The utilization of machine learning techniques in earthquake prediction is promising due to their ability to uncover hidden patterns in data that may not be easily identifiable through traditional methods. Time series analysis, specifically the ARIMA model, is a commonly used approach in predicting earthquake occurrences, as it considers the temporal dependencies of the data. The random forest algorithm, on the other hand, is a popular supervised learning algorithm that can handle both categorical and numerical data and has the potential to capture complex interactions between variables. The proposed research aims to develop a machine learning-based earthquake prediction model using time series analysis and the random forest algorithm. The data for the study has been acquired from the United States Geological Survey and has undergone exploratory data analysis (EDA) techniques to obtain key observations. The data’s stationarity has been analyzed, and the auto-regressive integrated moving average (ARIMA) model has been identified as suitable for time series prediction. Additionally, the random forest algorithm has been implemented as another prediction model. A comparative study has been conducted to determine the effectiveness of both models. The proposed models have potential applications in earthquake prediction, which can have significant societal benefits. Early prediction of earthquakes can help save lives by providing an advanced warning to affected areas, allowing people to evacuate and take precautionary measures. Additionally, civil engineers can use these predictions to design structures that are more resilient to seismic activity, reducing the risk of structural failures and damage. The comparative study of the ARIMA, linear regression, and random forest models will provide insights into the effectiveness of each approach in earthquake prediction and can contribute to the existing body of research in this field.