Learning reputation in an authorship network

Seventh IEEE International Conference on Data Mining (ICDM 2007), 2007

The problem of evaluating scientific publications and their authors is important, and as such has attracted increasing attention. Recent graph-theoretic ranking approaches have demonstrated remarkable successes, but most of their applications are limited to homogeneous networks such as the network of citations between publications. This paper proposes a novel method for co-ranking authors and their publications using several networks: the social network connecting the authors, the citation network connecting the publications, as well as the authorship network that ties the previous two together. The new coranking framework is based on coupling two random walks, that separately rank authors and documents following the PageRank paradigm. As a result, improved rankings of documents and their authors depend on each other in a mutually reinforcing way, thus taking advantage of the additional information implicit in the heterogeneous network of authors and documents. The proposed ranking approach has been tested using data collected from CiteSeer, and demonstrates a great improvement in author ranking quality compared with ranking by the number of publications, the number of citations and the PageRank calculated in the authors' social network. This paper proposes a framework for co-ranking entities of different kinds in a heterogeneous network connect

2020

We consider a co-authorship network in “Information security” field. The network is constructed using Scopus data for the Ukrainian affiliated authors. We define the key centrality indicators: centrality degree, betweenness centrality and weighted centrality. We have made rankings of authors by centrality indicators and citations and aggregated this data with decision support methods. We demonstrate the methodology and possibility of defining expert groups and academic schools using the scientific database’s content. We propose to use decision support methods to define most communicative and cited scientists within co-authorship networks and demonstrate the way of ordinal factorial analysis usage for defining the relative weights of different centrality indicators within complex networks. Empirical results, obtained in the paper, indicate that there are no strong connection between given author’s centrality indicators and the number of citations to the author’s works. However, some ...

Expert finding is an information retrieval task that is concerned with the search for the most knowledgeable people with respect to a specific topic, and the search is based on documents that describe people's activities. The task involves taking a user query as input and returning a list of people who are sorted by their level of expertise with respect to the user query. Despite recent interest in the area, the current state-of-the-art techniques lack in principled approaches for optimally combining different sources of evidence. This article proposes two frameworks for combining multiple estimators of expertise. These estimators are derived from textual contents, from graph-structure of the citation patterns for the community of experts, and from profile information about the experts. More specifically, this article explores the use of supervised learning to rank methods, as well as rank aggregation approaches, for combing all of the estimators of expertise. Several supervised learning algorithms, which are representative of the pointwise, pairwise and listwise approaches, were tested, and various state-of-the-art data fusion techniques were also explored for the rank aggregation framework. Experiments that were performed on a dataset of academic publications from the Computer Science domain attest the adequacy of the proposed approaches.

This paper investigates the problem of finding author interest in coauthor network through topic modeling with providing several performance evaluation measures. Intuitively, there are two types of explicit grouping exists in research papers (1) authors who have co-authored with author A in one document (subgroup) and (2) authors who have co-authored with author A in all documents (group). Traditional methods use graph-link structure by using keywords based matching and ignored semantics-based information, while topic modeling considered semantics-based information but ignored both types of explicit grouping e.g. State-of-the-art Author-Topic model used only one kind of explicit grouping single document (subgroup) for finding author interest. In this paper, we introduce Group-Author-Topic (GAT) modeling which exploits both types of grouping simultaneously. We compare four different topic modeling methods for same task on large DBLP dataset. We provide three performance measures for method evaluation from different domains which are; perplexity, entropy, and prediction ranking accuracy. We show the trade of between these performance evaluation measures. Experimental results demonstrate that our proposed method significantly outperformed the baselines in finding author interest. The trade of between used evaluation measures shows that they are equally useful for evaluating topic modeling methods.

