WHATSAPP CHAT ANALYZER USING MACHINE LEARNING

WhatsApp has emerged as one of the most widely adopted and efficient platforms for digital communication, facilitating a broad spectrum of interactions within individual and group conversations. These chat logs encompass diverse topics and serve as a rich source of unstructured textual data, which holds substantial value for contemporary technologies such as machine learning (ML) and natural language processing (NLP). For machine learning models to perform optimally, the availability of relevant and high-quality training data is paramount, as the learning outcomes are directly influenced by the input data. The objective of this research is to develop a tool capable of conducting comprehensive exploratory data analysis (EDA) on WhatsApp chat data. The proposed system is agnostic to the subject matter of the conversation and can be uniformly applied to extract actionable insights from any chat dataset. The implementation leverages widely-usedPython libraries including Pandas, Matplotlib, Seaborn, and NLTK for sentiment analysis. These tools facilitate the transformation of raw textual data into structured formats (e.g., DataFrames), enable statistical visualization, and support sentiment classification. The analytical results are rendered via a Flutter-based front-end application, chosen for its lightweight architecture and efficient resource consumption, thus enabling scalability and applicability to large-scale datasets.