Hybrid Computational Intelligence System for Fashion Design: A Case of Genetic-Fuzzy Systems With Interactive Fitness Evaluation

Hybrid Computational Intelligence System for Fashion Design: A Case of Genetic-Fuzzy Systems With Interactive Fitness Evaluation

Copyright: © 2022 |Pages: 16
DOI: 10.4018/IJSDA.20221101.oa2
Article PDF Download
Open access articles are freely available for download


The domain of fashion design evolves continuously and is highly personalised, demanding intelligent and customised recommendation. The traditional artificial intelligence-based systems offer solutions based on stored knowledge; hence they can be quickly obsolete and require high effort. To meet the fashion designers’ needs and provide tailor-made recommendations effectively, a hybrid genetic-fuzzy system is proposed with interactive fitness functions. The system is based on generic hybrid architecture using fuzzy logic and genetic algorithm, which can be used to evolve various products in different domains and tested with interactive fuzzy fitness functions. The design of the generic architecture meets the research gap identified through an in-depth literature survey. To prove the utility of the architecture, an experiment is carried out showing encoding scheme, genetic operators, fuzzy membership functions, and fuzzy rules. The results are also discussed, along with the comparison, advantages, applications, and possible future enhancements.
Article Preview

1. Introduction

It is said that the dressing sense of a person conveys the person’s character, mood, attitude, faith, and status. Garments are essential not only for safety and protection but also for the exhibition of personality and to show a specific aspect such as religious beliefs. People may choose from ready-made clothes or tailor-made depending on various factors such as design, price, fitting, etc. In the post-Covid-19 era, where people cannot afford to explore places like design studios and boutiques physically for a unique and customised product, it is difficult to quench the thirst for fashion. Further, a personal designer is also not an affordable solution for everybody. On the other hand, the designers also need to find innovative designs for the garment to get a competitive advantage on the global market and to lock in their loyal and valuable consumers. In this post-pandemic era, when the virtual has become a new standard, there is a need for an intelligent system that recommends the design of garments to consumers and fashion designers.

Many ideas and styles appear on the World Wide Web (Web), social media and catalogues. Finding an appropriate and customised design is a real difficulty in the domain. Traditionally, the creation of a garment is mainly based on designers’ experience and users’ choices. Sometimes, besides the common people, the fashion designers also face a challenge to find innovative fashions in various cloth categories. Such decision making about fashion is more creative, artistic, and mundane in nature. As observed, the machines are not very good at such innovative and mundane tasks. Another challenge from the user’s side is to decide on a purchase based on standard ratings and recommendations made by different categories of users. The authenticity of such a review is also needed to be taken into consideration.

Suggestions about fashion in the domain of ready-made garment are always challenging considering the involvement of many parameters and the diversity of the users. This is one of the significant reasons why such systems need customisation. Earlier, traditional artificial intelligence-based systems were used to provide customised and effective solutions through a recommendation system. As intelligence requires knowledge, such intelligent systems are primarily knowledge-based. The most popular knowledge-based systems experimented by professionals are the expert system. However, acquiring and representing knowledge are challenging and tedious tasks. Often, such expert systems are not adaptive enough and do not offer advantages such as self-learning and automatic knowledge update. Because of the stored knowledge in the knowledge base, which can not update itself, such systems also tend to be quickly obsolete. Here, a computational intelligence/machine learning-based system helps a lot. With the advancements of modern computational intelligence techniques, it is possible to meet these challenges. Techniques such as artificial neural network, evolutionary algorithms, and fuzzy logic are considered as the significant components of the modern machine learning consortium. The artificial neural networks learn in an automatic manner from big amount of data in supervised and unsupervised manners. The evolutionary algorithms help evolve better and optimised solution components according to the customised fitness functions. Fuzzy logic-based systems especially handle uncertainty and approximate reasoning. Each technique has its pros and cons. Artificial neural networks and genetic algorithms do not handle uncertainty and reasoning, as they do not explicitly store knowledge. A fuzzy logic-based system does not have the virtue of self-learning or evolutionary advantages. Hence, hybridisation of more than one technique is suggested in recommending fashion designs to the users in the most customised manner.

Complete Article List

Search this Journal:
Volume 12: 1 Issue (2024): Forthcoming, Available for Pre-Order
Volume 11: 5 Issues (2022)
Volume 10: 4 Issues (2021)
Volume 9: 4 Issues (2020)
Volume 8: 4 Issues (2019)
Volume 7: 4 Issues (2018)
Volume 6: 4 Issues (2017)
Volume 5: 4 Issues (2016)
Volume 4: 4 Issues (2015)
Volume 3: 4 Issues (2014)
Volume 2: 4 Issues (2013)
Volume 1: 4 Issues (2012)
View Complete Journal Contents Listing