Institutional Repository of Vilnius University
Lietuviškų pradinių klasių matematikos uždavinių automatinis klasifikavimas į kategorijas ir tipus /
Abstract
dc:descriptionMathematical word problems are an important part of mathematics education; however, their analysis is often performed manually, making it complex and time-consuming. This study investigates the possibilities of automating the classification of Lithuanian primary school mathematics word problems using natural language processing and machine learning methods. In the experiments, different text preprocessing methods, text vectorization approaches (bag-of-words and TF-IDF vectors), and classification models were examined. The results showed that for category classification, the bag-of-words vectorization method combined with a support vector machine classifier performs best, achieving 93.33% accuracy. For type classification, the best results were obtained using a three-stage hierarchical architecture with bag-of-words vectors and language model embeddings, reaching a macro F1 score of 0.68. This research contributes to more efficient analysis of Lithuanian mathematical word problems.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Bareikaitė, Emilija,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD308334084&prefLang=en_US