arXiv subjects

Óscar Corcho

Publications and source records attributed to Óscar Corcho.

4 recordsLinked to original sources

arXiv2021

Interpretable Machine Learning Models for Predicting and Explaining Vehicle Fuel Consumption Anomalies

Identifying anomalies in the fuel consumption of the vehicles of a fleet is a crucial aspect for optimizing consumption and reduce costs. However, this information alone is insufficient, since fleet operators need to know the causes behind anomalous fuel consumption. We combine unsupervised anomaly detection techniques, domain knowledge and interpretable Machine Learning models for explaining potential causes of abnormal fuel consumption in terms of feature relevance. The explanations are used for generating recommendations about fuel optimization, that are adjusted according to two different user profiles: fleet managers and fleet operators. Results are evaluated over real-world data from telematics devices connected to diesel and petrol vehicles from different types of industrial fleets. We measure the proposal regarding model performance, and using Explainable AI metrics that compare the explanations in terms of representativeness, fidelity, stability, contrastiveness and consistency with apriori beliefs. The potential fuel reductions that can be achieved is round 35%.

cs.LG↗

arXiv2021

Vehicle Fuel Optimization Under Real-World Driving Conditions: An Explainable Artificial Intelligence Approach

Fuel optimization of diesel and petrol vehicles within industrial fleets is critical for mitigating costs and reducing emissions. This objective is achievable by acting on fuel-related factors, such as the driving behaviour style. In this study, we developed an Explainable Boosting Machine (EBM) model to predict fuel consumption of different types of industrial vehicles, using real-world data collected from 2020 to 2021. This Machine Learning model also explains the relationship between the input factors and fuel consumption, quantifying the individual contribution of each one of them. The explanations provided by the model are compared with domain knowledge in order to see if they are aligned. The results show that the 70% of the categories associated to the fuel-factors are similar to the previous literature. With the EBM algorithm, we estimate that optimizing driving behaviour decreases fuel consumption between 12% and 15% in a large fleet (more than 1000 vehicles).

cs.AI↗

arXiv2021

Rule Extraction in Unsupervised Anomaly Detection for Model Explainability: Application to OneClass SVM

OneClass SVM is a popular method for unsupervised anomaly detection. As many other methods, it suffers from the black box problem: it is difficult to justify, in an intuitive and simple manner, why the decision frontier is identifying data points as anomalous or non anomalous. Such type of problem is being widely addressed for supervised models. However, it is still an uncharted area for unsupervised learning. In this paper, we evaluate several rule extraction techniques over OneClass SVM models, as well as present alternative designs for some of those algorithms. Together with that, we propose algorithms to compute metrics related with eXplainable Artificial Intelligence (XAI) regarding the "comprehensibility", "representativeness", "stability" and "diversity" of the extracted rules. We evaluate our proposals with different datasets, including real-world data coming from industry. With this, our proposal contributes to extend XAI techniques to unsupervised machine learning models.

cs.LG↗

arXiv2014

The Research Object Suite of Ontologies: Sharing and Exchanging Research Data and Methods on the Open Web

Research in life sciences is increasingly being conducted in a digital and online environment. In particular, life scientists have been pioneers in embracing new computational tools to conduct their investigations. To support the sharing of digital objects produced during such research investigations, we have witnessed in the last few years the emergence of specialized repositories, e.g., DataVerse and FigShare. Such repositories provide users with the means to share and publish datasets that were used or generated in research investigations. While these repositories have proven their usefulness, interpreting and reusing evidence for most research results is a challenging task. Additional contextual descriptions are needed to understand how those results were generated and/or the circumstances under which they were concluded. Because of this, scientists are calling for models that go beyond the publication of datasets to systematically capture the life cycle of scientific investigations and provide a single entry point to access the information about the hypothesis investigated, the datasets used, the experiments carried out, the results of the experiments, the people involved in the research, etc. In this paper we present the Research Object (RO) suite of ontologies, which provide a structured container to encapsulate research data and methods along with essential metadata descriptions. Research Objects are portable units that enable the sharing, preservation, interpretation and reuse of research investigation results. The ontologies we present have been designed in the light of requirements that we gathered from life scientists. They have been built upon existing popular vocabularies to facilitate interoperability. Furthermore, we have developed tools to support the creation and sharing of Research Objects, thereby promoting and facilitating their adoption.

cs.DL↗