a challenging task with continuing growth in the size of computationally analysable data and dynamic nature of scholarly networks. This study ranks the authors not only on the basis of their own work but, also measures the impact of mutual influence of authors when they work in collaboration. The work of an author is influenced by the work of his or her co-authors, more in the case if co-authors are senior and less otherwise. This study proposes MuInf Rank (Mutual Influence based Ranking algorithm), which is calculated via a series of experiments. First, we calculate Mutual Influence with respect to number of papers, second, the number of citations and third, we calculate the combined effect of papers and citations with respect to the mutual influence of authors. It is shown empirically that the proposed methods produce satisfying results. have more chances to flourish in future. A researcher who is collaborating with more researchers receives influence by more researchers. Hence, he or she will have high impact of other's work if the collaborators are seniors. Contemplating the impact of an author's own work along with the influence of his or her collaborators give a complete picture of the standing of an author. Existing methods compute the rank of authors based on their in-links information. The proposed method not only involves the in-links information of a node, but also considers out-links of an author. These out-links are determined from the co-authorship relation from a network. This study considers a bibliography network and its essential contributions are finding the

Proceedings of the 20th …, 2004

We introduce the author-topic model, a generative model for documents that extends Latent Dirichlet Allocation (LDA; Blei, Ng, & Jordan, 2003) to include authorship information. Each author is associated with a multinomial distribution over topics and each topic is associated with a multinomial distribution over words. A document with multiple authors is modeled as a distribution over topics that is a mixture of the distributions associated with the authors. We apply the model to a collection of 1,700 NIPS conference papers and 160,000 CiteSeer abstracts. Exact inference is intractable for these datasets and we use Gibbs sampling to estimate the topic and author distributions. We compare the performance with two other generative models for documents, which are special cases of the author-topic model: LDA (a topic model) and a simple author model in which each author is associated with a distribution over words rather than a distribution over topics. We show topics recovered by the authortopic model, and demonstrate applications to computing similarity between authors and entropy of author output.

Finding rising stars is a challenging and interesting task which is being investigated recently in co-author networks. Rising stars are authors who have a low research profile in the start of their career but may become experts in the future. This paper introduces a new method Weighted Mutual Influence Rank (WMIRank) for finding rising stars. WMIRank exploits influence of co-authors' citations, order of appearance and publication venues. Comprehensive experiments are performed to analyze the performance of WMIRank in comparison to baseline methods, which have ignored weighted mutual influence. AMiner 1 data for years 1995-2000 is used for experiments. List of top 30 authors as per proposed and baseline methods are compared for their average number of papers, average number of citations and achievements. Experimental results provide convincing evidence of the effectiveness of the investigated weighted mutual influence.

2016

The prediction of relationships in a social network is a complex and extremely useful task to enhance or maximize collaborations by indicating the most promising partnerships. In academic social networks, prediction of relationships is typically used to try to identify potential partners in the development of a project and/or co-authors for publishing papers. This paper presents an approach to predict coauthorships combining artificial intelligence techniques with the state-of-the-art metrics for link predicting in social networks.

Future Generation Computer Systems, 2018

Recently, link prediction studies on large-scale and complex networks have particularly become the focus of interest for researchers in various scientific fields. Many complex networks created from the real-world data contain bipartite structure by nature. Bipartite networks are a kind of complex networks that represent the interactions between the different node groups. Almost all of the previous studies on link prediction in bipartite networks focus on using the properties of projection networks to predict the relations between the node pairs. In this study, a novel similarity-based link prediction method based on strengthening weighted projection is proposed to predict the potential links between the authors and the topics in the large-scale bipartite academic information network created from the real-world data. Since information loss occurs when bipartite networks are converted to unimodal networks, it should be noted that when making a link prediction in this paper, both the bipartite network and the information on strengthening unimodal network obtained from the bipartite network are used. To evaluate the proposed method, a bipartite network was first created from a real data set consisting of authors and their work. Then the method was tested on this network. Experimental results demonstrate that it is possible to obtain faster and more accurate link prediction results by the proposed method.

2004

We propose a new unsupervised learning technique for extracting information from large text collections. We model documents as if they were generated by a two-stage stochastic process. Each author is represented by a probability distribution over topics, and each topic is represented as a probability distribution over words for that topic. The words in a multi-author paper are assumed to be the result of a mixture of each authors ’ topic mixture. The topic-word and author-topic distributions are learned from data in an unsupervised manner using a Markov chain Monte Carlo algorithm. We apply the methodology to a large corpus of 160,000 abstracts and 85,000 authors from the well-known CiteSeer digital library, and learn a model with 300 topics. We discuss in detail the interpretation of the results discovered by the system including specific topic and author models, ranking of authors by topic and topics by author, significant trends in the computer science literature between 1990 and...