USING GRAPH EMBEDDINGS FOR WIKIPEDIA LINK PREDICTION
Journal Title: Вісник Національного технічного університету «ХПІ». Серія: Системний аналiз, управління та iнформацiйнi технологiї - Year 2019, Vol 0, Issue 1
Abstract
Link prediction is an important area of study in network analysis and graph theory which tries to answer the question of whether two nodes in the graph might have an association in the future. Nowadays, graphs are ubiquitously present in our lives (social networks, circuits, roads etc.), which is why the problem is crucial to the development of intelligent applications. In the past, there have been proposed methods of solving link prediction problem through algebraic formulations and heuristics, however, their expressive power and transferability fell short. Recently, graph embedding methods have risen to popularity because of their effectiveness and the ability to transfer knowledge between tasks. Inspired by the famous in machine learning and natural language processing research Word2Vec approach, these methods try to learn a distributed vector representation, called an embedding, of graph nodes. After that a binary classifier given a pair of embeddings predicts the probability of the existence of a link between the encoded nodes. In this paper, we review several graph embedding approaches for the problem of Wikipedia link prediction, namely Wikipedia2vec, Role2vec, AttentionWalk and Walkets. Wikipedia link prediction tries to find pages that should be interlinked due to some semantic relation. We evaluate prediction accuracy on a hold-out set of links and show which one proves to be better at mining associations between Wikipedia concepts. The results include qualitative (principal component analysis dimensionality reduction and visualization) and quantitative (accuracy) differences between the proposed methods. As a part of the conclusion, further research questions are provided, including new embedding architectures and the creation of a graph embedding algorithms benchmark.
Authors and Affiliations
Roman Shaptala, Gennadiy Kyselev
Keywords
Related Articles
- EP ID EP603781
- DOI 10.20998/2079-0023.2019.01.09
- Views 96
- Downloads 0
ІНТЕЛЕКТУАЛЬНИЙ АНАЛІЗ ПРОПОЗИЦІЙ ТОВАРІВ НА ОСНОВІ КОНТЕКСТНИХ РЕКОМЕНДАЦІЙ
<p class="304">Інтернет-технології є невід’ємною складовою відносин, які виникають у сучасному суспільстві. Через швидке впровадження та зручність електронних майданчиків, прогнозовано зростає попит на ринку IT-продуктів...
MODELS FOR THE FORMATION OF IT COMPANY STRATEGIC PORTFOLIO OF PROJECTS
<p class="304">Increasing of unpredictability, novelty and complexity of the external environment of modern enterprises of IT industry has led to the need of creating such management mechanisms that can ensure making of...
Forecasting the results of football matches on the Internet based information
<span>The purpose of the article is making a model of results forecasting for football matches, which works better than bookmakers organizations. Lately the popularity of football forecasting has been increased. The exis...
PROBLEMS OF ELICITATION AND ANALYSIS OF REQUIREMENTS TO THE PROGRAM MODULE OF MONITORING ON THE BASIS OF THE CONDITIONS OF COGNITIVE ANALYSIS
There are risks of obtaining result that does not answer a work purpose statement when developing the software product for new subject area. It is necessary to pay attention on requirements to the software for eliminatio...
COST AND COMPLEXITY RESEARCH OF SOFTWARE DEVELOPMENT TO SOLVE THE PROBLEM OF INVENTORY MANAGEMENT
The article describes the process of estimating the cost and complexity of software development for the task of inventory management of a commercial enterprise to improve the decision-making process at the stage of forma...