Research topics can be seen from Abstraction research documents, for example, reports Scientific Writing (KTI) in the form of Final Project, Thesis and Dissertation. Research Topics of KTI is a collection of important words that indicate the area / field of study of the KTI. A guided KTI some supervisor, and a lecturer normally would guide some particular topic. Some lecturers have the same field of research formed a research group within the Department, but some courses are lecturers who exhibit similarities field of research. At this thesis proposed a method for classifying Writer (Lecturer) based on common research topics in Co-Authorship Graph using the Hypergraph Partitioning, making it possible to create a research group within the scope of inter Programs or college level. The method is divided into three stages: extraction of research topics, pembentuksn Co-Authorship Graph, and grouping author. Extraction of research topics, get the topic of EI by Title and Abstract using Latent Dirichlet Allocation (LDA). Formation of Co-Authorship Graph, where the nodes are the author, edge is the collaborative relationship and similarity / resemblance of research topics, and the weighting edge is Jaccard and cosine values similary research topics between author. Grouping Writers on Co-Authorship Graph using the Hypergraph Partitioning. Test method uses data from the Research Institute of Research and Community Service (LPPM) ITS. Grouping the results are validated using the Silhouette and Entropy. The final results showed that the grouping has been formed group Authors whose members come from the Department or a different field, with high similarity topic.
How to Cite
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.