This research utilizes machine learning algorithms to categorize crime reports, aiming to identify the relevant sections of the Indian Penal Code (IPC) associated with reported crimes. They developed a comprehensive dictionary of words related to various crime categories and analyzed crime reports to compile relevant IPC sections and terminology. By employing machine learning techniques, they classified the dataset based on the presence of these terms in the reports, facilitating efficient and accurate classification of reported incidents according to legal standards. In this paper, different machine learning approaches are compared to predict the applicability of particular IPC sections to crime reports. The results of applying various machine-learning algorithms to our dataset have provided valuable insights into their performance. Among the algorithms tested, Naive Bayes and Support Vector Machine emerged as the top performers, both achieving an impressive accuracy score of 0.981481. This indicates their efficacy in classifying the data accurately. Right behind it, Random Forest and Logistic Regression demonstrated competitive performance with an accuracy score of 0.962963. K-Nearest Neighbors also demonstrated respectable accuracy at 0.944444. However, it is worth noting that the Decision Tree classifier lagged behind the others, registering the lowest accuracy of 0.888889.