Classification Performance of Online Social Media Indian Airlines Reliable Dataset Based on Sentiment Analysis
In India there are many airline services which provide different services to the passengers for different locations and Customer’s experience is one of the important concerns for airline service providers. This research focus is on analysis of tweets related to airlines based in south regions of India. There are many social media trip related websites such as tripadvisior, skytrax, mouthshut and Trustpiolet where flight travelers share their feedbacks in the form of opinions or comments. This study presents a machine learning approach to analyze the comments or tweets to improve the customer’s experience and also Airline Services for customer satisfaction. Features were extracted from the comments using Snowball Stemmer with Word Tokinizer approach. Further, Random Forest and several Random Tree architectures were considered to develop classification model that maps the tweet into five different classes or categories such as positive, very positive, negative, very negative and neutral. The results show different heights accuracies given by Random Forest and Random Tree for respective five airline service providers such as AIR ASIA INDIA (99.47%), Go Air (97.13%), Indigo (99.47%), VISTARA (99.89%), and Spice Jet (99.38%) that certainly helps the airline industries to improve their customer’s experience.