Aspect-Based Community Detection of Cultural Heritage Streaming Data

Abstract

Twitter is considered a major and very popular social network providing an abundance of data generated by users’ interactions through tweets. After an appropriate analysis of this information, sets consisting of users who share similar attributes, and preferences can be identified. Massive cultural content management is important because reviews can be analyzed for extracting significant representations. In this study, an aspect mining method of a cultural heritage approach by incorporating big data methods, is proposed. We propose the combination of a community detection algorithm, i.e., the Parallel Structural Clustering Algorithm for Networks (PSCAN), with topic modelling methods, i.e., the Latent Dirichlet Allocation (LDA), for performing large-scale data analysis in Twitter.