Symbolic AI for XAI: Evaluating LFIT Inductive Programming for Explaining Biases in Machine Learning

Machine learning methods are growing in relevance for biometrics and personal information processing in domains such as forensics, e-health, recruitment, and e-learning. In these domains, white-box (human-readable) explanations of systems built on machine learning methods become crucial. Inductive logic programming (ILP) is a subfield of symbolic AI aimed to automatically learn declarative theories about the processing of data. Learning from interpretation transition (LFIT) is an ILP technique that can learn a propositional logic theory equivalent to a given black-box system (under certain conditions). The present work takes a first step to a general methodology to incorporate accurate declarative explanations to classic machine learning by checking the viability of LFIT in a specific AI application scenario: fair recruitment based on an automatic tool generated with machine learning methods for ranking Curricula Vitae that incorporates soft biometric information (gender and ethnicity). We show the expressiveness of LFIT for this specific problem and propose a scheme that can be applicable to other domains. In order to check the ability to cope with other domains no matter the machine learning paradigm used, we have done a preliminary test of the expressiveness of LFIT, feeding it with a real dataset about adult incomes taken from the US census, in which we consider the income level as a function of the rest of attributes to verify if LFIT can provide logical theory to support and explain to what extent higher incomes are biased by gender and ethnicity.

Implementation of bi-directional searching Chemical Reaction Optimization (CRO) algorithm towards achieving cutting tool indexing time minimization in Automatic Tool Changer(ATC)

Determining the optimum cutting tool positions in a tool magazine is a challenging task which involves a huge searching space. Reduction of manufacturing time even by nanoseconds will highly be beneficial for all process industries. The prime focus of this research work is to reduce the total turret or tool indexing time in Automatic Tool Changer (ATC) which in turn curtails the idle non-cutting time of the bi-directional turret or tool magazine in a CNC machine. This variant of optimization problem falls under the category of non-deterministic hard polynomial (NP-hard) problem. Chemical Reaction Optimization (CRO) algorithm is proposed to determine the near optimal/global optimal solution with lesser computational cost. Five benchmark case studies are considered in this work to assess the robustness of the proposed algorithm. From the results, it is evident that the proposed algorithm yields solutions which matches or exceeds with the solution quality.

SOURCE: a semi-automatic tool for spring-monitoring data analysis and aquifer characterisation

It has become increasingly necessary to optimise mountain groundwater resource management and comprehend resource-recharging systems from a hydrogeological perspective to formulate adequate resource protection strategies. Analysing mountain spring behaviour and aquifer characteristics can be time-consuming, so new automated techniques and software tools are needed to estimate hydrogeological parameters and understand the exhaustion dynamics of groundwater resources. This paper introduces SOURCE, a new semi-automatic tool that automates the hydrogeological characterisation of water springs and provides proper estimations of the vulnerability index, as well as autocorrelation and cross-correlation statistical coefficients. SOURCE rapidly processed input data from the Mascognaz 1 spring (Aosta Valley) water probes and meteorological station to provide graphical outputs and values for the main hydrodynamic parameters. Having a single software package that contains all the main methods of water spring analysis could potentially reduce analysis times from a few days to a few hours.

Towards Accessibility and Inclusion of Native Mobile Applications Available for Ecuador in Google Play Store

This article aims to evaluate the level of compliance with the accessibility requirements of the most popular native Android mobile applications, for which a sample of 50 Google Play Store applications available in Ecuador was taken. A five-phase method using the Accessibility Scanner tool was used to evaluate the apps. The results revealed that 47.5% are related to problems with tactile orientation, followed by the labeling of elements with 28.2%, and text contrast with 9.2%. The highest number of barriers found in the evaluation of mobile applications corresponds to the principle of operability with 53.9%. This study reveals that, although social networks are widely used, they have 28.7% of accessibility problems. Basing accessibility analysis exclusively on an automatic tool is very limited since it neither detects all errors nor are the errors they detect accurate. However, we suggest complementing the automatic review evaluations with a manual method based on heuristics to ensure an adequate level of accessibility in mobile apps. In addition, we recommend using this study as a starting point to create a software tool using WCAG 2.1 based on artificial intelligence algorithms to help developers evaluate accessibility in mobile apps.

Natural language processing tool for automatic diseases and drugs recognition from electronic health records in polish- pilot study

Background An estimate of 80% of data gathered in electronic health records is unstructured, textual information that cannot be utilized for research purposes until it is manually coded into a database. Manual coding is a both cost and time- consuming process. Natural language processing (NLP) techniques may be utilized for extraction of structured data from text. However, little is known about the accuracy of data obtained through these methods. Purpose To evaluate the possibility of employing NLP techniques in order to obtain data regarding risk factors needed for CHA2DS2VASc scale calculation and detection of antithrombotic medication prescribed in the population of atrial fibrillation (AF) patients of a cardiology ward. Methods An automatic tool for diseases and drugs recognition based on regular expressions rules was designed through cooperation of physicians and IT specialists. Records of 194 AF patients discharged from a cardiology ward were manually reviewed by a physician- annotator as a comparator for the automatic approach. Results Median CHA2DS2VASc score calculated by the automatic was 3 (IQR 2–4) versus 3 points (IQR 2–4) for the manual method (p=0.66). High agreement between CHA2DS2VASc scores calculated by both methods was present (Kendall's W=0.979; p<0.001). In terms of anticoagulant recognition, the automatic tool misqualified the drug prescribed in 4 cases. Conclusion NLP-based techniques are a promising tools for obtaining structured data for research purposes from electronic health records in polish. Tight cooperation of physicians and IT specialists is crucial for establishing accurate recognition patterns. Funding Acknowledgement Type of funding sources: None.

Parametric Investigation on the Effectiveness of FRM-Retrofitting in Masonry Buttressed Arches

Buttressed arches represent some of the most vulnerable elements in historical masonry buildings under seismic actions. Given their structural, architectural and often artistic importance, it is paramount to investigate effective retrofitting measures which satisfy reversibility and compatibility requirements for historical heritage. Among these, Fibre-Reinforced Mortar (FRM), characterised by enhanced ductility, has recently emerged in the scientific literature. In this paper, a numerical investigation aimed at exploring the effectiveness of this seismic retrofitting technique, when applied at intrados or extrados of various typologies of buttressed arches, is presented. An automatic tool for limit analysis is described and validated against a nonlinear Discrete Macro-Element modelling approach. This precedes an extensive parametric analysis, which has highlighted the effect of various geometrical features of the system on both collapse mechanism and maximum acceleration in the unreinforced configuration, and the high increase in seismic capacity provided by the retrofitting. This is particularly remarkable in case of local mechanism, i.e., wholly within the arch, in which case the FRM may be responsible of an overall shifting to a semi-global failure type.