In a world increasingly influenced by online activity, analyzing search behaviour has become an essential tool for understanding public interest and predicting emerging trends. This project, titled “Google Search Analysis Using Machine Learning,” aims to forecast keyword popularity over time using data sourced from Google Trends. During Phase 1, our team focused on conceptualizing the approach, finalizing the methodology, and outlining the system architecture. The planned system begins by defining the scope and selecting appropriate keywords. After that, it links to the Google Trends API to get search data that is useful. This data is cleaned up and pre-processed to make sure it is consistent and of good quality. Next, trend analysis is performed to uncover underlying patterns and seasonal behaviour. Using this processed data, machine learning models— such as ARIMA or LSTM—will be trained to make future predictions. The output is then visualized using charts and confidence intervals to offer intuitive insights. Finally, the system will generate a comprehensive report summarizing the findings and providing actionable recommendations. This structured workflow lays the foundation for an intelligent, automated tool capable of tracking and forecasting online keyword trends, thereby assisting in data-driven decision-making
Sanghvi et al. (Tue,) studied this question.