Using Data Mining To Improve Consumer Retailer Connectivity

  • : Ms Word, Ms Word Format
  • : 70 Pages
  • : ₦3000
  • : 1-5 Chapters
  •  
  • Click to DOWNLOAD Materials
USING DATA MINING TO IMPROVE CONSUMER RETAILER CONNECTIVITY

 

ABSTRACT

 

Retailing is increasingly becoming a high performance sector in the  Nigeriaeconomy and retailers are fast seeking a competitive edge through technology. We describe the exploitation of Data Mining techniques and in particular association rule mining to analyze various baskets of a popular retail shop in Uyo. The aim of the  basket analysis was to allow retailers to quickly and easily look at  the size,  content  and  value of theircustomers1 products to understand patterns, affinities and associations with a viewto identifying cross-sell opportunities, improve shop iloor  layout  and organization, encourage impulse buying driving promotions  and  advertisement based on the database intelligence and identify new business opportunities. We used CRISP- DM (Cross-Industry Standard Process for Data  Mining) methodology for data mining and predictive analytics. CRISP DM was adopted because of its ability to  iteratively move back and forth in all the six stages of data miming  namely  business understanding, data understanding, data preparation, modeling, evaluation and deployment. The dataset used for this case study contained data for both loyal and non loyal customers. We cleaned the data and converted it to binary  format  for analysis.  The data was divided into two partitions in equal portions of two months periods. The results observed were that regularities in both partitions were fairly consistent such  as the rules and item set generated. Best transacted items revolved around basic  Fast Moving Consumer Goods. With the regularities observed a tloor plan and a stimulus response model were proposed for the retail shop with a view to improving impulse buying therefore improving sales.

 

 

CHAPTER ONE:

 INTRODUCTION

 

1.0    Overview

The exploitation of data mining techniques in  retail  industry is progressively being recognized  as a sure means of achieving success and retail shops that apply them will never go wrong. In this paper, we present a means of exploiting data mining techniques in a retail shop with a view to improving layout, encouraging impulse buying, promoting cross selling and revamping internal operations.

 

Data mining, or the efficient discovery of interesting patterns from large collections of data, has been recognized as an important area of database research. The most commonly sought patterns  are association rules which are a class of important regularities in data. Association Rules (Agrawal, Imielinski, &Swami, 1993; Agrawal &Srikant,  1994)  in data mining is a technique  that analyzes the correlations and patterns between sets of items. The key strength of association rule mining is that it can efficiently discover the complete set of associations that exist in data. These associations provide a complete picture of the underlying regularities in the domain. Different techniques, and algorithms have been proposed for solving  this  problem  (Tan, Steinbach & Kumar, 2004; Gregory Piatetsky-Shapiro,et al, 1996 ).These algorithms arc fully explained in chapter2.

 

1.1    Motivation

 

There is an increasing focus on data mining, which has been defined as the application of data analysis and discovery algorithms to large databases with the goal of discovering (predictive) models ( Gregory Piatetsky-Shapiro,et al, 1996). Business Week  (Berry  1994)  estimated  that over half of all retailers are using or planning to use database marketing, and those who do use it have good results.

 

Retail stores are massively collecting large volumes of data in their daily business operations and the retail shops in themselves stoke hundreds of thousands of items. One leading retail shop in Uyo had more than 250,000 unique items and  posted  an average of 7,000 transactions per  day. This volume of data was a sea of sitting knowledge that can yield strategic business

 

intelligence when extracted. Transactions at these stores were routinely captured at the point of sale. Furthermore major retail shops and  many more were using  Customer  loyalty cards  largely to retain their clients. These cards equally provided a means of understanding the customers' bio information and buying characteristics. Indeed they provided a means of understanding the  value of the customer to the shop and how to reach them.

With data mining analysis, retailers can drive more profitable advertisement and promotions (database driven), attract more customers into the stores, increase the size and value of basket purchases, improve loyalty card promotions with longitudinal analysis, test and learn by using a market place as a laboratory, empower planners and vendors to make smarter decisions, march inventory to needs by customizing layouts, assortments and pricing to the local demographics.

 

  • ProblemStatement

That most retail shops are sitting on enormous amount of information on  their  databases  is evident with the sprawling of retail shops and the massive  queues  of  customers  transacting within these shops. The introduction of customer loyalty cards, the  acceptance  of usage  of visa and credit cards are additional tools that capture customer  transaction  behavior  and demographics. Most managers plan their product placement and replenishment, promote their product and organize advertisement based largely on their experience rather than on the basis of database drivenintelligence.

 

1.3   Justification for thestudy

Retailers always assume risk every time they make decisions around buying, replenishment, advertising, promotions and assortment planning. These decisions need not  be  based  on experience or instincts. A 1 % lift in sales or 0.01% improvement in margin can tip the balance between success, survival or failure. Every retailer's top-line sales and success require constant fine tuning of controls available to the retailer. Most retailers still suffer from the old age retail problem of stocking too much of the wrong item and not enough of the right one.  The  right product moves and the wrong one sits until it is marked down. Customer's life is further complicated when he or she cannot get enough quantities of a popular product. Data mining therefore will leverage retailers on smarter decision makingprocess.

 

1.4               Objective

The main goal is to exploit data mining techniques to perform basket analysis with  a  view  to using the knowledge mined to improve on sales and assortmentplanning.

The specific objectives are:

  1. To examine algorithms for association rulemining
  2. To extract interesting and useful patterns from a retail shop
  • To develop models based on generatedpatterns.

 

1.5              Researchquestions

This study was guided by the following questions:

  1. What are the most frequently transacted items in the database within  the  setmetrics?
  2. What are the most interesting regularitiesin the database?
  • How can such regularities aid marketingstrategies?
  1. What model(s) can be generated from suchregularities?

 

1.6               Scope

The study was based on transactional data collected from a leading retail shop in UyoTown. The choice of this shop was informed by virtue of its location, at the centre of the city. Being that other branches are equally dispersed within and outside the city, we believed that the  finding  from the study will be a true reflection of what was applicable in other branches across thecity.

1.7              Structure of the report

The report was done in five chapters. Chapter one introduced data mining and the objective of mining retail outlet transaction data. Chapter Two part one reviewed data mining and part two reviewed association rule mining and the algorithms that implement it.  Chapter Three described the mining methodologies and or process, and  the choice of CRISP_DM employed  in this  work. In Chapter Four, data was described and the rules and item sets were generated and models developed. Chapter Five contained the analysis and discussions of the finding, conclusion and recommendation.

 

USING DATA MINING TO IMPROVE CONSUMER RETAILER CONNECTIVITY

Sharing is caring!

Leave a Reply