This paper presents an extension of the well-known Extreme Learning Machines (ELMs). The main goal is to provide probabilities as outputs for Multiclass Classification problems. Such information is more useful in practice than traditional crisp classification outputs. In summary, Gaussian Mixture Models are used as post-processing of ELMs. In that context, the proposed global methodology is kee...

Post-preprocessing is an important part in handwritten character recognition system. In postpreprocessing, the knowledge about address is used to verify the result of segmentation and ‘recognition. As a result, the best-matched output of segmentation and recognition is selected. The main problem in post-preprocessing is there are too many addresses needed to be handled, so that the processing t...

We present a new similarity measure tailored to posts in an online forum. Our measure takes into account all the available information about user interest and interaction — the content of posts, the threads in the forum, and the author of the posts. We use this post similarity to build a similarity between users, based on principal coordinate analysis. This allows easy visualization of the user...

This research paper explores the use of AdaBoost algorithm for predicting residential habitation status in aftermath Mount Merapi eruption. Using a dataset from Rehabilitation and Reconstruction Task Force, with 2516 instances 11 attributes, model was trained evaluated. The demonstrated robust performance an accuracy 92%, though it struggled correctly identifying all 'No Habited' instances. The...

We present a characterization of the approximation errors of the PostWidder and the Gamma operators in Lp(0,∞), 1 ≤ p ≤ ∞, with a weight x for any real γ. Two types of characteristics are used – weighted Kfunctionals of the approximated function itself and the classical fixed step moduli of smoothness taken on a simple modification of it. AMS classification: 41A25, 41A27, 41A35, 41A36.

This paper presents an approach to the automatic classification of diatoms based on image-processing techniques. The main contribution is the development of a pattern evaluated from a frequential analysis of images taken through microscope, post-processed to extract the relevant characteristics that distinguish the different species. The final classification is carried out using a competitive n...