Detecting and quantifying ambiguity: a neural network approach
In general, it is not possible to have access to all variables that determine the behavior of a system. Once a number of measurable variables is identified, there might still exist hidden variables which influence the behavior of the system. The result is model ambiguity in the sense that, for the same (or very similar) input values, distinct outputs are obtained. In addition, the degree of ambiguity may vary across the range of input values. Therefore, to evaluate the accuracy of a model it is important to devise a method to obtain the degree of reliability for each output result. In this paper, we present such a scheme composed of two coupled neural networks, the first one computing the average predicted value and the other the reliability of the output, which is learned from the error values of the first one. As an illustration, the scheme is applied to a model for tracking slopes in a straw chamber and to a credit scoring model.
KeywordsUncertainty Ambiguity Neural networks
This study was funded by Fundação para a Ciência e Tecnologia, Portugal.
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
This article does not contain any studies with human participants performed by any of the authors.
- Albalate A, Suchindranath A, Soenmez MM, Suendermann D (2010) On ambiguity detection and postprocessing schemes using cluster ensembles, ICAART 2010. In: Proceedings of the international conference on agents and artificial intelligence. INSTICC Press, pp 623–630Google Scholar
- Bailey-Kellogg C, Ramakrishnan N (2001) Ambiguity-directed sampling for qualitative analysis of sparse data from spatially-distributed physical systems. In: Proceedings of the IJCAIGoogle Scholar
- Bartlett PL, Maass W (2003) Vapnik–Chervonenkis dimension of neural nets. In: Arbib MA (ed) The handbook of brain theory and neural networks. MIT Press, Cambridge, pp 1188–1192Google Scholar
- Clarke CLA, Kolla M, Vechtomova O (2009) An effectiveness measure for ambiguous and underspecified queries. In: Advances in information retrieval theory, lecture notes in computer science, vol 5766, pp 188–199Google Scholar
- dos Santos E, Sabourin R, Maupin P (2007) Ambiguity-guided dynamic selection of ensemble of classifiers. In: 10th international conference on information fusionGoogle Scholar
- Denby B, Lessner E, Lindsey CS (1990) Test of track segment and vertex finding with neural networks. In: Proceedings of the 1990 conference on computing in high energy physics, Sante Fe, NM. AIP conference proceedings, vol 209, p 211Google Scholar
- Lin Y-M, Wang X, Ng WWY, Chang Q, Yeung DS, Wang X-L (2006) Sphere classification for ambiguous data. In: 2006 International conference on machine learning and cybernetics, IEEE conference publications, pp 2571–2574Google Scholar
- Martins J, Vilela Mendes R (2001) Neural networks and logical reasoning systems. A translation table. Int J Neural Syst 11:179–186Google Scholar
- Pacelli V, Azzollini M (2011) An artificial neural network approach for credit risk management. J Intell Learn Syst Appl 3:103–112Google Scholar
- Roul RK, Sahay SK (2012) An effective information retrieval for ambiguous query. Asian J Comput Sci Inf Technol 2:26–30Google Scholar
- Svendsen JGG (2003) A SearchFree approach to ambiguity resolution. In: Proceedings of the 16th international technical meeting of the satellite Division of the Institute of Navigation. Portland, pp 769–774Google Scholar
- UCI Machine Learning Repository. http://archive.ics.uci.edu/ml/
- Van Gestel T, Baesens B (2009) Credit risk management: basic concepts: financial risk components. Rating analysis, models, economic and regulatory capital. Oxford University Press, OxfordGoogle Scholar
- Wong SKM, Wang ZW (1993) Qualitative measures of ambiguity. In: Proceedings of the 9th conference on uncertainty in artificial intelligence, pp 443–450Google Scholar
- IEEE (1993) Proceedings of second international symposium on uncertainty modeling and analysis. In: IEEE conference publications. doi: 10.1109/ISUMA.1993.366801