Published Paper Details:

When it comes to researching a service provider's services, facilities, and market standing, sentiment analysis is a great resource due to the abundance of customer evaluations, opinions, and feelings that are accessible online. When it comes to connecting passengers with airlines, social media channels are crucial. As a means of measuring service quality and operational efficacy, evaluations and feedback from customers are becoming more important to airlines. NLP and other machine learning technologies have come a long way in the last several years, greatly easing the burden of huge data analysis. This research delves into the sentiment analysis of airline customer evaluations utilizing a range of ML models and NLP approaches, such as NB, LSTM, RF, and CNN. The dataset, sourced from Kaggle, comprises predominantly negative reviews, with neutral and other sentiments. Data preprocessing steps such as normalization, stop words removal, lemmatization, and stemming were employed to enhance model performance. The analysis revealed that the CNN model outperformed other models with a 97%accuracy, 97%precision, 96%recall, and a 96%F1-score, demonstrating its superior capability in comparison of the RF, LSTM, and NB models. This research highlights the effectiveness of CNNs in sentiment analysis of airline reviews and provides valuable insights for improving customer service based on automated sentiment detection