Decision templates (DT) are a technique for classifier fusion for continuous-valued individual classifier outputs. The individual outputs considered here sum up to the same value (e.g., statistical classifiers, yielding some estimates of the posterior probabilities for the classes). First, the DT fusion algorithm is explained. Second, we show that two similarity measures (S1 and S2) and two inc...

The article deals with the problem of noise effect on semi-supervised learning. goal this is to analyze impact accuracy binary classification models created using three learning algorithms, namely Simple Recycled Selection, Incrementally Reinforced and Hybrid Algorithm, Support Vector Machines build a base classifier. Different algorithms compute similarity matrices, Radial Bias Function, Cosin...

The present paper describes exploratory work in which a Sparse Distributed Memory (SDM) was applied as a text classifier. The SDM is a type of associative memory based on the properties of highdimensional spaces, where data are stored based on pattern similarity. The results obtained with the SDM are surprisingly good, for they can be achieved with little or none pre-processing. They are far su...

This paper considers the problem of document-level multi-way sentiment detection, proposing a hierarchical classifier algorithm that accounts for the inter-class similarity of tagged sentiment-bearing texts. This type of classifier also provides a natural mechanism for reducing the feature space of the problem. Our results show that this approach improves on state-of-the-art predictive performa...

In this paper, we present a new inductive learning method for multilabel text categorization. The proposed method uses a mutual information measure to select terms and constructs document descriptor vectors for each category based on these terms. These document descriptor vectors form a document descriptor matrix. It also uses the document descriptor vectors to construct a document-similarity m...

Case retrieval constitutes an interesting area of research which contributes to the evolution of several domains. The similarity measure module is a fundamental step in the retrieval process which affects remarkably on a retrieval system. In this context, we suggest in this paper a similarity measure applied to brain tumor cases retrieval. The rationale behind the proposed measure consists in q...

We consider an online learning problem (classification or prediction) involving disparate sources of sequentially arriving data, whereby a user over time learns the best set of data sources to use in constructing the classifier by exploiting their similarity. We first show that, when (1) the similarity information among data sources is known, and (2) data from different sources can be acquired ...

MOTIVATION Structural similarity search among small molecules is a standard tool used in molecular classification and in-silico drug discovery. The effectiveness of this general approach depends on how well the following problems are addressed. The notion of similarity should be chosen for providing the highest level of discrimination of compounds wrt the bioactivity of interest. The data struc...

In object recognition problems a two-stage system is usually adopted composed of a fast and simple detector and a more complex classifier. This paper studies a design of the second stage classifier based on the recently proposed trainable similarity measure which is specifically designed for supervised classification of images. Common global measures such as correlation suffer from uninformativ...

In this submission, audio features that approximate timbre, rhythm and tempo are used for genre classification and music similarity estimation. This abstract describes the feature set, distance computation method, and classifier model used for the submitted algorithms.