Development of a File Management and Similarity Checker Software Using Semi-Supervised Learning

This paper presents the development of a file management and document similarity detection software capable of sorting PDF files based on keyword frequency using five distinct semi-supervised learning methods and algorithms. The methods and algorithms include the Cosine Similarity algorithm, the Best Match (BM25) algorithm, the Naïve Bayes algorithm, the K-Means Clustering algorithm, and a Support Vector Machine (SVM). Moreover, a semi-supervised learning (SSL) approach that combines the methods and algorithms has been implemented and tested using document files from an IEEE conference to ensure efficient operation and optimal performance. The results highlight the efficiency of the proposed approach in handling file management operations, tracking changes, and performing plagiarism detection tasks.