TF-IDF and Inverse Document Frequency are both important algorithms in related fields of computer science and information retrieval. A large number of scholars and researchers have previously conducted a lot of research on Inverse Document Frequency and Term Frequency-Inverse Document Frequency. Because of their huge application value, Inverse Document Frequency and Term Frequency-Inverse Document Frequency have been widely used in different fields. Many people have also studied novel algorithms based on TF-IDF, such as the BM25 algorithm. In the field of education, people use Inverse Document Frequency and Term Frequency-Inverse Document Frequency to mine education-related data for insights and knowledge useful for education. In this research, we invented and proposed 7 novel algorithms based on TF-IDF and Inverse Document Frequency. The purpose of this research is to improve more and more accurate text mining-related algorithms for the fields of computer science and information retrieval, and to provide solutions to people's different needs. We use 7 novel algorithms that we invented and proposed in this study. By conducting experiments on 20 hand-made and educational-related background data, we found that through our novel algorithms we can improve the importance of words in documents, calculation accuracy, and find words whose importance is overestimated by TF-IDF, etc. In this study, we used research methods such as algorithm design and optimization, and application examples.