Logistic Regression based Sentiment Analysis System: Rectify
Harsh Pratap Singh, Nagendra Singh, Anuprita Mishra, Santosh Kumar Sen, Mamta Swarnkar, Deepak Pandey
Abstract
The detection of various reactions using computer vision, machine learning, and artificial intelligence is a rapidly growing field of research. In this paper, we present a sentiment analysis model based on the Python, NLTK (natural language toolkit) libraries, and machine learning algorithms that can detect multiple reactions from the public on a single platform. The proposed Sentiment Analysis System deals with the very usual problem faced by different companies, manufacturers, and sellers about knowing the customer review for their product. The model classifies the public reviews and the normal text as positive, negative, or neutral. This proposed system is categorised into two parts, one being a field-specific sentiment analysis and the other being a generalised system that can judge any particular word or sentence to be either positive, negative, or neutral. The proposed model is also capable of representing a huge dataset (input as a csv file) in the form of a graph, which can be easily understood by the desired person, and the graphical representation is possible with the Matplotlib library of Python. The model's performance is evaluated using several metrics, such as precision, recall, and others. The model's accuracy and efficiency make it a promising tool for sentiment analysis, which can be used by anyone in any field. In conclusion, the system can work for any specific field provided the data set, and the generalized way can help out with random sentences, and the model’s potential can be boosted with some further research.