TRApriori Classification Based Algorithm by Fuzzy Techniques to Reduced Time Complexity

. Our simple classification technique on this approach is also able to remove the unwanted data sets that are not useful for making the decision system. We can also combined the fuzzy techniques, TRApriori Algorithm and classification technique to provide the close output. Our classification based fuzzy mining association algorithm can also work on low support values. Due to online notes (Video based and Web based) for education also plays an important role for enhancement of their result. Students of this generation are smart due to internet. Also the cost of internet goes down , a low income family student can also used web based learning.Video from NPTEL can also be downloaded from you tube free of cost. Once downloaded the video is distributed amongst the students. Keywords—TRApriori, Classified data sets, fuzzy approach, quantitative data I.INTRODUCTION In data mining technique, association rules plays an important role in knowledge discovery technique. Data mining is a technique digging out the essential thing and leaving unimportant item. Similarly if there is a huge amount of data or quantitatively data then we need a suitable algorithm or technique to remove or hide unimportant data. If someone is suffering from any diseases then he/she will ensure from many different ways of health checkup, it means it will check its diseases from two or more doctor or two or more blood testing. Similarly I am using fuzzy logic, classification technique, TRApriori mining algorithm to get the close output. Here our classification based fuzzy mining algorithm help in reduced time complexity in later steps of this algorithm. So we can remove the quantitavely data that are not important for making knowledge discovery. Our classification based fuzzy mining association algorithm can also work on low support values. The remaining parts of this paper are organized as follows. Related research is reviewed in Section 2. The proposed fuzzy TRApriori data-mining algorithm is described in Section 3. An example is given to illustrate the proposed algorithm in Section 4. Experiments to demonstrate the performance of the proposed data-mining algorithm are stated in Section 5. Conclusions and future work are finally given in Section 6. Fuzzy logic are used for intelligent system like human similarity[25]. Several fuzzy system are used for the set of data with some domain [14,16-17,18,20,22-24,29]. Fuzzy approach with data mining approach has been used in [15,25,29] II. RELATED WORK As we know , the aim of data mining is to apply some kind of association rule on data sets. Getting this agrawal and his co-worker proposed some mining algorithm based on the large data sets to find association rule[1-10]. These break the mining steps into two phases. In the first phase candidate of itemsets are obtained and counted by scanning the transactions. The number of itemset must support the minimum pre-defined threshold value called minimum support. Then later we make the pair of item sets and apply the association rule for getting the required output. Srikant and agrawal also proposed mine association rule that are partitioned based[27].The fuzzy set was first introduced by zadah in 1965 [29]. Fuzzy set is used to define the exact answer of data set when human being is unable to provide answer. Hong et al, proposed a fuzzy mining algorithm to mine fuzzy rules from quantitative data[22]. They required each quantitative data into a fuzzy set and fuzzy steps to find fuzzy rule. Cai at al proposed weighted mining rule of data sets[15]. Yue et al, then extended the fuzzy concept based an vectors[28]. III. THE PROPOSED FUZZY DATA-MINING CLASSIFIED BASED ALGORITHM I read all the references[1-29] for classification based techniques. They have used the fuzzy based mining techniques using TRApriori algorithm. I added the extra classified techniques. This technique will not require to fuzzifies every itemsets. On later steps of mining algorithm. This will reduces some complexity to theprior work. The TRApriori algorithm also works on low support and low confidence. Our proposed methodology is based on two part. First part mostly deal with classification based TRApriori algorithm set. Then later we will also apply the Apriri algorithm dor important fuzzy values for finding the association rule. IV. AN EXAMPLE In this section, an example is given to illustrate the proposed Classification based TRA data-mining algorithm by fuzzy techniques. This is a simple example to our proposed model where I am taking the percentage of under graduate (Polytechnic) , graduate (Bachelor of Engineering)and post graduate (Master of technology) marks for their first and Rohit Miri et al, / (IJCSIT) International Journal of Computer Science and Information Technologies, Vol. 5 (3) , 2014, 3679-3683