Sentiment Analysis of Handwritten and Text Statement for Emotion Classification using Intelligent Techniques: A Novel Approach

Traditionally, many people still wish to write on pen and paper. However, it has some drawbacks, like accessing and storing physical documents efficiently, searching through them, and sharing them efficiently. Handwriting to Text Conversion (HTC) classifies and converts an individual’s handwriting into digital form. However, HTC removes all the mentioned problems as storing, retrieving, and using the text as and when required is easier. Emotions are a basic and particularly important aspect of one’s life. To understand this important aspect of an individual’s life, we must detect emotions using affect data like text, voice, and image. We have used text as the effect data for this work. We can find a person’s emotions behind his text by sentiment analysis. Sentiment recognition and analysis is a topic with wide research as many brands, companies, and even famous personalities are very much interested in getting feedback and thus do the evaluation of their performance and knowing what people think about them around the world. Authors have proposed a model where they collected data from social media reviews and classified it into three broad categories, which are positive, negative, and neutral. Find the emotions category viz. happiness, sadness, shame, anger, disgust, fear, surprise, or neutral from three types of classified sentiments. The proposed model combined machine learning, deep learning, and natural language processing techniques to achieve the best outcome.