Purpose




The purpose of this paper is to exploit the fullest potential and capability of different non-traditional machining (NTM) processes, it is often recommended to operate them at their optimal parametric combinations. There are several mathematical tools and techniques that have been effectively deployed for identifying the optimal parametric mixes for the NTM processes. Amongst them, grey relational analysis (GRA) has become quite popular due to its sound mathematical basis, ease to implement and apprehensiveness for multi-objective optimization of NTM processes.




Design/methodology/approach




In this paper, GRA is integrated with fuzzy logic to present an efficient technique for multi-objective optimization of three NTM processes (i.e. abrasive water-jet machining, electrochemical machining and ultrasonic machining) while identifying their best parametric settings for enhanced machining performance.




Findings




The derived results are validated with respect to technique for order preference by similarity to ideal solution (TOPSIS), and analysis of variance is also performed so as to identify the most significant control parameters in the considered NTM processes.




Practical implications




This grey-fuzzy logic approach provides better parametric combinations for all the three NTM processes with respect to the predicted grey-fuzzy relational grades (GFRG). The developed surface plots help the process engineers to investigate the effects of various NTM process parameters on the predicted GFRG values.




Originality/value




The adopted approach can be applied to various machining (both conventional and non-conventional) processes for their parametric optimization for achieving better response values.

Application of grey-fuzzy logic technique for parametric optimization of non-traditional machining processes

Objectives: This paper introduces an improvement over the use of Artificial Neural Networks (ANN) for breast cancer detection. Methods: It suggests use of frequent pattern mining for minimizing the dimensions of breast cancer database. After reduction step, the database is then input to an ANN for classification. Findings: We have shown through experimentation that the proposed model not only reduces the input database dimensions but also produces better classification results. Application: The proposed model will be highly beneficial in the field of medicine and will help in precise detection of breast cancer.

Using Frequent Itemset Mining for Breast Cancer Detection

Objectives: In data mining, finding frequent item set from voluminous databases is an important role. It is a challenging task to find item sets frequently arise in the exponential growth of the databases. The number of scans to find the frequent item set is assessed as more and created redundancy. Methods/Statistical analysis: In this research work, a new technique is applied to improve the reduction in number of scans and minimize the redundancy. An algorithm is developed to mine the frequent item set using clustering techniques. A cluster table of frequent item set is created with minimal number of scanning. Findings: The support count is fixed to eliminate the duplicates which minimize the redundancy. This algorithm is appreciable, because the intermediate data in the dataset can be reused. The efficiency of algorithm can be identified by seeing the experimental results that is significantly performed well than the exiting algorithms. Applications/Improvements: In future, other algorithms are applied for the same data set to test its effectiveness in order to minimize the redundancy which avoids wastage of the memory space.

Mining Frequent ItemSet Based on Clustering of Bit Vectors

Frequent item sets is one of the most investigated fields of data mining. The significant feature is to find new techniques to reduce candidate item sets in order to generate frequent item sets efficiently. This paper introduces an efficient algorithm called Enhanced Candidate Generation for Frequent item set Generation (ECG for FIG) for finding frequent item sets from large databases. The existing algorithm for frequent item set generation scan the original database more than once, use more storage space, take more processing time. The proposed algorithm gives a solution to this by representing the transactions in the database with decimal numbers instead of binary values and strings. The original database is scanned only once and is converted into an equivalent decimal value to reduce the storage space. The subset generation concept is used to generate frequent item sets. Thus the proposed algorithm reduces the scanning time, processing time and the storage space respectively. When compared with the existing algorithms, the proposed algorithm takes very less execution time and memory. When implemented the algorithm using java and tested with WEKA tool, for 400 transactions of twenty five items, ECG for FIG is taking only 800 bytes of memory and 2000000000 ns (two seconds), whereas all the other above mentioned algorithms are taking 20800 bytes of memory and more than two seconds.

K. Manivannan

Papers

Enhanced Candidate Generation for Frequent Item Set Generation