Knowledge Based MDSS using Data Mining
Taranath N.L1, Shanthakumar B. Patil2, Premajyothi Patil3, C. K. Subbaraya4

1Prof. Taranath N. L, Research Scolar, Department of CS & E, VTU, NCET, Bengaluru (Karnataka), India.
2Dr. Shantakumar B Patil, Professor and Head, Department of CS & E, NCET, Bengaluru (Karnataka), India.
3Dr. Premajyothi Patil, Professor, Department of CS & E, NCET, Bengaluru (Karnataka), India.
4Dr. C.K. Subbaraya, Principal & Professor, Department of CS & E, AIT, Chikkamagaluru (Karnataka), India.
Manuscript received on 17 July 2017 | Revised Manuscript received on 24 July 2017 | Manuscript Published on 30 July 2017 | PP: 4-8 | Volume-6 Issue-11, July 2017 | Retrieval Number: K24480761117/17©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: Medical Decision Support System (MDSS) links the patient information to promising diagnostic and treatment paths. It can be built either as Knowledge-based system or Learning-based system. Knowledge-based systems are human-engineered maps from best medical practices and patient data will be recommended. Learning-based systems derive the mapping techniques from data mining, statistical approaches and machine learning techniques. An Integrated decision support system integrates both Knowledge-based and Learning-based systems to provide a robust solution to the information challenge in the presence of partial information. In this work, we design a framework and concrete implementation of Integrated Medical Decision Support System to assist the Doctors in clinical decisions regarding the prescription of drugs. It uses the Knowledge base for prescribing the drugs to the patients, however if the available data is partial it employs the machine learning techniques to answer the query. It is suitable for many different healthcare settings and many different users. The framework is query-based and it can be adapted for use with many different end-user interfaces.
Keywords: Artificial Intelligence, Data Mining, Learning based Systems, Knowledge based Systems.

Scope of the Article: Data Mining