Automating E-Government Services Using Artificial Intelligence for Sentiment Analysis and Digital Recognition

Abstract

The rapid evolution of digital governance has necessitated intelligent systems capable of understanding citizen feedback and automating service delivery. This project presents an AI-driven framework for automating e-government services by integrating handwritten digit recognition, text-based sentiment analysis, and facial expression-based emotion detection into a unified platform. The proposed system leverages deep learning techniques, particularly Convolutional Neural Networks (CNNs), along with classical machine learning approaches to analyze structured and unstructured citizen inputs effectively.The system is designed as a graphical user interface (GUI) application using Python’s Tkinter library, enabling users to interact seamlessly with government service modules. It incorporates three major components: digit recognition, opinion sentiment analysis, and facial emotion detection. The digit recognition module uses a CNN model trained on grayscale images to classify handwritten digits, supporting automated data entry and verification in government records. The sentiment analysis module processes textual opinions submitted by users and classifies them into positive or negative categories using a pre-trained machine learning model enhanced with natural language preprocessing techniques such as stemming.In addition to textual analysis, the system incorporates facial expression recognition using a deep learning model trained on emotional datasets. By detecting facial regions using Haar cascade classifiers and applying a CNN-based emotion classifier, the system identifies emotions such as happiness, sadness, anger, and neutrality. This multimodal approach ensures comprehensive sentiment understanding by combining textual and visual cues.The system stores user feedback and images, enabling policymakers to review aggregated sentiments and make informed decisions. The integration of multiple AI models into a single application demonstrates the feasibility of intelligent automation in e-governance systems. Furthermore, the platform emphasizes user accessibility and real-time processing, making it suitable for large-scale deployment in public service environments. Overall, this project contributes to the advancement of smart governance by providing a scalable and efficient solution for analyzing public opinion and automating administrative processes. It highlights the potential of AI in bridging the gap between citizens and government by enhancing transparency, responsiveness, and decision-making efficiency.