One-Class Classification to Predict the Success of Private-Participation Infrastructure Projects in Europe
Foreign investment has significantly increased in infrastructure projects as governments have realized the potential advantages for the host countries in terms of capital and technical expertise. Private participation projects are a common vehicle for private firms to invest in infrastructures, although these projects are typically subject to pressures from the government, consumers, suppliers, regulatory institutions, and public opinion. Forecasting the success of these projects in advance is a key element to be taken into account when deciding about participation. To support this kind of decisions, present paper proposes the application of some one-class classifiers to check their ability to predict the final success of private participation projects involving infrastructures. To validate the proposed soft-computing models, they are applied to a real-life dataset from the World Bank, comprising information about projects in European countries within the Energy and Telecommunication sectors.
KeywordsClassification Support Vector Machines Random trees k nearest neighbors Private-participation projects Internationalization
- 8.Pinzón, C., Herrero, Á., De Paz, J.F., Corchado, E., Bajo, J.: CBRid4SQL: A CBR Intrusion Detector for SQL Injection Attacks, pp. 510–519. Springer, Heidelberg (2010)Google Scholar
- 13.Herrero, Á., Jiménez, A., Bayraktar, S.: Hybrid unsupervised exploratory plots: a case study of analysing foreign direct investment. Complexity (2019, in press)Google Scholar
- 18.Boser, B.E., Guyon, I.M., Vapnik, V.N.: A training algorithm for optimal margin classifiers. In: 5th Annual Workshop on Computational Learning Theory, pp. 144–152. ACM (1992)Google Scholar
- 21.Cunningham, P., Delany, S.J.: k-Nearest neighbour classifiers. Technical report UCD-CSI-2007-4 (2007)Google Scholar