Learning Semi-Naive Bayesian Classifiers from Numeric Data

  • Yang, Ying (Primary Chief Investigator (PCI))

    Project: Research

    Project Details

    Project Description

    Fast and accurate classification is critical to many aspects of our society, such as doctors classify whether a patient is sick or banks classify whether a transaction is fraud. Semi-na ve Bayesian classifiers (SNB) are a cutting-edge technique that holds a great potential to offer fast and accurate classification. However, one obstacle to SNB s extensive deployment is that SNB is sub-optimal when applied to numeric data, such as blood pressure readings in clinics or share prices in stock markets. This project aims at optimizing SNB's accuracy and efficiency in face of numeric data. It will hence deliver fast and accurate classification technologies that are of immediate and substantial impact, including life saving and crime prevention.
    StatusFinished
    Effective start/end date1/01/0730/11/08

    Funding

    • Australian Research Council (ARC): AUD187,378.00