Framework for Analysis of Thematic Interest in Scientific Networks using Fuzzy Time Series

Abstract

The active growth of the scientific social networks is determined by increase of researchers and directions of investigations. This activity generates big data and analysis of such data could give not obvious knowledge about researchers, their thematic interest and direction of scientific development. One of the ways to know the scientific development is forecasting of the thematic interest using data from scientific social networks and cites. The goal of the paper is to propose the framework and techniques of mining this information on the basis of the publication activity prediction and linguistic representation of knowledge in the space and in the temporal areas. The fuzzy time series models, linguistic variables and transformations are combined in the proposed framework to achieve the goal. The proposed framework is applied in case study to determine the trends in scientific interest in selected topics.