Please use this identifier to cite or link to this item: https://rda.sliit.lk/handle/123456789/2780
Title: A Web Application to Support Customer Churn Management for Retail Grocery Stores
Authors: Mallawarachchi, S. N
Rodrigo, M. N. D
Gunaratne, M. A. S. N
Gamage, M. P
Qamra, N. N
Keywords: Customer-Churn
Machine-learning
Recommendation-system
Market-basket
Product-analysis.
Issue Date: 2022
Publisher: United International Journal for Research & Technology |
Series/Report no.: United International Journal for Research & Technology;Volume 03, Issue 03
Abstract: — In the business world ‘Customer Churn’ is a principal issue. The retail grocery business holds a peak point in churning customers due to various reasons. Churn means gradually breaking every connection with the business by the customers. According to the experts, retaining the existing customers cost less, than attracting new customers. Therefore, a web-based prediction model; “CRetention” with some additional features is proposed as a solution. The main features in the proposed system are to analyze data and predict customers who are about to churn, manage the storage of inventory items, provide marketing strategies by market basket analysis, and offer personalized marketing recommendations to retain customers. Machine Learning and Deep Learning technologies are used to implement the solution. The main advantage and novelty of the product are that a definition for churn adjusted to a retail business is created and churners and results are obtained are based on a real scenario. It is clear that the retail grocery store owners highly recommend and appreciate the proposed system from a survey conducted.
URI: http://rda.sliit.lk/handle/123456789/2780
ISSN: 2582-6832
Appears in Collections:Research Papers - Open Access Research
Research Papers - SLIIT Staff Publications
Research Publications -Dept of Information Technology

Files in This Item:
File Description SizeFormat 
UIJRTV3I30026.pdf225.08 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.