A Hyper-Personalized Product Recommendation System Focused on Customer Segmentation: An Application in the Fashion Retail Industry

Yıldız, Emre; Şen, Ceyda Güngör; Işık, Eyüp Ensar

doi:10.3390/jtaer18010029

Cited by 24 publications

(10 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Recent studies such as Sulikowski [ 12 , 13 ] have emphasized the importance of visual aspects and layout in recommendation systems, which significantly impact user purchase decisions. Additionally, the work by Yildiz [ 14 ] introduces a hyper-personalized product recommendation system focused on customer segmentation, highlighting the evolving nature of e-commerce strategies.…”

Section: Literature Reviewmentioning

confidence: 99%

Advancing e-commerce user purchase prediction: Integration of time-series attention with event-based timestamp encoding and Graph Neural Network-Enhanced user profiling

Zhou,

Hudin

2024

PLoS ONE

View full text Add to dashboard Cite

In recent years, the global e-commerce landscape has witnessed rapid growth, with sales reaching a new peak in the past year and expected to rise further in the coming years. Amid this e-commerce boom, accurately predicting user purchase behavior has become crucial for commercial success. We introduce a novel framework integrating three innovative approaches to enhance the prediction model’s effectiveness. First, we integrate an event-based timestamp encoding within a time-series attention model, effectively capturing the dynamic and temporal aspects of user behavior. This aspect is often neglected in traditional user purchase prediction methods, leading to suboptimal accuracy. Second, we incorporate Graph Neural Networks (GNNs) to analyze user behavior. By modeling users and their actions as nodes and edges within a graph structure, we capture complex relationships and patterns in user behavior more effectively than current models, offering a nuanced and comprehensive analysis. Lastly, our framework transcends traditional learning strategies by implementing advanced meta-learning techniques. This enables the model to autonomously adjust learning parameters, including the learning rate, in response to new and evolving data environments, thereby significantly enhancing its adaptability and learning efficiency. Through extensive experiments on diverse real-world e-commerce datasets, our model demonstrates superior performance, particularly in accuracy and adaptability in large-scale data scenarios. This study not only overcomes the existing challenges in analyzing e-commerce user behavior but also sets a foundation for future exploration in this dynamic field. We believe our contributions provide significant insights and tools for e-commerce platforms to better understand and cater to their users, ultimately driving sales and improving user experiences.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Advancing e-commerce user purchase prediction: Integration of time-series attention with event-based timestamp encoding and Graph Neural Network-Enhanced user profiling

Zhou,

Hudin

2024

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…The K-means algorithm is a method designed to divide a collection of objects into K clusters. The number of subgroups has been established, and the centroids are assigned randomly to observations in the dataset [17]. In order to minimize the variance within clusters, the algorithm employs an iterative process that involves two steps.…”

Section: Customer Segmentationmentioning

confidence: 99%

Optimizing Decision-Making Through Customer-Centric Market Basket Analysis

Hoque,

Islam,

Mohtasim

2024

JOSA

View full text Add to dashboard Cite

In the realm of understanding consumer purchasing behaviors and refining decision-making across diverse sectors, Market Basket Analysis (MBA) emerges as a pivotal technique. Traditional algorithms, such as Apriori and Frequent Pattern Growth (FP-Growth), face challenges with computational efficiency, particularly under low minimal support settings, which precipitates an excess of weak association rules. This study introduces an innovative approach, termed Customer-Centric (CC)-MBA, which enhances the identification of robust association rules through the integration of consumer segmentation. By employing Recency, Frequency, and Monetary (RFM) analysis coupled with K-means clustering, customers are categorized based on their purchasing patterns, focusing on segments of substantial value. This targeted approach yields association rules that are not only more relevant but also more actionable compared to those derived from conventional MBA methodologies. The superiority of CC-MBA is demonstrated through its ability to discern more significant association rules, as evidenced by enhanced metrics of support and confidence. Additionally, the effectiveness of CC-MBA is further evaluated using lift and conviction metrics, which respectively measure the observed co-occurrence ratio to that expected by chance and the strength of association rules beyond random occurrences. The application of CC-MBA not only streamlines the analytical process by reducing computational demands but also provides more nuanced insights by prioritizing high-value customer segments. The practical implications of these findings are manifold; businesses can leverage this refined understanding to improve product positioning, devise targeted promotions, and tailor marketing strategies, thereby augmenting consumer satisfaction and facilitating revenue growth.

show abstract

“…The elbow method (EM) and silhouette validation index are commonly used to determine the optimal number of clusters and evaluate the quality of clustering. The EM formula calculates the sum of distances within each cluster and considers the number of points in each cluster [19]. On the other hand, the silhouette validation index, introduced by Rousseeuw [20], measures the proximity of each data point to others within the same cluster and evaluates how well clusters are separated.…”

Section: Unsupervised Machine Learning: Clustering and Association Ru...mentioning

confidence: 99%

“…Association rule mining (ARM) is widely utilized in a multitude of industries, such as market basket research [8,23], stock market analysis [24], recommendation systems [7,19,22,[25][26][27], healthcare [28,29], and more [30]. This powerful technique plays a pivotal role in aiding organizations in making informed decisions [8,22,25,26], improving customer experience [7,19], and implementing preventive strategies [28,29]. Data mining identifies frequent itemsets (groups of items that frequently appear together) and generates explanations for them.…”

Section: Unsupervised Machine Learning: Clustering and Association Ru...mentioning

confidence: 99%

Enhancing Retail Transactions: A Data-Driven Recommendation Using Modified RFM Analysis and Association Rules Mining

Chen,

Gunawan

2023

Applied Sciences

View full text Add to dashboard Cite

Retail transactions have become an integral part of the economic cycle of every country and even on a global scale. Retail transactions are a trade sector that has the potential to be developed continuously in the future. This research focused on building a specified and data-driven recommendation system based on customer-purchasing and product-selling behavior. Modified RFM analysis was used by adding two variables, namely periodicity and customer engagement index; clustering algorithm such as K-means clustering and Ward’s method; and association rules to determine the pattern of the cause–effect relationship on each transaction and four types of classifiers to apply and to validate the recommendation system. The results showed that based on customer behavior, it should be split into two groups: loyal and potential customers. In contrast, for product behavior, it also comprised three groups: bestseller, profitable, and VIP product groups. Based on the result, K-nearest neighbor is the most suitable classifier with a low chance of overfitting and a higher performance index.

show abstract

A Hyper-Personalized Product Recommendation System Focused on Customer Segmentation: An Application in the Fashion Retail Industry

Cited by 24 publications

References 67 publications

Advancing e-commerce user purchase prediction: Integration of time-series attention with event-based timestamp encoding and Graph Neural Network-Enhanced user profiling

Advancing e-commerce user purchase prediction: Integration of time-series attention with event-based timestamp encoding and Graph Neural Network-Enhanced user profiling

Optimizing Decision-Making Through Customer-Centric Market Basket Analysis

Enhancing Retail Transactions: A Data-Driven Recommendation Using Modified RFM Analysis and Association Rules Mining

Contact Info

Product

Resources

About