This research investigates WhatsApp group chat analysis, focusing on sentiment detection and topic modeling using machine learning and natural language processing (NLP) algorithms. With the increasing use of instant messaging platforms for social and professional communication, understanding group dynamics is critical. The study employs Random Forest and XGBoost classifiers to classify sentiments into Positive, Negative, and Neutral categories while using BERTopic module for topic modeling to uncover prevalent themes in conversations. The results show that XGBoost achieved a higher classification accuracy of 92.8% compared to 88.6% for the Random Forest Classifier, effectively addressing the class imbalance in the dataset. Sentiment distribution revealed that most group chats were Neutral (45%), followed by Positive (35%) and Negative (20%). Topic modeling identified key themes, such as event planning, work-related collaboration, and casual social interactions. These findings highlight the effectiveness of machine learning and NLP techniques in extracting valuable insights from group chat data, with applications ranging from user behavior analysis to enhancing communication strategies on digital platforms.