Sentiment Analysis using Cuckoo Search and Computational Intelligence
Gurpreet Kaur, Er.Poonam Kukana · 2020 International Conference on Smart Electronics and Communication (ICOSEC) · 2020
Sentiment analysis (SA) is a computational study focusing on analyzing an individual's attitude and beliefs based on textual information. Presently, it is considered to be the most popular field in business analytics reflecting customer's opinions by reviewing feedback on their products This unstructured data is collected and pre-processed to transform it into a usable form which involves removal of normalization, punctuation removal, and stop-word removal. It is followed by the tokenization of the text reflecting the feature extraction stage in the proposed methodology. Nature-inspired Cuckoo Search (CS) is implemented to optimize the feature data that is fed to Feed Forward Back Propagation Neural Network (FFBPNN) for classification of sentiment data to reflect positive, negative, and neutral reviews The performance of the proposed SA work is evaluated in terms of precision, recall, f-measure, and classification accuracy.