id: 05596151
dt: a
an: 05596151
au: Shaari, Faizah; Bakar, Azuraliza Abu; Hamdan, Abdul Razak
ti: A predictive analysis on medical data based on outlier detection method
    using non-reduct computation.
so: Huang, Ronghuai (ed.) et al., Advanced data mining and applications. 5th
    international conference, ADMA 2009, Beijing, China, August 17‒19,
    2009. Proceedings. Berlin: Springer (ISBN 978-3-642-03347-6/pbk).
    Lecture Notes in Computer Science 5678. Lecture Notes in Artificial
    Intelligence, 603-610 (2009).
py: 2009
pu: Berlin: Springer
la: EN
cc: 
ut: non-reduct; abnormal; outliers; non-interesting; redundant; superfluous;
    rare
ci: 
li: doi:10.1007/978-3-642-03348-3_62
ab: Summary: In this research, a new method to predict and diagnose medical
    dataset is discovered based on outlier mining method using Rough Sets
    Theory (RST). The RST is used to generate medical rules, while outliers
    are detected from the rules to diagnose the abnormal data. In detecting
    outliers, a computation of set of attributes or known as Non-Reduct is
    proposed by proposing two new formula of Indiscernibility Matrix Modula
    (iDMM D) and Indiscernibility Function Modulo (iDMFM D) based on RST.
    The results show that the proposed method is a fast detection method
    with lower detection rate. In conclusion, the computation of the
    Non-Reduct is expected to give medical knowledge that able to predict
    abnormality in dataset that could be used in medical analysis.
rv: