Intent Classification Dataset for Student Question

This dataset contains categorized questions collected from high school (secondary school) students to support research in intent classification and natural language processing (NLP). Each question was manually labeled into one of two main intent categories, providing a structured dataset for developing and evaluating chatbots that can interpret and respond to student inquiries effectively.The dataset was developed as part of the study titled “A Chatbot Intent Classifier for Supporting High School Students” by Suha Assayed, Khaled Shaalan, and Manar Alkhatib, published in <i>EAI Endorsed Transactions on Scalable Information Systems</i> (Volume 1, December 21, 2023).The categories and their corresponding question counts are as follows:Schools: 384 questions related to high school curriculum, subjects, and academic informationUniversities: 585 questions related to university admission, study programs, and related topicsTotal number of questions: 969This dataset was used to train and evaluate intent classification models, including Multinomial Naive Bayes and Random Forest, both achieving high accuracy levels exceeding 90% across key evaluation metrics (accuracy, precision, recall, and F1 score).ReferencesAssayed SK, Shaalan K, Alkhatib M. A Chatbot Intent Classifier for Supporting High School Students. EAI Endorsed Scal Inf Syst [Internet]. 2022 Dec. 21 [cited 2025 Oct. 20];10(3):e1. Available from: https://publications.eai.eu/index.php/sis/article/view/2948