Improving Methodology for Selecting Fiction Works for Grades 5-9

Selecting appropriate fiction for secondary school students (grades 5–9) is critical to maintaining their interest and intellectual growth. This paper presents a new approach using natural language processing (NLP) techniques to improve the book selection process. We collected a specialized “School Literature Corpus” of the Uzbek School Literature textbooks from the website http://kitob.uz, and then pre-processed the text data, creating a unique dictionary and using TF-IDF for vector representation. We implemented cosine similarity to determine the most suitable books for each grade. Our results indicate that NLP can significantly improve the selection of relevant literature, which will ultimately benefit curriculum development and improve student learning. By ensuring that students read materials that are appropriate to their intellectual level, we can foster a deeper love of reading, improve literacy skills, and foster a strong reading culture. This, in turn, plays a vital role in their future academic success, critical thinking, and lifelong learning. Improving the methodology of book selection not only contributes to the achievement of immediate educational goals but also contributes to the development of well-rounded, informed, and intellectually curious people.