|
Title:
|
KNOWLEDGE-BASED AID FOR THERAPEUTIC INTERVENTION |
|
Author(s):
|
K.E. Burn-Thornton and T. Burman |
|
ISBN:
|
978-972-8939-09-0 |
|
Editors:
|
Miguel Baptista Nunes, Pedro Isaías and Philip Powell |
|
Year:
|
2010 |
|
Edition:
|
Single |
|
Keywords:
|
Decision-Support, Therapeutic Intervention, Addiction, Knowledge Extraction. |
|
Type:
|
Full Paper |
|
First Page:
|
129 |
|
Last Page:
|
136 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
We describe the first decision support tool (DST) for use by counsellors during support of alcohol addicts during therapeutic intervention. The DST is novel both in its domain of application and the engine which is used to provide the decision support. This Knowledge-Based Aid to Therapeutic Intervention (KATI) provides a high accuracy, of 92%, in predicting the next direction, and magnitude, of recovery movement during the intervention process. The high accuracy of movement prediction is achieved by a combination of two approaches to knowledge extraction. The first approach is an empirically derived decay based model which is commonly used to model survival times. The second approach is the application of rules, produced by applying a rule-based Data Mining algorithm to the domain data, which describes the different recovery behaviours which are exhibited. We demonstrate that KATI has the potential for being used to aid all therapeutic intervention processes by addition of an additional knowledge module in which the two knowledge extraction approaches, used in the tool, are applied to drug addict data. We also show the potential ease of use of KATI in other areas of intervention. |
|
|
|
|
|
|
