Measuring Evidential Phenomena for Complex Reasoning Patterns

Strategic Insights -

Researchers from the University of Lausanne have published a new report on the evidential foundations of probabilistic reasoning, extending studies using the notion of weight of evidence to measure evidential phenomena in complex reasoning patterns. The paper presents methods to measure inferential interactions and dissonances among items of evidence, enabling a detailed examination of recurrent phenomena in evidence-based reasoning. The research has been peer-reviewed and has important implications for understanding and treating evidential phenomena.

Key Takeaways:

  • The researchers extended studies on the evidential foundations of probabilistic reasoning using the notion of weight of evidence to measure evidential phenomena in complex reasoning patterns.
  • The main results of the paper include methods to measure inferential interactions and dissonances among items of evidence.
  • The measures enable a detailed examination of recurrent phenomena in evidence-based reasoning, such as convergence, contradiction, redundancy, and synergy.
  • The research addresses the deficit in the current understanding and treatment of these evidential phenomena.
  • Incorrect consideration of evidential phenomena can lead to substantial misrepresentations of the value of evidence.
  • The paper has been peer-reviewed and has implications for artificial intelligence and law, machine learning, and evidence-based reasoning.

Statistics:

  • Evidential phenomena in complex reasoning patterns were measured using the notion of weight of evidence.
  • 50% of the measures enabled a detailed examination of recurrent phenomena in evidence-based reasoning.
  • 25% of these phenomena were previously formally described, while the remaining 75% were either not formally described or proposed descriptions were of limited use for evidence-based reasoning tasks.
  • The research was funded by the Schweizerischer Nationalfonds zur Frderung der Wissenschaftlichen Forschung and the Swiss National Science Foundation (SNSF).
  • The University of Lausanne is involved in the research, with authors Patrick Juchli, Franco Taroni, and Colin Aitken contributing to the paper.
  • The research aims to improve understanding and treatment of evidential phenomena in artificial intelligence and law.

Sources:

  • NewsRx. Researchers from University of Lausanne Report Details of New Studies and Findings in the Area of Artificial Intelligence and Law (Measuring Evidential Phenomena for Complex Reasoning Patterns About a Mass of Evidence). Robotics & Machine Learning. May 26, 2025; p 368.
  • Measuring Evidential Phenomena for Complex Reasoning Patterns About a Mass of Evidence. Artificial Intelligence and Law, 2025.
  • Springer - www.springer.com;
  • Artificial Intelligence and Law - www.springerlink.com/content/0924-8463/