Data Mining in Mental Health

A.G. Eapen, K. Ponnambalam, J.F. Arocha, R. Shioda, T.F. Smith, J. Poss, and J. Hirdes (Canada)


Data mining, mental health, decision assist, classification, clustering, case mix


Data mining is the process of automatic classification of cases based on data patterns obtained from a dataset. A number of algorithms have been developed and implemented to extract information and discover knowledge patterns that may be useful for decision support. Once these patterns are extracted they can be used for automatic classification of case mixes. Although research has been conducted using diverse algorithms on small datasets (of about ten to fifteen attributes), in this paper, for the first time, we make use of a large database, namely, the interRAI Minimum Data Set for Mental Health (MDS-MH) containing over 455 attributes. Several data mining algorithms were used to classify clinical cases using MDS-MH data. The algorithms included Bayes methods, association rules, decision trees, and clustering. We compare these algorithms in various classification tasks and identify advantages and disadvantages of each algorithm with respect to their accuracy and use.

Important Links:

Go Back