Development of Industrial Knowledge Management Applications with Case-Based Reasoning

Studies in Computational Intelligence: Successful Case-Based Reasoning Applications – 1, Volume 305/2 010. P 53-82. Springer
Development of Industrial Knowledge Management Applications with Case-Based Reasoning
Mehmet H.Goker, Catherine Baudin, Michel Manago, Mehmet H.Goker, Catherine Baudin, Michel Manago

The successful development, deployment and utilization of Case-Based Reasoning Systems in commercial environments require the development team to focus on aspects that go beyond the core CBR engine itself. Characteristics of the Users, the Organization and the Domain have considerable impact on the design decisions during implementation and on the success of the project after deployment.

If the system is not technically and organizationally integrated with the operating environment, it will eventually fail. In this chapter, we describe our experiences and the steps we found useful while implementing CBR applications for commercial use. We learned these lessons the hard way. Our goal is to document our experience and help practitioners develop their own approach and avoid making the same mistakes.

Another publication from the same category: Machine Learning and Data Science

IEEE Computing Conference 2018, London, UK

Regularization of the Kernel Matrix via Covariance Matrix Shrinkage Estimation

The kernel trick concept, formulated as an inner product in a feature space, facilitates powerful extensions to many well-known algorithms. While the kernel matrix involves inner products in the feature space, the sample covariance matrix of the data requires outer products. Therefore, their spectral properties are tightly connected. This allows us to examine the kernel matrix through the sample covariance matrix in the feature space and vice versa. The use of kernels often involves a large number of features, compared to the number of observations. In this scenario, the sample covariance matrix is not well-conditioned nor is it necessarily invertible, mandating a solution to the problem of estimating high-dimensional covariance matrices under small sample size conditions. We tackle this problem through the use of a shrinkage estimator that offers a compromise between the sample covariance matrix and a well-conditioned matrix (also known as the "target") with the aim of minimizing the mean-squared error (MSE). We propose a distribution-free kernel matrix regularization approach that is tuned directly from the kernel matrix, avoiding the need to address the feature space explicitly. Numerical simulations demonstrate that the proposed regularization is effective in classification tasks.