CONSISTENCY AND STRUCTURE ANALYSIS OF SCHOLARLY PAPERS USING BASED ON NATURAL LANGUAGE PROCESSING
This research presents a comprehensive similarity analysis of the consistency of authors in crafting papers and providing simple conclusions or meanings in a journal. Machine learning technique are employed to assess the similarity and interpretation of these sentences. The study attempts to mine text data, making it more structured and easily understood, introducing an approach to identifying relevant author consistency in an extensive collection by utilizing text analysis and the understanding of the meanings of new words using Natural Language Processing. In the interim, the weight analysis was conducted through the validation of TF-IDF and Cosine Similarity. By conducting an in-depth analysis across the corpus dataset consisting of 60 to 150 journal documents, this research utilizes classification patterns, preprocessing patterns, similarity calculations, and interpretation of results. This study is able to provide information about how consistent researchers are in writing assembled journals. The results underscore the effectiveness of NLP in processing natural language, enhanced by the incorporation of TF-IDF and Cosine Similarity, which refine the representation of relevance in journal content.
Text Mining, Natural Language Processing, TF-IDF, Corpus, Cosine Similarity.