Academic Journals Database
Disseminating quality controlled scientific knowledge

An Improved Single and Multiple association Approach for Mining Medical Databases

ADD TO MY LIST
 
Author(s): Mr. Sachin sohra | Mr. Narendra Rathod

Journal: International Journal of Advanced Computer Research
ISSN 2249-7277

Volume: 2;
Issue: 2;
Start page: 70;
Date: 2012;
VIEW PDF   PDF DOWNLOAD PDF   Download PDF Original page

Keywords: Data Mining | Apriori | Association Rule | Medical Diagnosis

ABSTRACT
The main aim of data mining is to extract usefulpatterns from huge amount of data. For thispurpose some effective techniques like Apriorialgorithm is presented. The major drawback ofApriori algorithm is Generate huge candidatesets:104 frequent 1-itemset will generate 107candidate 2-itemsets.To discover a frequent patternof size 100, e.g., {a1, a2, …, a100}, one needs togenerate 2100 which is approx 1030 candidates.Candidate Test incur multiple scans of database:each candidate. To remove the above drawback, wepresent an improved non candidate single andmultiple association approach for mining medicaldatabases. The developed approach generatesassociation rules for determining the relationshipsamong the diseases observed synchronously. Thegenerated association rules are too significant formaking early diagnosis for the correlated diseases.Some types of diseases can have triggering effectson different kinds of diseases. The symptoms anddiseases which have stronger effect on each othercan be determined and interpreted by theconstructed system and the large and extendeddatabases can be scanned effectively with thepruning property of the developed system.
Why do you need a reservation system?      Save time & money - Smart Internet Solutions