Human knowledge models: Learning applied knowledge from the data.

Artificial intelligence and machine learning have demonstrated remarkable results in science and applied work. However, present AI models, developed to be run on computers but used in human-driven applications, create a visible disconnect between AI forms of processing and human ways of discovering and using knowledge. In this work, we introduce a new concept of "Human Knowledge Models" (HKMs), designed to reproduce human computational abilities. Departing from a vast body of cognitive research, we formalized the definition of HKMs into a new form of machine learning. Then, by training the models with human processing capabilities, we learned human-like knowledge, that humans can not only understand, but also compute, modify, and apply. We used several datasets from different applied fields to demonstrate the advantages of HKMs, including their high predictive power and resistance to noise and overfitting. Our results proved that HKMs can efficiently mine knowledge directly from the data and can compete with complex AI models in explaining the main data patterns. As a result, our study reveals the great potential of HKMs, particularly in the decision-making applications where "black box" models cannot be accepted. Moreover, this improves our understanding of how well human decision-making, modeled by HKMs, can approach the ideal solutions in real-life problems.

Discovering structural alerts for mutagenicity using stable emerging molecular patterns.

This study is dedicated to the introduction of a novel method that automatically extracts potential structural alerts from a data set of molecules. These triggering structures can be further used for knowledge discovery and classification purposes. Computation of the structural alerts results from an implementation of a sophisticated workflow that integrates a graph mining tool guided by growth rate and stability. The growth rate is a well-established measurement of contrast between classes. Moreover, the extracted patterns correspond to formal concepts; the most robust patterns, named the stable emerging patterns (SEPs), can then be identified thanks to their stability, a new notion originating from the domain of formal concept analysis. All of these elements are explained in the paper from the point of view of computation. The method was applied to a molecular data set on mutagenicity. The experimental results demonstrate its efficiency: it automatically outputs a manageable number of structural patterns that are strongly related to mutagenicity. Moreover, a part of the resulting structures corresponds to already known structural alerts. Finally, an in-depth chemical analysis relying on these structures demonstrates how the method can initiate promising processes of chemical knowledge discovery.