Visual Analysis of Citation Context-Based Article Influence Ranking

Chen Shi,Haoxuan Wang,Binjie Chen,Zhiguang Zhou,Yuhua Liu

doi:10.1109/access.2019.2932051

Chen Shi, Haoxuan Wang + Show 3 more

Open Access

https://doi.org/10.1109/access.2019.2932051

Copy DOI

Abstract

Article influence ranking is an effective way to reduce information redundancy and improve the efficiency of article retrieval. A large number of ranking models for network items have been employed for the ranking of article influence, such as PageRank and Spamming-resistant Expertise Analysis and Ranking. However, the effectiveness of article influence ranking based on the models of PageRank and SPEAR declines with the rapid growth of academic datasets, because of the increasing complexity of citation network. In order to take a rich set of contextual structures of citation context into consideration, we propose a visualization system VAIR for the citation context-based article influence ranking. Firstly, the word2vec model, a renowned technique in the field of natural language processing, is applied to transform articles into vectorized representations according to citation context. Then, a novel citation context-based article influence ranking model is designed according to the complex relationships quantified in a semantic vectorised space. Several visual designs are implemented, allowing users to perceive and compare the ranking results visually and intuitively. A set of user-friendly interactions are provided in the visualization framework, enabling users to explore the desirable article influence and obtain deep insights into the ranking model. Moreover, a series of case studies and comparison experiments are carried out based on real-world datasets, which further demonstrate the effectiveness of our algorithm for article influence ranking.

Highlights

With the development of the academia and the increase in publishing scale, a large amount of academic articles has been shared, effectively promoting academic communication and innovation [1], [2]
EVALUATION A web-based system implemented using Javascript supports the real-time exploration of visual analysis of citation context based on article influence ranking
1) VISUAL EXPLORATION OF CITATION CONTEXT An analogy between citation contexts and natural language processing (NLP) terms is used to explore the implicit semantic relationship in a citation network, which is trained by the word2vec model and projected into a two-dimensional space for intuitive display

Summary

INTRODUCTION

With the development of the academia and the increase in publishing scale, a large amount of academic articles has been shared, effectively promoting academic communication and innovation [1], [2]. Citation context in structure features may be an important clue for optimising the ranking of article influence, and studies are required to investigate this effect. As a popular representation learning model with good learning quality and fast computing speed, word2vec can be used to learn representations of contextual features in various data by a flexible analogy between domain interactions and NLP terms [30], [32] It constructs a vectorized space where the geometric distribution of vectors is able to reflect contextual information of original data. For large-scale academic data, we design a visual analysis framework to evaluate the academic influence of articles from the perspective of citation context in network structure.

RELATED WORKS

CITATION CONTEXT-BASED RANKING MODEL

VISUAL RANKING SYSTEM

CONCLUSION