iLog Update Presentation at the ARDUOUS Conference
The 8th International Workshop on Annotation of useR Data for UbiquitOUs Systems (ARDUOUS 2024) focused on the role of user data, scoring, and ground truth data in machine learning and AI systems. Besides, the discussion was also dedicated to regulatory challenges, model bias, and security concerns. The conference workshop aims to bridge academic research with practical policy solutions, addressing transparency and accountability in AI systems.
DataScientia had the chance to present the latest update on iLog, and Ivan Kayongo presented the paper ‘A Methodology and System for Big-Thick Data Collection,’ co-authored by Haonan Zhao, Leonardo Malcotti, and Fausto Giunchiglia, highlighting their collaborative work on improving data collection methodologies for AI systems.
This article describes a new system that improves on iLog by collecting high-quality Big-Thick data. The system integrates human input with machine learning capabilities, allowing it to gather objective (Big Data) from sensors and subjective (Thick Data) through user feedback.
Key advancements over iLog include adaptive scheduling, where the system learns optimal times to ask context-specific questions, minimizing participant disruptions. The system also features re-planning options for participants and a comprehensive dashboard for visualizing real-time data. It provides a deeper, more nuanced view of human behavior by understanding participants’ personal contexts through sensor data and context questions.
This approach ensures continuous monitoring, accuracy, and relevance in data collection, making the system valuable for understanding user needs and preferences. By promoting knowledge exchange, DataScientia drives the advancement of AI applications through the continuous improvement of systems like iLog, which aim to address societal needs.
Here below the paper’s abstract:
Pervasive sensors have become essential in research for gathering real-world data. However, current studies often focus solely on objective data, neglecting subjective human contributions. We introduce an approach and system for collecting big-thick data, combining extensive sensor data (big data) with qualitative human feedback (thick data).
This fusion enables effective collaboration between humans and machines, allowing machine learning to benefit from human behavior and interpretations. Emphasizing data quality, our system incorporates continuous monitoring and adaptive learning mechanisms to optimize data collection timing and context, ensuring relevance, accuracy, and reliability. The system comprises three key components: a) a tool for collecting sensor data and user feedback, b) components for experiment planning and execution monitoring, and c) a machine-learning component that enhances human-machine interaction.
The full paper is available here.
Keywords: pervasive sensors, big data, thick data, machine learning, human-machine interaction, data quality, adaptive learning, experiment planning, context awareness.
────────────────
Tetiana Bihun
Author, Content Creator
Tag:data collection, ilog, LiveData, presentation, research, tools
