Quantitative Analysis of the Romanian Private Security Market. A Machine Learning Approach

Alexandru-Costin Băroiu ()
Additional contact information
Alexandru-Costin Băroiu: Bucharest University of Economic Studies

A chapter in Digital Economy and the Green Revolution, 2023, pp 1-15 from Springer

Abstract: Abstract Market segmentation and analysis have benefited greatly from advancements in Machine Learning. Supervised and unsupervised learning techniques have been applied with great success in market analysis. This paper proposes such an approach that aims to first introduce a new dataset and identify the groups in which the market is segmented, by applying k-means++ clustering, and then to develop a well performing classifier that would correctly identify future companies and place them in the previously identified clusters for a disregarded industry, the Romanian private security market. First, a clustering algorithm is applied to group the companies into clusters. Then, the results are analyzed and findings are discussed about the market segmentation. 6 distinct groups are identified and the main factors that differentiate the companies are number of employees and turnover. Second, multiple classification algorithms are trained and benchmarked in order to find the best performing model. Due to the nature of the data, which is heavily imbalanced, sampling algorithms and weights adjustments are applied in order to improve model performance. Lastly, the best combination of classification algorithm and sampling technique is presented for the given data. The best performing model is a Multi-Layer Perceptron network, with an average F-score of 71.8. The best performing technique used to counter imbalanced data is ADASYN, with an average F-score of 69.9. The best performing combination and the best result overall is achieved by the Multi-Layer Perceptron in tandem with Random Oversampling, with an F-score of 83.3.

Keywords: Market segmentation; Market analysis; Machine learning (search for similar items in EconPapers)
Date: 2023
References: Add references at CitEc
Citations:

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:prbchp:978-3-031-19886-1_1

Ordering information: This item can be ordered from
http://www.springer.com/9783031198861

DOI: 10.1007/978-3-031-19886-1_1

Access Statistics for this chapter

More chapters in Springer Proceedings in Business and Economics from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().