Recommendation system development for fashion retail e-commerce
Introduction
Recently, e-commerce has become an important channel for many retail businesses. The eMarketer (2017), an e-business marketing company, estimates worldwide retail e-commerce sales will increase from $2.29 trillion in 2017 to $4.48 trillion by the end of 2021. In spite of its success, e-commerce has a significant market limitation. While there is staff available to assist customers in offline stores, there is no staff to help buyers in online stores. In order to overcome this limitation, online stores provide various features, for instance, a “search and directory” to assist customers. Although these services can enhance purchase experience, online customers can only take advantage of them if they use them.
Recommendation systems are an innovative solution that overcomes the limitations of e-commerce services. Recommendation systems use customer behavior and information, and product information to identify customer preferences, and proactively suggest products that they are likely to buy. Many studies have been conducted to develop such recommendation systems and many practical systems have been successfully implemented in various businesses (Choi et al., 2012, Koren, 2009a, Linden et al., 2003, Wei et al., 2016).
Recommendation systems used in e-commerce have been developed to reflect unique domain characteristics (Portugal et al., 2015, Schafer et al., 1999, Sivapalan et al., 2014, Wang and Zhang, 2013, Zhao et al., 2015). This study aims to develop a recommendation system for a company – referred to as Company K in the following discussion – that sells fashion products through an online shopping mall as well as through offline shopping outlets. Company K has an average of 5 million members and sells around 40,000 products per year in the online shopping mall. There are around 1.5 million clicks and around 10,000 transactions per month. Company K also operates about 1300 offline stores in Korea and sells around 20,000 products per year.
Company K possesses the following unique operation environment:
- (1)
When purchasing fashion products, customers buy to replace or supplement their previous purchases, or preferred products.
- (2)
Demand for fashion products generally decreases over time due to seasonal changes. In general, people buy fashion products appropriate to the current season. Such a pattern in purchases is found frequently in fashion items, while purchases of other products like books and music do not display any significant relationship with the changing of the seasons.
- (3)
Fashion products can be sold in both online and offline stores. Most previous studies focus mainly on online stores. However, online and offline stores usually sell the same fashion products. Usually, customers first decide on potential purchases at online stores and then purchase them online.
Company K has already recognized that recommendation systems are a key success factor for its business and has used a system that employs a conventional item-based collaborative filtering for its online shopping mall. However, Company K wants a new recommendation system to be developed to reflect the fashion industry-specific characteristics discussed above. In the development of this new recommendation system, we address the following recommendation requirements.
First, the recommendation system should reflect the decline in preference for fashion products over time. Previous temporal recommendation studies assume that the intensity of preference decreases as time passes (Campos et al., 2013, Ding and Li, 2005, Ding et al., 2006, Hong et al., 2012, Koren, 2010, Larrain et al., 2015, Lathia et al., 2010, Xu et al., 2016). That is, these studies assume that recent preference-indicating behaviors, such as clicks or purchases for the same product, reflect stronger preferences than older ones. However, this study focuses on the decline in preference for fashion products which occurs over time following their release.
Second, the recommendation system needs to combine both offline customer preference data and online customer preference data. Purchases in offline shopping malls reflect the preferences of offline customers. Therefore, combining offline purchase data with that of online customers can improve the performance of online recommendation. Some researchers emphasize that online and offline information can help predict customer preferences, but have not applied this finding to the development of recommendation systems. (Cheema and Papatla, 2010, Dzyabura et al., 2016). Only a few studies consider the problem of integrating online and offline shopping mall data into recommendation systems (Adomavicius and Tuzhilin, 2001, Cantador et al., 2015, Kim et al., 2016, Nilashi et al., 2014a). The most important reason seems to be that it is difficult to find a domain where it is important to integrate both online and offline preference data. However, fashion products are sold through both online and offline stores and therefore preference data can be collected from these two stores. This study therefore aims to propose a way to combine online and offline preferences for recommending fashion products.
Third, customer purchase intent should be reflected by the recommendation system. When buying a product, the customer chooses an item that can be used with, or an item that replaces something that he or she had previously preferred. In this paper, the former is called a complementary product, and the latter is called a substitute product. This study proposes a method that recommends complementary products and substitute products separately using product category information.
