Design and Application of Coordinator App Based on Needs Survey

2020-02-01 08:56MALing马玲LIUChi刘驰XUBugao

MALing(马玲),LIUChi(刘驰),XUBugao

School of Apparel and Art Design, Xi’an Polytechnic University, Xi’an 710000, China

Abstract: The clothing industry is booming, and clothing prices are more affordable today. People’s requirements for clothing are no longer limited to the question of enough clothing to wear. In order to solve the main problem that consumers have a lot of clothes but don’t know how to match and look good, this article aims to design a matchmaker application (app)—Coordinator app. Through analyzing the current situation of existing apparel matching apps and typical case, the paper summarizes the existing clothing, the advantages and disadvantages of matching apps, and the combine with user needs survey feedback, to design the main four modules required by the Coordinator app which are elaborated and displayed separately.

Key words: Coordinator app; clothing matching; interface design; user needs; data analysis

Introduction

With the continuous influx of fast fashion and other apparel industries into our country’s apparel market[1], the number and frequency of consumers buying apparel has increased significantly[2]. However, the seemingly large amount of clothing and apparel has increased the time that consumers spend on their daily wear. There are many clothes, but they don’t know how to match and look good. Consumers wear and take off clothes over and over again, wasting a lot of time, and finally can only conclude that there are still too few clothes. When such helplessness comes, there are still a group of consumers who choose to browse major websites or use mobile apps to watch dressing instructional videos, summarize street photos, watch fashion Internet celebrity live broadcasts and other channels. However, they cannot find resonance with online demonstrations because of personal stature, dressing style and other reasons. Then how to match a set of satisfactory looks for their daily travel has become an important issue , which cannot be ignored under the dual pressure of massive clothing and fast pace of life.

1 Current Status of Existing Apparel Matching Apps

It is mainly based on application(app) store under Apple system as the search engine. When entering keywords such as wearing, clothing collocation, clothing design and collocation, fitting room and other keywords, 10 apps for related clothing collocations can be searched[3]. They are Haoda box, Dressing assistant, Men’s clothing, Fashion dressing, Shining, Mogujie, Clothing matching, Clothing designing and matching, Virtual fitting room, Shopping and matching. According to the feedback provided by users who have used them, it is shown that the expression of the core functions of these 10 apps are not clear, and the integration of teaching, shopping, advertising, and social interaction cannot highlight them core value. Most of the clothes demonstrated on the platform are very different from the consumer’s own clothes. Therefore, they will guide the purchase. In the end, users still failed to master the wearing skills with their existing clothes. Most platforms can not utilize the user’s own body size for virtual fittings, and often the ideal matching effect is not satisfactory.

1.1 Actual case analysis

This section analyzes the three typical clothing matching apps with the largest number of users: Haoda box, Virtual fitting room and Men’s clothing from the main functions, interface design, and operability of different apps. This section describes how to show the main functions of the program to users, and to better understand the thinking and algorithms behind apps by investigating the core parts of different clothing matching apps[4]. These studies can provide some theoretical guidance for the Coordinator app.

1.1.1Haodaboxapp

Haoda box is committed to solving the daily wear problem for users in five main aspects: virtual model, online trying, personalized recommendations, brand pavilion, discount hall and wear notes. Users can upload their own facial photos, and then combine their height and weight to generate an exclusive avatar, and try on clothes of various brands intuitively and effectively. Users can choose their favorite dressing styles in advance, and the system will recommend corresponding matching schemes according to their personal dressing preferences. When users encounter their favorite dresses, they can store them in “my wardrobe” for easy guidance on dressing later. Each set of collocations is composed of three basic items: top, bottom and shoes. On the right hand side of the virtual model, the specific information of the clothes will be displayed, including the price of each item and the purchase links, as shown in Fig. 1.

Fig. 1 Main interface of the Haoda box app

For users who have mastered certain clothing matching skills, Haoda box is an app that can provide inspiration for wearing and stimulate shopping desire. But in essence, Haoda box is a shopping software. It cannot explain each set of collocations in text, and it cannot make users effectively apply the clothes they have learned on the software to their existing clothing.

1.1.2Virtualfittingroomapp

Virtual fitting room app is not for playing, but the app hopes that users can find a way of matching clothes that suits them in daily life by creating real fitting models and manually trying various matchings. The main modules of the virtual fitting room are the image library and the fitting room, as shown in Fig. 2. Among them, users in the image library can customize the body, skin color, hairstyle, scene,etc. The fitting room module has a certain amount of full set of clothes for users to choose to try on. The detailed clothing types are shown in Fig. 3.

