2023-10-21, 15:10–15:40 (Europe/Athens), Α115-117
The Recommender Metrics Framework (RMF) is an open-source software that monitors, analyses, and evaluates recommendation mechanisms. The European Open Science Cloud (EOSC) uses a modern recommender System (RS) in the EOSC Marketplace for suggesting various resources. Utilizing RMF for gauging the effectiveness of the EOSC RS is essential for acquiring valuable insights into numerous factors that impact the user experience. The inclusion of supplementary diagnostic metrics and visualizations provides in-depth and occasionally unexpected perspectives into the model's performance. The RMF is currently developed in the National Infrastructures for Research and Technology (GRNET) and used as an open-source solution within the EOSC-Future project. The evaluation is quantitatively performed by processing information such as resources, user actions, and recommendations to measure the impact of the AI-enhanced services and user satisfaction as well as to incorporate this feedback and improve the services provided, via a user-friendly Application Programming Interface (API) and a User Interface (UI) dashboard. The framework supports both real-time and offline ingestion of data, multiple resource types, and recommendation engines as sources. The software is responsible for processing the collected data, computing the designated evaluation metrics, and producing the necessary information in a homogenized manner. The RESTful API along with the rich UI dashboard presents reports as a web service and visualizes statistics, metrics and Key Performance Indicators (KPI)s. The RS evaluation framework continually evolves, incorporating additional features and utilities, to foster the development of more reliable and high-quality RS designs.
Nikolaos Triantafyllis, PhD candidate, (male) holds the position of HPC high-level Support Engineer in GRNET. Currently, he is involved in the Copernicus - eoSC AnaLytics Engine (C-SCALE) and EOSC-Future EU research projects. Before that, he was engaged in several EU and national research projects such as the European Open Science Cloud (EOSC) under the National Observatory of Athens (NOA). He is the main developer of an HPC-oriented application for automatic moment tensor retrieval in real-time, operationally in use by the Institute of Geodynamics (GI) of NOA, which received an award from the EGU General Assembly 2021. Until lately, he held the position of the Technical Commission at the European Integrated Data Archive (EIDA) in GI-NOA, while in the past, he participated in software development projects on security-bound applications at the Cyber Defense Department of the Hellenic National Defense General Staff. He graduated from the Computer Engineering and Informatics Department (CEID) of the School of Engineering at the University of Patras in 2012, and in 2014 he received his MSc in Computer Science and Engineering from the same department. At the moment, he is working on his PhD on HPC Job Scheduling optimizations at the Computing Systems Laboratory (CSLAB) of the School of Electrical and Computer Engineering (ECE) of the National Technical University of Athens (NTUA).