Social Network Analysis Text Mining pada Jaringan Sitasi Publikasi

Abstract

The problem of analysing publication citation networks has recently been widely researched, but many studies only analyse one or several types of journals due to differences in the metadata of each journal. Apart from that, there are also those who only count the number of citations without taking into account the distribution of citations, which includes how far apart they are based on the distance between the author's university and the university of the author citing the publication. This research analyses the publication citations of University of North Sumatra lecturers from Google Scholar using a text mining approach, and then the text mining results obtained visualise the distribution of publication citations through the publication citation network formed from each node. Due to the complexity of the network formed, including differences in metadata for each journal and the wide distribution of publication sources, it is difficult to build network citations. This research aims to build a visualization model of the distribution of lecturer publications at the Universitas Sumatera Utara. Text mining was performed by web scraping on Google Scholar with Python. The data used is lecturer data on the SINTAScience and Technology Index website, which includes publication data from lecturers at the University of North Sumatra, consisting of 1730 records. The method used is a quantitative approach with computational text mining and web scraping algorithms. Social Network Analysis (SNA) is used to build the network formed between authors. The research results show that the proposed method can solve the problem of publication citation network visualization. The resulting visualisation of network citation publications can provide valuable knowledge about the structure and patterns of connectivity in network citation publications.