Pemodelan Topik Skripsi Menggunakan Metode LDA

Abstract

Writing is one of the most important activities to express ideas in visual form. However, many students have difficulty in writing research. The difficulty experienced by students lies in choosing a topic that will be developed in the background. The topics needed in research writing are topics that are currently being discussed by the community so that initial observations are needed to get the appropriate topic. However, not infrequently the observation stage will take a long time, so a solution is offered by building an LDA model to assist students in choosing a thesis topic. This study uses an unsupervised learning algorithm, namely LDA or (Latent Dirichlet Allocation) for topic modeling. The best model produced is the LDA model with a combination of 40 topics, alpha 0.4, beta symmetric, and corpus tf-idf with a coherence score of 0.43 and a perplexity of 199.09.