Dictionary Vectorized Hashing of Emotional Recognition of Text in Mutual Conversation

Emotion detection is a subset of sentiment classification that interacts with emotion processing and analysis. The condition of just being emotional is frequently associated with making sensible qualitative stimulation of feelings or with environmental influence. With increase in the social media usage, people tend to have frequent conversation through several applications. Even police department tend to analyze the victim of any suicidal cases through the personal chat conversation. Machine learning could be used to analyze emotional detection of the person through text processing of their personal conversation. The text conversation dataset with 7480 conversations from KAGGLE warehouse and is used in the execution analysis to detect the emotional analysis. The text conversation dataset is preprocessed by removing the stop words. The tokens are extracted from the text using NGram method. The emotional labels are assigned for the tokens and the machine is trained to identify the emotions during testing. The emotional labels are converted into features to form corpus text for classifying the emotions in the conversation. The corpus is splitted to form training and testing dataset and is fitted to Dictionary Vectorizer, Feature Hashing, Count Vectorizer and Hash Vectorizer to extract the important features from the text conversation. The extracted features from the text conversation is the subjected to all the classifiers to analyze the performance of the emotion prediction. The scripting is written in Python and implemented with Spyder in Anaconda Navigator IDE, and the experimental results shows that the random forest classifier with dictionary vectorizer is exhibiting 99.8% of accuracy towards predicting the emotions from the personal conversations.