Section snippets
Collaborative filtering recommendation systems
Recommendation systems are one of the most important applications in big data analytics and have performed excellently for numerous businesses (Bobadilla et al., 2013, Shi et al., 2014, Su and Khoshgoftaar, 2009). Many online companies, such as Amazon (Linden et al., 2003), Netflix (Koren, 2009a), Google (Das et al., 2007), and Facebook (Shapira et al., 2013), are using recommendation systems as part of their business.
Recommendation systems are broadly categorized into content-based systems and
Data preparation subprocess
Our system generates a product list, product metadata, purchase history data (offline) and click history data (online). First, the system constructs a product list from the online shopping mall database and generates product metadata, which includes a product code (individual and group code), gender type (male, female, unisex), product status (e.g., active and inactive), sale type (e.g., general sale and set sale), product type (clothing, shoes, etc.), and a production year. If any product does
Implementation
A recommendation system, called K-RecSys, was developed using ORACLE PL/SQL. The system runs on IBM Flex 240 (CPU 16 core (8 core ∗2), RAM 196 GB, HDD 1.2 T). The overall system architecture of K-RecSys is illustrated in Fig. 1.
Raw data are collected from the offline shop management system and online shopping mall system. Offline purchase history data spanning one year before the recommendation process begins is collected from the Enterprise Data Warehouse (EDW). Online click history data
Experimental design
For the experiment, K-RecSys was implemented into Company K’s shopping mall in addition to the existing system. Then, the A / B test was conducted over three weeks (May 20, 2015–June 1, 2015) to compare the performance of K-RecSys with the existing recommendation system. In the following discussion, customers who are recommended by the existing system are referred to as the control group, and those who are recommended by K-RecSys are referred to as the experimental group.
The experiment was
Click results
A total of 1,076,394 clicks occurred during the experimental period in the online shopping mall. From amongst this number, the control group clicked 532,598, which is 49.5% of all clicks, and the experimental group clicked 543,796, or 50.5% of all clicks. Fig. 2 illustrates daily click trends during the experimental period and Table 1 statistically displays the daily clicks by the two user groups.
According to the Fig. 2 and Table 1, there is no significant difference in daily clicks between two
Conclusion
In this study, we proposed a new method of recommending fashion products to customers by extending the existing collaborative filtering method to reflect the characteristics of fashion products. First, we considered the fact that fashion products are sold online and offline, and preferences for fashion products also appear online and offline by using online click data and offline purchase data to generate recommendations. Second, customer preference for fashion products generally tends to
Acknowledgements
This research was supported by the Bisa Research Grant of Keimyung University in 2017.
References (57)
- et al.
Recommender systems survey
Knowl.-Based Syst.
(2013) - et al.
Relative importance of online versus offline information for Internet purchases: product category and Internet experience effects
J. Business Res.
(2010) - et al.
A hybrid online-product recommendation system: combining implicit rating-based collaborative filtering and sequential pattern analysis
Electron. Commerce Res. Appl.
(2012) - et al.
Product recommendation with temporal dynamics
Expert Syst. Appl.
(2012) A personalized recommendation system based on product taxonomy for one-to-one marketing online
Expert Syst. Appl.
(2005)- et al.
Learning recency based comparative choice towards point-of-interest recommendation
Expert Syst. Appl.
(2015) - et al.
Multi-criteria collaborative filtering with high accuracy using higher order singular value decomposition and Neuro-Fuzzy system
Knowl.-Based Syst.
(2014) - et al.
A hybrid approach for movie recommendation via tags and ratings
Electron. Commer. Res. Appl.
(2016) - et al.
New recommendation techniques for multicriteria rating systems
IEEE Intell. Syst.
(2007) - et al.
Multidimensional recommender systems: a data warehousing approach
Time-aware recommender systems: a comprehensive survey and analysis of existing evaluation protocols
User Model. User-Adap. Inter.
Evaluating collaborative filtering recommender systems
ACM Trans. Inf. Syst.
GroupLens: applying collaborative filtering to Usenet news
Commun. ACM
The BellKor Solution to the Netflix Grand Prize
Cited by (179)
Service recommendation based on contrastive learning and multi-task learning
2024, Computer CommunicationsEnhanced fractional prediction scheme for effective matrix factorization in chaotic feedback recommender systems
2023, Chaos, Solitons and FractalsProduct recommendation using enhanced convolutional neural network for e-commerce platform
2024, Cluster Computing