Target Customer Selection Method Based on Data Mining in Big Data Environment

There are rising interests in developing techniques for data mining. One of the important subfield in data mining is itemset mining, which consists of discovering appealing and useful patterns in transaction databases. In a big data environment, the problem of mining infrequent itemsets becomes more complicated when dealing with a huge dataset. Infrequent itemsets mining may provide valuable information in the knowledge mining process. The current basic algorithms that widely implemented in infrequent itemset mining are derived from Apriori and FP-Growth. The use of Eclat-based in infrequent itemset mining has not yet been extensively exploited. This paper addresses the discovery of infrequent itemsets mining from the transactional database based on Eclat algorithm. To address this issue, the minimum support measure is defined as a weighted frequency of occurrence of an itemsets in the analysed data. Preliminary experimental results illustrate that Eclat-based algorithm is more efficient in mining dense data as compared to sparse data.

Download Full-text

Big Data Clustering: Applying Conventional Data Mining Techniques in Big Data Environment

Innovations in Computer Science and Engineering - Lecture Notes in Networks and Systems ◽

10.1007/978-981-13-7082-3_58 ◽

2019 ◽

pp. 509-516 ◽

Cited By ~ 3

Author(s):

P. Praveen ◽

Ch. Jayanth Babu

Keyword(s):

Data Mining ◽

Big Data ◽

Data Clustering ◽

Data Mining Techniques ◽

Data Environment

Download Full-text

Optimization of Intelligent Data Mining Technology in Big Data Environment

Journal of Advanced Computational Intelligence and Intelligent Informatics ◽

10.20965/jaciii.2019.p0129 ◽

2019 ◽

Vol 23 (1) ◽

pp. 129-133 ◽

Cited By ~ 2

Author(s):

Wei Wang ◽

Keyword(s):

Data Mining ◽

Big Data ◽

Association Rules ◽

Confidence Level ◽

Strong Association ◽

Processing Efficiency ◽

Compression Technique ◽

Data Set ◽

Mining Technology ◽

Data Environment

At present, storage technology cannot save data completely. Therefore, in such a big data environment, data mining technology needs to be optimized for intelligent data. Firstly, in the face of massive intelligent data, the potential relationship between data items in the database is firstly described by association rules. The data items are measured by support degree and confidence level, and the data set with minimum support is found. At the same time, strong association rules are obtained according to the given confidence level of users. Secondly, in order to effectively improve the scanning speed of data items, an optimized association data mining technology based on hash technology and optimized transaction compression technology is proposed. A hash function is used to count the item set in the set of waiting options, and the count is less than its support, then the pruning is done, and then the object compression technique is used to delete the item and the transaction which is unrelated to the item set, so as to improve the processing efficiency of the association rules. Experiments show that the optimized data mining technology can significantly improve the efficiency of obtaining valuable intelligent data.

Download Full-text