Conference Paper

Identifying complex and multi-step customer projects: A graph mining market basket approach

September 2016

Conference: Proceeding of AMBS Big Data Forum
At: Manchester, United Kingdom

Authors:

Anastasia Griva

Ollscoil na Gaillimhe – University of Galway

Apostolos N. Giovanis

University of West Attica

Customer projects are major do-it-yourself (DIY) undertakings, involving a considerable amount of planning, money, and effort. Examples of customer projects include installing a paver patio, tiling kitchen walls, building a backyard football toss, and renovating the bathroom. Such projects require several cross-category purchases through multiple shopping trips (Wolf and McQuitty 2011). Identifying and understanding customer projects provides novel insights into customer behavior, which in turn, improve advertising planning and the development of effective marketing decisions. In this work, we seek to contribute to the growing literature that utilizes graph mining techniques to identify persistent multi-trip purchase patterns (Dhar et al. 2014; Kim, Kim, and Chen 2012; Oestreicher-Singer et al. 2013; Videla-Cavieres and Ríos 2014). We do so, by presenting an analytical method for the identification of customer projects from retail sales transactions and by demonstrating the utility, validity, and replicability of our approach using a data set of 20,000 customers across two years from a Fortune 500 specialty retailer.

ResearchGate has not been able to resolve any citations for this publication.

ResearchGate has not been able to resolve any references for this publication.

Article

Full-text available

Penerapan Metode Data Mining Market Basket Analysis Terhadap Data Penjualan Produk Pada Toko Iblite...

June 2024

Data mining is a technique to dig up new information from a data warehouse, information is seen as very important and valuable because by mastering information it is easy to achieve a goal, this makes everyone compete to obtain information, as well as in trading businesses such as the Iblite Luxury store. This store is located in Medan close to residents' houses, Sales transaction data will ... [Show full abstract] continue to grow, causing data storage to be even larger. Sales transaction data is only used as an archive without being properly utilized. Basically, a dataset has very useful information. Market basket analysis with a priori algorithm is one of the data mining methods that aims to find association patterns based on consumer shopping patterns, so that it can be known what items of goods are purchased in a At the same time, the results of this study found that the highest support and confidence values were Ysl and Chanel with a support value of 50% and confidence of 75%.

Article

Full-text available

Penerapan Metode Data Mining Market Basket Analysis Terhadap Data Penjualan Produk Pada Toko Dimsum...

July 2024 · Jurnal Sistem Informasi dan Ilmu Komputer

Data mining is a technique for extracting new information from data warehouses, information is seen as very important and valuable because by mastering information it is easy to achieve a goal, this makes everyone compete to obtain information, as is the case with the Dimsum business at Dimsum Madani.toko. This is located on Jalan Lampu gg. Pelita 4, Brayan Bengkel, East Medan, the location is ... [Show full abstract] close to many Brayan Resident's Houses. This of course affects sales levels. Increasing daily sales activity results in an accumulation of sales transaction data that continues to increase, thereby burdening data storage. Unfortunately, this data is only stored without further processing. In fact, this data collection holds valuable information.This research uses Market Basket Analysis with the Apriori Algorithm to find association patterns based on consumer shopping behavior. The goal is to identify items that are often purchased together. The research results showed that the combination of Seaweed Dimsum with Tofu Skin Spring Rolls had the highest support value (50%) and the highest confidence (75%).

Article

Full-text available

Penerapan Metode Data Mining Market Basket Analysis Terhadap Data Penjualan Produk Pada Toko Tas Bra...

November 2023 · Jurnal Sistem Informasi dan Ilmu Komputer

Data mining is a technique to extract new information from the data warehouse, information is considered very important and valuable because by mastering the information so easily to achieve a goal, this makes everyone competing to obtain information, as well as on trading businesses such as bag store BRANDED. store is located close to the home of the population, this certainly affects the level ... [Show full abstract] of sales, with the daily sales activities, sales transaction data will continue to grow, causing data storage is greater. Sales transaction data is only used as an archive without being put to good use. Basically the data set has very useful information. The analysis of market basket with Apriori Algorithm is one method of data mining which aims to find the pattern of association based on consumer spending pattern, so that it can be known what items are purchased simultaneously. The result of this research found that the highest support and confidence value is Ysl and Chanel with a support value of 50% and confidence of 75%.

Article

Full-text available

Mining Market Basket Data Using Share Measures and Characterized Itemsets

May 2001

We propose the share-confidence framework for knowledge discovery from databases which addresses the problem of mining itemsets from market basket data. Our goal is two-fold: (1) to present new itemset measures which are practical and useful alternatives to the commonly used support measure; (2) to not only discover the buying patterns of customers, but also to discover customer profiles by ... [Show full abstract] partitioning customers into distinct classes. We present a new algorithm for classifying itemsets based upon characteristic attributes extracted from census or lifestyle data. Our algorithm combines the Apriori algorithm for discovering association rules between items in large databases, and the AOG algorithm for attribute-oriented generalization in large databases. We suggest how characterized itemsets can be generalized according to concept hierarchies associated with the characteristic attributes. Finally, we present experimental results that demonstrate the utility of the shareconfidence framework. 1