A self-adaptive k-means classifier for business incentive in a fashion design environment

Abstract

An incentive mechanism to target market for fashion designers is proposed. Recent researches have been focused on the art, style or the design; while a few were based on traditional practice. In this study, economy is considered as a major liberation in the fashion world by analyzing six attributes, namely, style, color, fabric, brand, price and size that could bring about commercial success. Dataset of 1000 customers’ records were used and categorized as original, combined and new designs using self-adaptive k-means algorithm, which extract common attributes that would foster better business from the dataset. The results would be useful to designers in knowing the type of designs usually ordered by customers with the design code, and which combinations of the attributes have high patronage. In addition, customers would have easy access to the best and current designs invoke from a combination of highest patronized designs.