Clustering Global Entrepreneurship through Data Mining Technique

Clustering Global Entrepreneurship through Data Mining Technique

Paula Odete Fernandes (Polytechnic Institute of Bragança, Portugal) and Rui Pedro Lopes (Polytechnic Institute of Bragança, Portugal)
DOI: 10.4018/978-1-4666-8348-8.ch027
OnDemand PDF Download:
No Current Special Offers


The purpose of this chapter is to contribute for the identification of groups of countries that share similar patterns regarding the characteristics of Global Entrepreneurship and capturing features of entrepreneurship by focusing on entrepreneurial attitudes and entrepreneurial activity. For this purpose, 67 countries from 2013 GEM survey were selected, and Data Mining Methodology was used. In particular, evolutionary computation is used to determine a finite set of categories to describe the data set according to multi-dimensional similarities among its objects. In other words, several clustering algorithms where used, to get the best categories possible. The results show four clusters with different entrepreneurial attitudes among the countries - very high, medium and low entrepreneurial attitudes and entrepreneurial activities.
Chapter Preview

1. Data Mining Techniques

The way we work and live has been shaped by the advances of technology. As devices become smaller, they tend to be with us anywhere, anytime. The storage, processing capacity, autonomy has been constantly increasing, which makes us less afraid of relying on their functionality and on keeping including them in our daily routine.

These omnipresent devices make it easy to save things previously discarded. Our decisions, holiday pictures, documents, supermarket choices, walking tours are all registered in the devices and uploaded for future reference to the huge information repository in clouds everywhere. The generation of data is growing much faster than our capacity to understand it.

Key Terms in this Chapter

Global Entrepreneurship Monitor: Study that analyses the level of entrepreneurship in several world countries.

Clustering: Analysis of large set of data to provide a finite set of categories to describe similarities among its objects.

DBSCAN: Clustering algorithm to identify the number and shape of clusters.

Genetic Algorithm: Abstraction and implementation of evolutionary principles and theories in computational algorithms to search optimal solutions to a problem.

Entrepreneurship: A polysemic concept originated in French economics in the 17th and 18th century and typically associated to one who undertakes an activity or a significant project.

Data Mining: Techniques used to extract knowledge from unstructured data.

High Status Successful Entrepreneurship: The percentage 18-64 population who agree with the statement that in their country successful entrepreneurs receive high status.

Total Early-Stage Entrepreneurial Activity: The percentage of 18-64 population who are either a nascent entrepreneur or owner-manager of a new business.

Complete Chapter List

Search this Book: