AQCLab has participated in QUACTIC 2024, 17th International Conference on the Quality of Information and Communications Technology, presenting the paper "An Environment for the Assessment of the Functional Suitability of AI Systems". QUATIC 2024 took place in Pisa (Italy) from 11-13 September, serving as a forum for practitioners and researchers on the dissemination of advanced methods, techniques and tools to support quality approaches in ICT engineering and management.
In this work, our colleagues Jesús Oviedo, Moisés Rodríguez and Mario Piattini propose an environment for measuring and evaluating the functional suitability of Artificial Intelligence systems. This is a research work carried out by AQCLab in collaboration with the Alarcos group in which they propose: a model with subcharacteristics, properties and metrics to evaluate functional suitability aligned with the ISO/IEC 25059 standard; an evaluation process aligned with ISO/IEC 25040; and a technological environment with tools to support the execution of the evaluation.
Our colleagues Jesús Oviedo, Moisés Rodríguez and Mario Piattini, in collaboration with Andrea Trenta, Dino Cannas and Domenico Natale from Data Quality Lab and UNINFO in Italy, have published the article ISO/IEC quality standards for AI engineering in volume 54 of the journal Computer Science Review.
This article provides an extensive review of the various existing international standards, with a special focus on those defined by ISO/IEC, related to the quality of Artificial Intelligence systems. Thus, standards are addressed in different dimensions of quality for this type of systems: the quality of the software processes used in their development, the quality of the software products themselves, and the quality of the data of the applications that integrate AI systems. The article also describes the different existing proposals for assessing the quality of AI systems in relation to the above dimensions.
AQCLab has participated in the XXVII Ibero-American Conference on Software Engineering (CIbSE 2024), the leading Software Engineering research forum in Ibero-America, which took place in Curitiba (Brazil) between 6 and 10 May 2024. AQCLab presented two research papers regarding quality in AI systems and hybrid (classical-quantum) systems:
Our colleagues Ana Díaz, Moisés Rodríguez and Mario Piattini have published the article Implementing an Environment for Hybrid Software Evaluation in the journal Science of Computer Programming.
This paper presents the first technological environment for measuring and evaluating the analyzability of hybrid software. As quantum software gains ground and relevance in various domains, it is essential to address the evaluation of hybrid systems that combine classical and quantum elements to ensure diverse quality characteristics.
Thus, the environment proposed in this paper stablishes a model, metrics and tools applicable in the field of quantum software to assess its analyzability, one of the subcharacteristics of maintainability defined in ISO/IEC 25000.
Furthermore, real-world examples of hybrid software are provided to showcase the functionality of the different tools in the environment, yielding readable and representative results for the evaluator.
Our colleagues Javier Verdugo, Jesús Oviedo, Moisés Rodríguez and Mario Piattini have published the article Connecting research and practice for software product quality evaluation and certification: a software laboratory's 25-year journey in the Volume: 41 Issue: 3 May-June-2024 of the IEEE Software journal, in which they present the trajectory of AQCLab, a laboratory for software quality evaluation, over 25 years, starting from the research work of the Alarcos Research Group.
The article details the challenges and difficulties involved in establishing the first accredited laboratory for software quality assessment based on the ISO/IEC 25000 family of international standards. Thus, AQCLab has meant the transfer of the research on software quality carried out over the years to the industry, allowing organisations that develop or acquire software to evaluate their products, and even obtain a certificate based on the results of these laboratory evaluations.
In addition to the challenges and lessons learned during this journey in software product evaluation and certification, the article provides details on the quality models used in the laboratory for the evaluation of Maintainability and Functional Suitability.
+34 676 97 57 83
