Search

This comprehensive review of concept-supported interpretation methods in Explainable Artificial Intelligence (XAI) navigates the multifaceted landscape. As machine learning models become more complex, there is a greater need for interpretation methods that deconstruct their decision-making processes. Traditional interpretation techniques frequently emphasise lower-level attributes, resulting in a schism between complex algorithms and human cognition. To bridge this gap, our research focuses on concept-supported XAI, a new line of research in XAI that emphasises higher-level attributes or 'concepts' that are more aligned with end-user understanding and needs. We provide a thorough examination of over twenty-five seminal works, highlighting their respective strengths and weaknesses. A comprehensive list of available concept datasets, as opposed to training datasets, is presented, along with a discussion of sufficiency metrics and the importance of robust evaluation methods. In addition, we identify six key factors that influence the efficacy of concept-supported interpretation: network architecture, network settings, training protocols, concept datasets, the presence of confounding attributes, and standardised evaluation methodology. We also investigate the robustness of these concept-supported methods, emphasising their potential to significantly advance the field by addressing issues like misgeneralization, information overload, trustworthiness, effective human-AI communication, and ethical concerns. The paper concludes with an exploration of open challenges such as the development of automatic concept discovery methods, strategies for expert-AI integration, optimising primary and concept model settings, managing confounding attributes, and designing efficient evaluation processes. [ABSTRACT FROM AUTHOR]

Payment cards offer a simple and convenient method for making purchases. Owing to the increase in the usage of payment cards, especially in online purchases, fraud cases are on the rise. The rise creates financial risk and uncertainty, as in the commercial sector, it incurs billions of losses each year. However, real transaction records that can facilitate the development of effective predictive models for fraud detection are difficult to obtain, mainly because of issues related to confidentially of customer information. In this paper, we apply a total of 13 statistical and machine learning models for payment card fraud detection using both publicly available and real transaction records. The results from both original features and aggregated features are analyzed and compared. A statistical hypothesis test is conducted to evaluate whether the aggregated features identified by a genetic algorithm can offer a better discriminative power, as compared with the original features, in fraud detection. The outcomes positively ascertain the effectiveness of using aggregated features for undertaking real-world payment card fraud detection problems. [ABSTRACT FROM AUTHOR]

In this paper, a hybrid model consisting of the fuzzy ARTMAP (FAM) neural network and the classification and regression tree (CART) is formulated. FAM is useful for tackling the stability-plasticity dilemma pertaining to data-based learning systems, while CART is useful for depicting its learned knowledge explicitly in a tree structure. By combining the benefits of both models, FAM-CART is capable of learning data samples stably and, at the same time, explaining its predictions with a set of decision rules. In other words, FAM-CART possesses two important properties of an intelligent system, i.e., learning in a stable manner (by overcoming the stability-plasticity dilemma) and extracting useful explanatory rules (by overcoming the opaqueness issue). To evaluate the usefulness of FAM-CART, six benchmark medical data sets from the UCI repository of machine learning and a real-world medical data classification problem are used for evaluation. For performance comparison, a number of performance metrics which include accuracy, specificity, sensitivity, and the area under the receiver operation characteristic curve are computed. The results are quantified with statistical indicators and compared with those reported in the literature. The outcomes positively indicate that FAM-CART is effective for undertaking data classification tasks. In addition to producing good results, it provides justifications of the predictions in the form of a decision tree so that domain users can easily understand the predictions, therefore making it a useful decision support tool. [ABSTRACT FROM AUTHOR]