FORMATION OF A DATABASE FOR SENTIMENT ANALYSIS OF TEXTS IN THE UZBEK LANGUAGE

In sentiment analysis of user comments, we first need to start with pre-processing the comments. Because the commentary texts were written by different people in different languages, with different spelling mistakes in the writings. If the input texts for classification algorithms in data mining are pre-processed, the accuracy of the sentiment analysis algorithm will increase and we can achieve the expected result. Solving such problems is an important task of natural language processing. In this article, we have prepared a Dataset using feedback given to restaurants located in the city of Tashkent on the Google map and analyzed Sentiment using logistic regression models. Overall evaluation results show that the system performs well by performing pre-processing steps such as stemming for agglutinative languages