No abstract

The paper approaches the task of automatically reading and recognition of registered data on the utility meters of the users and is a part of a more complex project of our team concerning the remote data acquisition from industrial processes. A huge amount of utility meters in our country is of mechanical type without remote acquiring facilities and as an intermediate solution we propose an intelligent optical acquisition system which will store the read values in desktop and mobile devices. The main requirements of such a system are: portability, data reading accuracy, fast processing and energy independence. The paper analyses several solutions (including Artificial Neural Networks approach) tested by our team and present the experimental results and our conclusions.

No abstract

A problem arises in data mining, when classifying unbalanced datasets using Support Vector Machines. Because of the uneven distribution and the soft margin of the classifier, the algorithm tries to improve the general accuracy of classifying a dataset, and in this process it might misclassify a lot of weakly represented classes, confusing their class instances as overshoot values that appear in the dataset, and thus ignoring them. This paper introduces the Enhancer, a new algorithm that improves the Cost-sensitive classification for Support Vector Machines, by multiplying in the training step the instances of the underrepresented classes. We have discovered that by oversampling the instances of the class of interest, we are helping the Support Vector Machine algorithm to overcome the soft margin. As an effect, it classifies better future instances of this class of interest.

No abstract