborhood. The final node representations are global-local neighborhood based node embeddings. Empirical experiments prove the effectiveness of the GLNNR on a real-world information citation networks, i.e. CiteSeer and DBLP. Besides, we conduct experiments on a web page citation network, i.e. Wikipedia, to prove extensibility and portability of the proposed model. [ABSTRACT FROM AUTHOR]

Citation recommendation systems mainly help researchers find the lists of references that related to their interests effectively and automatically. The existing approaches face the issues of data sparsity and high-dimensional in large-scale bibliographic network representation, which hinder the citation recommendation performance. To address these problems, we proposed a Content-Sensitive citation representation approach for Citation Recommendation, named CSCR. Firstly, the Doc2vec model is used to generate a paper embedding according to paper content. Then, utilizing the similarity between the paper content embeddings to select the assumed neighbours of the target paper, append the auxiliary links between target paper and its new neighbours in the bibliographic network. Thirdly, distributed network representation method is implemented on appended bibliographic network to obtain the paper node embedding, which can learn interpretable lower dimension embedding for paper nodes. Finally, the embedding vectors of these papers can be used to conduct citation recommendation. Experimental results show that the proposed approach significantly outperforms other benchmark methods in Normalized Discounted Cumulative Gain (NDCG) and the positive rate (Recall). [ABSTRACT FROM AUTHOR]

Researchers frequently encounter the following problems when writing scientific articles: (1) Selecting appropriate citations to support the research idea is challenging. (2) The literature review is not conducted extensively, which leads to working on a research problem that others have well addressed. The study focuses on citation recommendation in the related studies section by applying the term function of a citation context, potentially improving the efficiency of writing a literature review.

When writing academic papers, referencing statements, claims and previous studies is always an important activity. However, it is considered challenging for scientists to find relevant and appropriate scientific articles which are closely related to their current works in order to reference them in their research. As a consequence of the rapid growth in scientific papers being published every year, researchers might easily get overwhelmed within a huge number of resources. One way to help them find the desired references more easily is to use context-aware citation recommendation. The citation recommender system can automatically provide a list of suitable papers as references based on specified inputs which reflect the researchers' interests. Among the outstanding achievements of deep learning and natural language processing in recent years, the utilization of deep neural learning architectures have supported to address the problem of citation recommendation. As the result, the neural citation recommendation area has received much attention from the academic community, with the aim of enhancing the precision and correctness of the results of existing citation recommendation systems. Following this research direction, in our paper we present a novel context-aware citation recommendation model, called RHN-DualLCR (Recurrent Highway Networks – Dual Local Citation Recommendation), which integrates Recurrent Highway Networks (RHN), an improved model of the original Bidirectional Long Short-Term Memory (BiLSTM) architecture, and uses a SciBERT-based (Science text of Bidirectional Encoder Representations from Transformers) embedding layer to build up the efficiency of the state-of-the-art local citation recommendation model, which enriches context representation with global information. Our research demonstrates its originality and relevance because we used one of the latest achievements of deep models (RHN model) and natural language processing (SciBERT) applied to the citation recommendation problem. We have conducted experiments on the RHN-DualLCR model on 3 widely known datasets for the citation recommendation problem: ACL-200 (Association for Computational Linguistics), ACL-600 and RefSeer, and also used 2 common evaluation standards Mean Reciprocal Rank (MRR) and the Recall@K (R@K for short) to evaluate the performance of our model. Experimental results show that our proposed model is 3% to 16% better than the original models or state-of-the-art models. [ABSTRACT FROM AUTHOR]

Proper citation is of great importance in academic writing for it enables knowledge accumulation and maintains academic integrity. However, citing properly is not an easy task. For published scientific entities, the ever-growing academic publications and over-familiarity of terms easily lead to missing citations. To deal with this situation, we design a special method Citation Recommendation for Published Scientific Entity (CRPSE) based on the cooccurrences between published scientific entities and in-text citations in the same sentences from previous researchers. Experimental outcomes show the effectiveness of our method in recommending the source papers for published scientific entities. We further conduct a statistical analysis on missing citations among papers published in prestigious computer science conferences in 2020. In the 12,278 papers collected, 475 published scientific entities of computer science and mathematics are found to have missing citations. Many entities mentioned without citations are found to be well-accepted research results. On a median basis, the papers proposing these published scientific entities with missing citations were published 8 years ago, which can be considered the time frame for a published scientific entity to develop into a well-accepted concept. For published scientific entities, we appeal for accurate and full citation of their source papers as required by academic standards. [ABSTRACT FROM AUTHOR]

Efficiently making adequate citations is becoming more challenging due to the rapidly increasing volume of publications. In practice, citing the appropriate references is a time-consuming and skill-required task. Accordingly, many studies have tried to help by providing citation-oriented support. In this field, citation recommendation is a significant research area because it addresses the problems of required profound skills and information overload. In this paper, we propose a sentence-level citation recommender, SentCite, that can identify the sentences that need links to references and can recommend citations. SentCite employs the convolutional recurrent neural network to extract the citing sentences and recommends citations based on the salient similarity between the sentences among the abstract, full text, and in-link context of the target papers. Unlike some other research in the big data domain, the recommended quality papers in this application are very limited. We proposed undersampling inlink context awareness to avoid overfitting problems. SentCite can recommend the most appropriate papers for the given sentences and outperforms other context-based methods in terms of improvement in mean reciprocal rank (MRR) 31.8%, mean average precision (MAP) 30.1%, and normalized discounted cumulative gain (NDCG) 33.8%. [ABSTRACT FROM AUTHOR]

Citation recommendation is a task that aims to automatically select suitable references for a working manuscript. This task has become increasingly urgent as the typical pools of candidates continue to grow, in the order of tens or hundreds of thousands or more. While several approaches to citation recommendation have been proposed in the literature, they generally seem to lack principled mechanisms to ensure diversity and other global properties among the recommended citations. For this reason, in this paper we propose a novel citation recommendation approach that leverages a submodular scoring function and a deep document representation to achieve an effective trade-off between relevance to the query and diversity of the references. To optimally train the scoring function and the deep representation, we propose a novel training objective based on a structural/multiclass hinge loss and incremental recommendations. The experimental results over three popular citation datasets have showed that the proposed approach has led to remarkable accuracy improvements, with an increase of up to 1.91 pp of MRR and 3.29 pp of F1@100 score with respect to a state-of-the-art citation recommendation system.

Automatic recommendation of citations has been a focal point of research in scholarly digital libraries. Many graph-based citation recommendation algorithms have been proposed; however, most of them utilize local citation behavior from the citation network that results in recommending papers in the same proximity as the query article. In this paper, we propose to capture the global citation behavior in the citation network and use it to enhance the citation recommendation performance. Specifically, we develop a novel citation network embedding algorithm, ConvCN, to encode the citation relationship among papers. We then propose to enhance existing graph-based citation recommendation algorithms by incorporating ConvCN to improve the recommendation efficacy. ConvCN has been shown to improve the citation recommendation performance by 44.86% and 34.87% on average in terms of Bpref and F-measure@20, respectively. The findings from this research not only confirm that global citation behavior could be additionally useful for improving the performance of traditional citation recommendation algorithms but also shed light on the possibility to adapt the proposed ConvCN algorithm for other recommendation tasks that rely on graph-like information such as items recommendation in social networks and people recommendation in referral networks. [ABSTRACT FROM AUTHOR]

In the era of rapid paper publications in various venues, automatic citation recommendations would be highly useful to researchers when they write papers. Local citation recommendation aims to recommend possible papers to cite given local citation contexts. Previous work mainly computes the similarity score between citation contexts and cited papers on a one-to-one basis, which is quite time-consuming. We train a pair of neural network encoders that map citation contexts and all possible cited papers to the same vector space, respectively. After that, we index the positions of all cited papers in the vector space. This makes our process for searching recommended papers considerably faster. On the other hand, existing methods tend to recommend papers that are highly similar to each other, which makes recommendations lack diversity. Therefore, we extend our algorithm to perform multi-topic recommendations. We generate multi-topic training examples based on the index we mentioned earlier. Furthermore, we specially design a multi-group contrastive learning method to train our model so that it can distinguish different topics. Empirical experiments show that our model outperforms previous methods by a wide margin. Our model is also light weighted and has been deployed online so that researchers can use it to obtain recommended citations for their own paper in real-time. • Proposed FLCR algorithm with sentence-transformer & k-d tree. • Introduced multi-topic citation recommendation for diverse contexts. • Developed large-scale dataset with 1.7 million citation contexts for evaluation. • Demonstrated significant performance improvement over previous methods. • Deployed demo for real-time citation recommendations. [ABSTRACT FROM AUTHOR]

Citation recommendation has gained increasing attention in recent years. In practice, researchers usually prefer to cite the most topic-relevant articles. Nevertheless, how to model the implicit correlations between topics and citations is still a challenging task. In this paper, we propose a novel citation recommendation model, called TopicCite, which mines such fine-grained correlations. We extract various citation features from citation network, and integrate the learning process of feature regression with topic modeling. At the recommendation stage, we expand the folding-in process by adding the topic influence of papers that correlated with user-provided information. TopicCite can also be considered a technique for extracting topic-related citation features from manually defined citation features, which can essentially improve the granularity of pre-extracted features. In addition, the unsupervised topic model is supervised and mutually reinforced by abundant citation features in TopicCite; thus, the proposed model can also extract more reliable topic distributions from citation data, which brings a new perspective to topic discovery on linked data. The experimental results on the AAN and DBLP datasets demonstrate that our model is competitive with the state-of-the-art methods.

With an increasing number of scholarly publications, accessing and retrieving appropriate papers is becoming an essential task for researchers. Citation recommendation, which can automatically provide a reference list based on a text segment, can overcome this problem. In this paper, we first construct a heterogeneous bibliographic network and deem citation recommendation as edge prediction problem, and then we develop a network representation-based edge prediction (NREP) model, which can simultaneously learn the edge prediction knowledge and the predictive representation for efficient citation recommendation. For personalized recommendation, we incorporate author information. We conduct extensive experiments on two datasets; the experimental results show that the NREP-based approach outperforms the other four state-of-the-art baseline approaches in terms of recall, mean average precision, and normalized discounted cumulative gain.

Recommendation systems assist web users with personalized suggestions based on past preferences for products, or items including documents, books, movies, and research papers. The plethora and variety of research papers on the Web and digital libraries make it challenging for researchers to find relevant publications to their scholarly interests. To cope with this inevitable challenge, various models and algorithms have been proposed to assist researchers with personalized citation recommendations. Nevertheless, so far, no research study has exploited the validity and suitability of evaluations conducted for these models to find the most promising among them. This study investigates and examines the existing citation recommendation algorithms based on the following criteria: evaluation methods adopted, comparative baselines employed, the complexity of the proposed algorithm, reproducibility of the experimental results, and consistency and universality of the evaluation methods. Besides this, our study presents a generic architecture and process of a typical citation recommendation system and provides a brief overview of information filtering methods used in the existing models. The findings of the study have implications for researchers and practitioners working on research paper recommendation and related areas. [ABSTRACT FROM AUTHOR]

Researchers need to formulate their achievements as research papers. Representative references are essential to high-quality papers. Academic citation recommendation refers to providing the recommendation of citations for the author of papers when they write. With the help of citation recommendation, researchers can improve the efficiency of writing academic papers and reduce the omission of important related literature. To achieve this goal, some methods were proposed. Many of them used citation networks to learn the representation of papers and chose references, they tended to ignore the content properties of papers. There are also some methods used partial properties to recommend citation. But their performance can be further improved. In this paper, we propose a citation recommendation method based on context correlation. We use two neural network models to learn the representations of papers and their references, then calculate the context similarity of them. Besides, we also introduce the publishing time and authority of papers, two key properties of papers for citation evaluation. In the experiment section, we compare our method with other methods and evaluate the performance of different properties choice in our method, it shows that our method outperforms some baselines and the combination of the dimensions including time, authority and context performs better. [ABSTRACT FROM AUTHOR]

Many state-of-the-art citation recommendation methods have been proposed for finding a list of reference papers for a given manuscript, among which the graph-based method has gained particular attention, due to its flexibility for incorporating various information that embodies user's preferences. To achieve a more synthetic, accurate, and personalized recommendation result than the previous graph-based methods, this paper proposes a new graph-based recommendation framework that exploiting diversified link information in a bibliographic network and the concise query information that embodies the specific requirement of user comprehensively. The proposed framework not only performs mutual reinforcement rules on all available multiple types of relations in a multi-layered graph but also incorporates the query information into the multi-layered mutual reinforcement schema to construct a multi-layered mutually reinforced query-focused (MMRQ) citation recommendation approach. Extensive experiments have been conducted on a subset of anthology network data set. Experimental results of Recall measures, normalized discounted cumulative gain measures, and case study all demonstrate that our MMRQ method obtains a superior citation recommendation.

Scientific articles store vast amounts of knowledge amassed through many decades of research. They serve to communicate research results among scientists but also for learning and tracking progress in the field. However, scientific production has risen to levels that make it difficult even for experts to keep up with work in their field. As a remedy, specialized search engines are being deployed, incorporating novel natural language processing and machine learning methods. The task of citation recommendation, in particular, has attracted much interest as it holds promise for improving the quality of scientific production. In this paper, we present the state-of-the-art in citation recommendation: we survey the methods for global and local approaches to the task, the evaluation setups and datasets, and the most successful machine learning models. In addition, we overview two tasks complementary to citation recommendation: extraction of key aspects and entities from articles and citation function classification. With this survey, we hope to provide the ground for understanding current efforts and stimulate further research in this exciting and promising field. [ABSTRACT FROM AUTHOR]

With the tremendous growth in the number of scientific papers being published, searching for references while writing a scientific paper is a time-consuming process. A technique that could add a reference citation at the appropriate place in a sentence will be beneficial. In this perspective, the context-aware citation recommendation has been researched for around two decades. Many researchers have utilized the text data called the context sentence, which surrounds the citation tag, and the metadata of the target paper to find the appropriate cited research. However, the lack of well-organized benchmarking datasets, and no model that can attain high performance has made the research difficult. In this paper, we propose a deep learning-based model and well-organized dataset for context-aware paper citation recommendation. Our model comprises a document encoder and a context encoder. For this, we use graph convolutional networks layer, and bidirectional encoder representations from transformers, a pre-trained model of textual data. By modifying the related PeerRead dataset, we propose a new dataset called FullTextPeerRead containing context sentences to cited references and paper metadata. To the best of our knowledge, this dataset is the first well-organized dataset for a context-aware paper recommendation. The results indicate that the proposed model with the proposed datasets can attain state-of-the-art performance and achieve a more than 28% improvement in mean average precision and recall@k. [ABSTRACT FROM AUTHOR]

Citation recommendation systems play an important role to alleviate the dilemma that scholar users spend a lot of time and experiences for literature survey. With the burgeoning computational models and open data movement, scientific repository can provide more evidence in support of recommendation. On the one hand, recommenders are applying better algorithms to understand the text of user queries and candidate citations. On the other hand, more types of data such as citation network and co-author relationship are aggregated to enrich the citation contextual information. The available data used for recommendation has been extended from textual content to enriched context. This review is conducted to identify the information and methods used for recommendations recently. We begin by introducing definitions of the task, recommending factors along with the corresponding problems and some application platforms. Then, we classify existing recommenders according to user query types and review representative methods for each type. We also elaborate on different strategies applied in three main stages of citation recommendation. Finally, a few open issues for future investigations are proposed. [ABSTRACT FROM AUTHOR]

In the era of big scholarly data, citation recommendation is playing an increasingly significant role as it solves information overload issues by automatically suggesting relevant references that align with researchers' interests. Many state-of-the-art models have been utilized for citation recommendation, among which graph-based models have garnered significant attention, due to their flexibility in integrating rich information that influences users' preferences. Co-authorship is one of the key relations in citation recommendation, but it is usually regarded as a binary relation in current graph-based models. This binary modeling of co-authorship is likely to result in information loss, such as the loss of strong or weak relationships between specific research topics. To address this issue, we present a fine-grained method for co-authorship modeling that incorporates the co-author network structure and the topics of their published articles. Then, we design a three-layered graph-based recommendation model that integrates fine-grained co-authorship as well as author-paper, paper-citation, and paper-keyword relations. Our model effectively generates query-oriented recommendations using a simple random walk algorithm. Extensive experiments conducted on a subset of the anthology network data set for performance evaluation demonstrate that our method outperforms other models in terms of both Recall and NDCG.

Researchers cite papers in order to connect the new research ideas with previous research. For the purpose of finding suitable papers to cite, researchers spend a considerable amount of time and effort. To help researchers in finding relevant/important papers, we evaluated textual and topological similarity measures for citation recommendations. This work analyzes textual and topological similarity measures (i.e., Jaccard and Cosine) to evaluate which one performs well in finding similar papers? To find the importance of papers, we compute centrality measures (i.e., Betweeness, Closeness, Degree and PageRank). After evaluation, it is found that topological-based similarity via Cosine achieved 85.2% and using Jaccard obtained 61.9% whereas textualbased similarity via Cosine on abstract obtained 68.9% and using Cosine on title achieved 37.4% citation links. Likewise, textual-based similarity via Jaccard on abstract obtained 35.4% and using Jaccard on title achieved 28.3% citation links.

Due to the development of academic, more and more attentions are paid to citation recommendation. To solve the citation recommendation problem, researchers begin to focus on the network representation, because it fuses semantic information and structural information well. It is a big challenge that how to map articles in a heterogeneous information network into a low-dimensional space while preserving the potential associations between articles. We propose a novel citation recommendation algorithm based on citation tendency, named CIRec which learns more about the potential relationship of articles in the process of network embedding. Citation tendency means if an article can be selected as a reference, it probability satisfies some kinds of conditions. In our algorithm, five weight matrices which represent the probability of entity-to-entity migration based on citation tendency are defined to build weighted heterogeneous network first. Second, we design a biased random walk procedure which efficiently explores articles' characteristics and citations information. Finally, the skip-gram model is used to learn the neighborhood relationship of the nodes in the walk sequence and map the nodes to the vector space. Comparing with existing state-of-the-art technique, experiment results show that our algorithm CIRec has better recall, precision, NDCG on AAN and DBLP dataset. [ABSTRACT FROM AUTHOR]

Citation recommendation is an interesting and significant research area as it solves the information overload in academia by automatically suggesting relevant references for a research paper. Recently, with the rapid proliferation of information technology, research papers are rapidly published in various conferences and journals. This makes citation recommendation a highly important and challenging discipline. In this paper, we propose a novel citation recommendation method that uses only easily obtained citation relations as source data. The rationale underlying this method is that, if two citing papers are significantly co-occurring with the same citing paper(s), they should be similar to some extent. Based on the above rationale, an association mining technique is employed to obtain the paper representation of each citing paper from the citation context. Then, these paper representations are pairwise compared to compute similarities between the citing papers for collaborative filtering. We evaluate our proposed method through two relevant real-world data sets. Our experimental results demonstrate that the proposed method significantly outperforms the baseline method in terms of precision, recall, and F1, as well as mean average precision and mean reciprocal rank, which are metrics related to the rank information in the recommendation list.

With the rapid proliferation of information technology, researchers find it more and more difficult to rapidly find appropriate reference papers for an authoring paper. Citation recommendation aims to overcome this problem by providing a list of reference papers given a query document. There exist various aspects in bibliographic literature acting as paper's scholarly roles, such as paper's content, paper's author, citation behavior, paper's topic. We argue that combining different kinds of paper's scholarly roles can enhance citation recommendation performance. Based on it, we propose a network correlation based query-oriented citation recommendation approach. We first construct a semantic network and a citation network, these two networks consist of the same vertices but different edge connection. Then we build correlations of these two networks and select the top features to calculate the semantic similarities of the query paper and scientific papers. Finally, we choose the top ranked scientific papers as the recommended citation list. When evaluating on the AAN dataset, the experimental results demonstrate the efficacy of the proposed approach. [ABSTRACT FROM AUTHOR]

Citation recommendation is the task of suggesting a list of references for an author given a manuscript. This is important for academic research for it provides an efficient and easy way to find relevant literatures. In this paper, we propose a novel probabilistic topic model to automatically recommend citations for researchers. The model considers not only text content similarity between papers but also community relevance among authors for effective citation recommendation. To fully utilize content and diversified link information in a bibliographic network, we extend LDA with matrix factorization, so that semantic topic learning and community detection are essentially reinforcing each other during parameter estimation. We also develop a flexible way to generate a family of citation link probability functions, which can substantially increase the model capacity. Experimental results on the ANN and DBLP dataset show that our model outperforms baseline algorithms for citation recommendation, and is capable of generating qualified author communities and topics. [ABSTRACT FROM AUTHOR]