Environmental Protection Department
Consumer Council
In recent years, Artificial Intelligence (AI) have been applied in various aspects of human life, with touch on different types of data. Explainable Artificial Intelligence (XAI) is an emerging area of research in AI. XAI can explain how AI obtained a particular answer and why it is so. Traditional AI solution usually ends when a prediction or classification is made. But for many applications, users are required to communicate with others with the AI-generated answer. It is important to help explain how and why an answer is obtained, gain user’s trust in AI solution and help organization implement AI solution with responsibility. In this project, our primary objective is to deliver an explainable AI solution platform which is needed to fully facilitate sophisticated work on text assessment and build user’s trust on it. We will research and develop a framework for explainable AI and decision visualization with domain expertise addressing specific industry. It is first trial of its kind in HK society for explainable AI on document assessment. After successful completion, we can help our partners enhance efficiency with trust in AI solution and encourage production use of AI.