Abstract
New study results on artificial intell igence have been published. According to news reporting originating from Dublin, Ireland, by NewsRx correspondents, research stated, "Time series classification is a challenging research area where machine learning and deep learning techniq ues have shown remarkable performance. However, often, these are seen as black b oxes due to their minimal interpretability." Our news reporters obtained a quote from the research from Technological Univers ity Dublin: "On the one hand, there is a plethora of eXplainable AI (XAI) method s designed to elucidate the functioning of models trained on image and tabular d ata. On the other hand, adapting these methods to explain deep learning-based ti me series classifiers may not be straightforward due to the temporal nature of t ime series data. This research proposes a novel global post-hoc explainable meth od for unearthing the key time steps behind the inferences made by deep learning -based time series classifiers. This novel approach generates a decision tree gr aph, a specific set of rules, that can be seen as explanations, potentially enha ncing interpretability. The methodology involves two major phases: (1) training and evaluating deeplearning- based time series classification models, and (2) ex tracting parameterized primitive events, such as increasing, decreasing, local m ax and local min, from each instance of the evaluation set and clustering such e vents to extract prototypical ones. These prototypical primitive events are then used as input to a decision-tree classifier trained to fit the model prediction s of the test set rather than the ground truth data."
NETL
NSTL