Research on Korean literature corpus processing based on computer system improved TF-IDF algorithm

Article type: Research Article

Authors: Xue, Jing

Affiliations: North China University of Water Resources and Electric Power, Zhengzhou, China | E-mail: [email protected]

Correspondence: [*] Corresponding author: North China University of Water Resources and Electric Power, Zhengzhou, China. E-mail: [email protected].

Abstract: Art is a symbol of people’s thoughts, and among many forms of artistic expression, literature is the most direct one, which can present art directly to people. How to correctly understand language materials in literature is crucial for understanding literary works and realizing their artistic value. Therefore, in order to strengthen the understanding of Korean literature and analyze its core ideas, this article utilizes modern computer technology and improved Term Frequency-Inverse Document Frequency (TF-IDF) algorithm to process the corpus of Korean literature, in order to quickly extract valuable textual information from Korean literature and facilitate reading and understanding. At the same time, a Korean literature corpus processing model was constructed based on deep learning algorithms. This model is based on the Natural Language Processing (NLP) algorithm, selecting Word Frequency Inverse Document Frequency (TF-IDF) as the feature to calculate the feature weight of keywords. By weighting the naive Bayesian algorithm, it achieves the classification and processing of expected text data in Korean literature. The results of multiple experiments show that the classification accuracy of the model exceeds 97.7%, and the classification recall rate is as high as 94.2%, indicating that the model can effectively achieve corpus processing in Korean literature.

Keywords: Korean literature, art, expect, TF-IDF algorithm, NLP, computer system

DOI: 10.3233/IDT-230772

Journal: Intelligent Decision Technologies, vol. Pre-press, no. Pre-press, pp. 1-14, 2024