Research paper

CUSTOMER SEGMENTATION ANALYSIS OF ELECTRONIC GOODS

A MACHINE LEARNING APPROACH - Machine Learning and Optimization

Abstract

There seems to be a lot of competition among competing firms to attract new customers and hold on to existing ones as a result of the creation of numerous competitors and entrepreneurs. Because of the above, providing excellent customer service is necessary regardless of the size of the company. Furthermore, the capacity of any company to comprehend the requirements of each of its clients will enhance client assistance in offering focused client services and creating unique client care strategies. Structured customer service makes it possible to have this understanding. Customers in each segment are similar in terms of market characteristics.

Introduction

Data mining algorithms are now frequently used to uncover vital and strategic information that is concealed in organizational data as a result of growing company competition and the accessibility of large-scale historical data. (Rajaraman, 2011) Data mining is the process of removing logical information from a dataset and presenting it for decision support in a way that is easily understandable by humans. Data mining techniques set apart disciplines like statistics, AI, machine learning, and data systems. Applications for data mining range from biology to weather forecasting, fraud detection, financial research, and customer insights, among others. The main goal of this article is to use data mining to find client segments in a commercial enterprise.

These differences are based on factors that directly or indirectly affect the market or business such as product preferences or expectations, location, behavior and so on. The importance of customer segmentation includes, inter alia, the ability of a business to customize market plans that would be appropriate for each segment of its customers.Support for business decisions based on risky environments such as credit relationships with its customers (Rogers, 2016); Identify products related to individual components and how to manage demand and supply power; Interdependence and interaction between consumers, between products, or between customers and products are revealed, which the business may not be aware of; The ability to predict customer declines, and which customers are likely to have problems and raise other market research questions and provide clues to find solutions. Buried in a database of integrated data proved to be effective for detecting subtle but subtle patterns or relationships.

This mode of learning is classified under supervised learning. Integration algorithms include the KMeans algorithm, K-nearest algorithm, sorting map (SOM), and more (Morrisson, 2014). These algorithms, without prior knowledge of the data, are able to identify groups in them by repeatedly comparing input patterns, as long as static aptitude in training examples is achieved based on subject matter or process. Each set has data points that have very close similarities but differ greatly from the data points of other groups.

Results:

Data frame output

Cash Advance vs Balance

Customer ID vs Balance Frequency

Balance vs Purchases Observation

Installment Purchase Vs Cash Advance

Purchases Vs Cash Advance Frequency

Credit Limit Vs Payments

Machine Learning K means Clustering Output:

Conclusion

Regardless of the information provided, the results provide a practical opportunity.For retailers to carry out marketing campaigns or similar segmentation consumers. Despite the usefulness of the status quo analysis, there are many ways to improve and grow. While there was a motive to maintain it. The number of functions is small, and another function is added in consideration of timeliness. Consumer percentages provide a clearer indicator of whether a particular purchase decision has been made. Profiles are more common now than the store's past. In a similar way find a way to cluster your customers faster (for example, one or two visits are more likely) (Mattson, 2014). Not only provide insights into the evolutionary aspects of (3 or more) Although it is clustering, there is also a potential outflow of customers. Same analysis using many other clustering algorithms such as k- Means clustering or deep learning provides insight into stability of cluster formation.

References:

1. Mattson, M. P. (2014), ‘Superior pattern processing is the essence of the evolved human brain’, Frontiers in Neuroscience 8, 265.

2. Morrison, C., Gruenewald, P. J., Freisthler, B., Ponicki, W. R. & Remer, L. G. (2014), ‘The economic geography of medical cannabis dispensaries in california’, International Journal of Drug Policy 25(3), 508 – 515. URL: http://www.sciencedirect.com/science/article/pii/S0955395913002387

3. Rajaraman, A. & Ullman, J. D. (2011), Mining of Massive Datasets, Cambridge University Press, New York, NY, USA.

4. Rogers, S. & Girolami, M. (2016), A First Course in Machine Learning, Second Edition, Chapman & Hall/CRC