Mental illness is a set of health problems that affect the way individuals perceive themselves, relate to others, and interact with the world around them. Due to the myriad of underlying causes and subsequent effects of mental illness, these conditions often trigger fear and misunderstanding among the general population. Common mental illnesses such as depression and anxiety disorders often affect an individual's thoughts, feelings, abilities, and behaviours. Anxiety disorder is characterized by an irrational fear of certain things or events. It is often attributed as the feeling of worry about anticipated events and fear in response to current events. This work has identified several related research efforts on the general well-being and psychological distress using data mining. However, there is inadequate research done using a similar method on specific mental health issues, especially related to generalized anxiety disorder (GAD). In view of this gap, this study focuses on implementing a novel feature selection and data mining classifier system. Under the proposed method, Shapley value will be implemented as the feature selection of the data mining classifier on the mental health data. The approach is used to predict GAD among women. The methodology for this research is adapted from the process of Knowledge Discovery in Databases (KDD). This methodology consists of 5 main phases; namely data acquisition, data pre-processing, feature selection, classification prediction, and evaluation. Using this enhanced prediction algorithm, any women can get help if they are perceived to be suffering from GAD. By designing an effective way of identifying individuals who may be suffering from mental illnesses, we hope that our work would improve the awareness surrounding mental health issues especially among women and enable them to undertake autonomous decision in seeking mental health services.