Analyzing customer reviews for ARBA Travel using sentiment analysis

Sentiment analysis is a natural language processing (NLP) that categorizes customer feedback into as positive, negative, or neutral. The project entails gathering a dataset of customer reviews from Google Reviews and Facebook, cleaning the text to eliminate any noise, and analyzing sentiments using three machine learning algorithms; Naive Bayes, Support Vector Machine, and Logistic Regression. Key phrases are extracted from 400 customer reviews on Google Reviews and 300 from Facebook to reveal common themes in customer experiences. Data visualization techniques are used to present trends, such as the frequency of positive and negative reviews over time and the most commonly mentioned aspects of ARBA Travel's services. The findings offer actionable recommendations for ARBA Travel to improve its offerings, address issues, and reinforce strengths. The integration of sentiment analysis into the feedback evaluation process at ARBA Travel will enhance its decision-making processes, refinement in customer engagement approaches, and sustain competitiveness within the travel industry. This research illustrates the potential of sentiment analysis to convert unstructured customers' feedback into meaningful insights that drive improved business results and stronger customer relationships. Three machine learning algorithms which are Naive Bayes, Logistic Regression, and Support Vector Machine, were implemented and evaluated using cross-validation and performance metrics such as accuracy, precision, recall, and F1- score. Among these, Naive Bayes achieved the highest performance with an accuracy of 93.67% and an F1-score of 93.54%, making it the most effective model for sentiment classification in this context. The results were visualized using an interactive dashboard developed in Power BI, allowing users to explore sentiment trends, keyword frequency, and review distributions by gender, platform, and time. Future enhancements include expanding the analysis to other platforms like TripAdvisor and Instagram for broader sentiment coverage. Implementing real-time sentiment analysis would enable faster responses to customer feedback. Additionally, enhancing the dashboard with features like trend forecasting, customer segmentation, and automated alerts can support more proactive and data-driven decision-making for ARBA Travel.