Application of fuzzy clustering data mining approach to garment sizing problem

ABSTRACTCurrently owing to the annual increase of data, there is urgent need for anthropometric database and useful approaches, in designing machine, equipment, facilities, workspace and other products. Fuzzy clustering which is an important way of clustering is an approach where an object is grouped into more than one partitions. The approach is used to bridge the gap between the anthropometry dimension of an individual and garment size which is a problem attributed to the size of the garment because product are designed to fit a user and not the user fitting the product.The difference between the anthropometric dimensions and dimensions of products designed for human body is the major problem of unfit. In finding the solution to the problem of garment sizing, the fuzzy clustering data mining approach was applied. The aim of the study is to develop a garment sizing chart for men in Nigeria for the purpose of designing and manufacturing clothing that will make a good fit.

The anthropometry survey involves taking the measurement of 220 subjects between the ages 18 to 60years in accordance with ISO 8859/1989 body dimension standard.  This involves taking permission from the subjects for their body dimensions to be measured. Twelve variables were measured for the production of the male garment chart.  The collected data were statistically analyzed to identify the relationships between the variables. Principal component analysis was done to get the vital body measurements which then undergo fuzzy clustering for the development of the size chart.

 The body dimensions were identified and were distributed into five clusters within the population. The neck girth, shoulder width and chest girth were seen as the critical dimensions that determine good fit for tops and all the variables measured were discovered to be important for the design of the trouser.  Validation of the result was done by using different numbers of cluster data, this is a measure known as adjusting the fuzzy overlap with the fuzzy c-means clustering technique.

The visible difference seen in the anthropometry of the population proves that there is need for further studies in other regions and part of the country for ergonomic design, industrial and architectural design and as well as  garment design. These measurements are critical element for designers and for the garment producers, it will help to produce less unfit garments and also the consumer will have less difficulty in choosing a more fitting garment.

 

Keywords: Fuzzy clustering; Data mining; Anthropometric data; Principal Component Analysis; Garment Sizing.

Overall Rating

0

5 Star
(0)
4 Star
(0)
3 Star
(0)
2 Star
(0)
1 Star
(0)
APA

Enoh, D. (2019). Application of fuzzy clustering data mining approach to garment sizing problem. Afribary. Retrieved from https://afribary.com/works/application-of-fuzzy-clustering-data-mining-approach-to-garment-sizing-problem

MLA 8th

Enoh, Daniel "Application of fuzzy clustering data mining approach to garment sizing problem" Afribary. Afribary, 02 Jun. 2019, https://afribary.com/works/application-of-fuzzy-clustering-data-mining-approach-to-garment-sizing-problem. Accessed 23 Dec. 2024.

MLA7

Enoh, Daniel . "Application of fuzzy clustering data mining approach to garment sizing problem". Afribary, Afribary, 02 Jun. 2019. Web. 23 Dec. 2024. < https://afribary.com/works/application-of-fuzzy-clustering-data-mining-approach-to-garment-sizing-problem >.

Chicago

Enoh, Daniel . "Application of fuzzy clustering data mining approach to garment sizing problem" Afribary (2019). Accessed December 23, 2024. https://afribary.com/works/application-of-fuzzy-clustering-data-mining-approach-to-garment-sizing-problem