Instruments and Systems: Monitoring, Control, and Diagnostics Annotation << Back Comparative Analysis of the Effectiveness of VADER NLTK and ROBERTA in the Context of Sentiment Analysis A.U. MENTSIEV, F.P. NGUEN, R.S. ZARIPOVA The main objective of this study is to compare and contrast the performance of two different sentiment analysis models – VADER NLTK and ROBERTA – on customer reviews and product ratings on the Ozon platform. VADER NLTK is selected for its simplicity and effi ciency in sentiment analysis, and ROBERTA, a transformer-based model, is selected for its state-of-the-art capabilities in NLP tasks. By assessing the accuracy and effi ciency of these models, this study aims to provide empirical insight into their strengths and weaknesses, shedding light on their potential usefulness for businesses seeking to leverage sentiment information from textual content. Keywords: natural language processing, sentiment analysis, VADER, ROBERTA, machine learning. DOI: 10.25791/pribor.11.2023.1451 Pp. 10-13.