Темир йўл транспорти: долзарб масалалар ва инновациялар 6-jild 2-son (2025) · 119-124-betlar
QUANTUM ONE-HOT ENCODING ALGORITHM FOR TEXT VECTORIZATION: QUANTUM ONE-HOT ENCODING ALGORITHM FOR TEXT VECTORIZATION
Ниёзматова, Н.А., Маматов, Н.С., Турғунова, Н.М.
Annotatsiya
This article focuses on the vectorization of textual data using a one-hot encoding algorithm based on quantum computing. The proposed quantum one-hot encoding algorithm enables each word to be represented as a quantum state through vectors. For the experiments, real-world complaints were collected and classified by service type. Features generated via both classical and quantum one-hot algorithms were used for classification through models such as Decision Tree, Random Forest, Naive Bayes, and Passive Aggressive. The research results indicate that the quantum one-hot encoding algorithm provides approximately 10% higher accuracy compared to the classical approach. Among the classifiers, the Passive Aggressive model achieved the highest accuracy of 94%. These findings confirm that the quantum one-hot encoding method is highly efficient and holds promise for natural language processing tasks. The approach is recommended as an effective complement to traditional encoding techniques.
Квантовое one-hot кодирование, классификация текста, машинное обучение, Decision Tree, Random Forest, Naive Bayes, Passive Aggressive, текстовые жалобы.Quantum one-hot encoding, text classification, machine learning, Decision Tree, Random Forest, Naive Bayes, Passive Aggressive, textual complaints.
Metadata manbasi: jurnal OAI-PMH arxivi · Sindex toʻliq matnni saqlamaydi, manbaga havola beradi.