Abstract:
In this thesis, we mentioned a system that assigns scores indicating positive or negative to translated data. Much works have been done on sentiment analysis, document clustering for newspaper data in bangla language. News from one Bengali newspaper is used for the purpose of the project. we are using web crawler to get necessary news to make a dataset to use for this project. With the significant increase of person interactions through outstanding advances of the Web, sentiment analysis has acquired more focus from an educational and a industrial point of view. Recently, sentiment analysis in the Bangla language is step by step being regarded as an vital task, for which previous methods have tried to notice the universal polarity of a Bangla document. This can be described as being due to the lack of reachable datasets for bangle text analysis .These high massive unstructured web contents will be utilized to create smarter tools to help people by natural language Processing (NLP). Though Bengali NLP tools are still insufficient because of its natural complexities, research on Sentiment Analysis in Bengali is flourishing as a challenging area and is getting researcher‟s attention at a rapid pace. In this paper, we apply publicly reachable datasets to operate sentiment analysis in Bangla text. One of the datasets consists of human-annotated person remarks on dataset consists of person opinions of overall review. We additionally describe a baseline approach for the subtask of component class extraction to consider our datasets. Through this thesis, our destination is to establish a system, which can identify positive an negative news accurately.
Description:
This thesis submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Information and Communication Engineering of East West University, Dhaka, Bangladesh.