Fig. 2 Main interface of Virtual fitting room app

Fig. 3 Clothing types of Virtual fitting room app

1.1.3Men’sclothingapp

Men’s clothing is mainly composed of four modules(shown in Fig. 4): fashion guide, online exclusive matching assistant, carefully selected style brand partners and high quality of life.

Compared with other clothing matching software, the personalized service for male consumer groups and the clothing dressing teaching for different body types (fat,thin, short,etc.) promoted by Men’s clothing are ingenious and practical. However, summarizing the user who have used the Men’s clothing, it is obvious that the app has not deviated from its nature of online shopping platform as its core. The matching skills learned from the app are difficult to translate into practice and apply to the users. It’s not a long-term solution to blindly imitate fashion outfits or buy the whole set.

Fig. 4 Main interface of Men’s clothing app

1.2 Summary of this section

Through the analysis of the builder model of the existing clothing matching apps, we can effectively learn the deficiencies and advantages of the apps.

According to the summary of the three typical clothing matching apps in Table 1, the research on users’ existing clothing matching methods can use the functions of existing apps to summarize and improve the minimum knowledge base. It includes learning the matching and virtual fitting of users’ existing clothes in the virtual fitting room, learning the personalized matching suggestions that can be given in the Haoda box app, and learning the matching which based on user’s need in the Men’s clothing app, it is also necessary to improve the important fashion measurement system of scoring match, which is not available in these three apps.

There comes a time in every boy s life when he must take on responsibilities. This was important. I might even get to back the car out of the driveway.

Table 1 Summary of three clothing matching apps

2 User Needs Survey and Analysis

2.1 Questionnaire design and distribution

The implementation of the Coordinator app project is inseparable from the needs and recognition of consumers[5]. In order to understand the current status of clothing, the matching of consumers and the problems consumers face, as well as the use of subsequent app products, this questionnaire survey was conducted through internet platforms such as WeChat and Weibo. We distributed 520 copies, received 512 valid questionnaires with an effective rate of 98.5%. Among them, the questionnaire is divided into three main parts: basic information, current situation survey, and product usage survey.

2.2 Analysis of survey results

2.2.1Basicinformation

The survey respondents came from 16 provinces, cities and regions across the country. The main sources of the questionnaires were Shaanxi, Beijing, Ningxia, Guangdong and Hubei. And almost covers all age groups and education levels. A frequency analysis of the age and gender of the surveyed showed that women accounted for 74%, and the 18-25 age group accounted for 76%.

2.2.2Currentsituationsurvey

Based on the early interviews on the problems encountered by consumers in dressing and matching in major colleges and universities, shopping malls and other densely populated places in Xi’an, nine typical issues were summarized. The corresponding frequency analysis is shown in Table 2. According to the analysis, 65.23% of consumers have been aware of the phenomenon of pressing the bottom of their wardrobe. Although there are a lot of clothing, they still can’t generate enough inspiration to wear before they out, and even 61.91% of consumers are in a situation that they take a long time to wear but have a general clothing matching effect. Analytic hierarchy process (AHP) level analysis[6]and weight values are shown in Table 3. According to Q1, Q2 and Q3, a third-order judgment matrix is constructed and the corresponding weight values of the three items are: 23.517%, 47.156% and 29.327%. Among them, the weight value of Q2 is significantly higher than the other two, which indicates that consumers generally agree and value the time spent on dressing and matching before traveling.

These three multiple-choice questions are analyzed and summarized by multiple responses analysis. The multiple response analysis of Q4 (What do you think is the reason why many clothes in the closet press on the bottom of the box?) is shown in Table 4. Regarding whether the proportional distribution of each option of the multiple choice question is uniform, the chi-square goodness-of-fit test is used for analysis. The goodness of fit test in Table 4 shows significance (χ2=271.963,p=0.000<0.05), which means that the selection ratio of each item is significantly different, and the difference can be specifically compared by response rate or penetration rate[7]. Specifically, clothes are updated quickly and bought frequently; new clothes are unsightly after being worn once or twice; new clothes cannot be used to match the clothes in the closet to achieve a satisfactory effect. Over time, the new clothes become old. The response and penetration rate of these three questions are significantly higher. By analyzing Q5 and Q6 in the same way, we can see that most consumers spend a lot of time on different colors of clothing, weather and attendance before going out, and the way to improve the efficiency of clothing match is to ask friends and relatives for advice, and to refer to teaching videos and street pictures.

Table 2 Current frequency analysis

Table 3 AHP level analysis and weight value

Table 4 Multiple response analysis of Q4

2.2.3Productdemandandusagesurvey

Table 5 Product demand and frequency analysis

3 Coordinator App Design

Based on the preliminary analysis and summary of typical clothing matching app cases and user needs and usage surveys, the Coordinator app was designed and developed. Its main modules and functions are as follows.

3.1 Cloakroom module

Cloakroom module is mainly responsible for the storage of existing clothing in the user’s own wardrobe. The clothing storage interface is shown in Fig. 5. Users manually take photos and upload existing clothes to form user’s own unique cloakroom in Coordinator app. The specific division of clothing types in the cloakroom is shown in Fig. 6. This module is designed as the core module of the entire app at the beginning of the design. It should make full use of the existing clothing that consumers can continue to wear. When uploading clothing, users need to set the type, color, occasion and season of each piece of clothing. And then, using the expert knowledge matching system behind the Coordinator app to match the desired clothes for users, so as to achieve the goal of saving time and improving efficiency.

Fig. 5 Clothing storage interface

Fig. 6 Clothing types in cloakroom

3.2 Virtual image module

Virtual image module is mainly to form a personal virtual image through the collection of the user’s personal physical characteristics. The virtual image creation interface is shown in Fig. 7, and the formed personal virtual image is shown in Fig. 8. The establishment of the personal virtual image is to prepare for the matching of the latter module, to show the effect of the matched clothing on the virtual image, and to easily try it on with one button.

Fig. 7 Virtual image creation interface

Fig. 8 Personal virtual image

3.3 Clothing matching module

The matching module of the Coordinator app realizing match is mainly by the expert knowledge in the clothing intelligent matching system[10-12]. The process of the clothing recommendation system is as follows. Firstly, the system maintains interaction with customers through a interactive interface. Secondly, the clothing recommendation system uses the reasoning engine, and continuously draws conclusions from the known information based on the clothing matching knowledge and the customer’s data. Matching knowledge is mainly based on clothing color principles and clothing style knowledge[13-14]. The proposed system can search for suitable clothing matching for customers according to their skin color and body shape. In addition, it stores the intermediate results in the knowledge base for further inference, and converts the unknown state of the problem into a known state. Finally, the system searches the fact base for information about clothing for sale based on the results of the inference engine. The system generates the final clothing recommendation result and gives an explanation through an explanation mechanism[15-16].

3.3.1Clothingmatchingprinciple

The system structure of the Coordinator app collocation method is shown in Fig. 9. It consists of five parts: interactive interface, knowledge base, inference engine, knowledge acquisition mechanism and interpretation mechanism. Among them, the structure of the knowledge base is shown in Fig. 10, which consists of two parts: a fact base and a rule base. The fact base includes two parts about the data of existing clothing (clothing color, style, suitable season,etc.) and user personal information (users’ personal dressing style preferences,etc.), the rule base includes the clothing matching rule table and the index table generated by template[17].

Fig. 9 User’s existing apparel recommendation system structure based on the expert system

Fig. 10 Knowledge base structure

3.3.2Clothingmatchingrealization

Based on the analysis and summary of the existing clothing matching apps in the early stage, it is concluded that the Coordinator app can solve problems of more existing clothes and difficult matching. It can not only improve the efficiency of life, but also get matching according to the needs of users. A matching and scoring mechanism is designed for users to score and feedback according to the clothing recommendation and matching scheme given by the Coordinator app. And then, the system will improve the quality and efficiency of the clothing matching recommendation program based on the scoring situation. The specific operation process of the match module is shown in Fig. 11.

3.4 Personal center module

The Personal center module is the user center of Coordinator app[18], which mainly includes the user’s basic personal information (such as gender, age, dressing style preferences). The user registration and login interface of personal center module is shown in Fig. 12.

Fig. 12 User registration and login interface

4 Conclusions

So far, there is no similar app in the mobile application mall. Therefore, this is a risk and a bold attempt for the preliminary investigation and development of the Coordinator app, and it is also a great opportunity. The problem such as crowd positioning, later publicity and promotion, continuous optimization of design and after-sales service guarantee of app are serious, which are the challenges that application developers must face in the development process.