Blog tags are usually considered to be supplementary information for blog post classification tasks. Due to the sparsity of tag features, improving performance of classifiers merely using tags is not a trivial operation. This paper presents a blog post classification approach based on the tag propagation strategy. Using a dataset of blog posts gleaned from the Internet, tags of a blog post are ...

In many pattern classification problems, an estimate of the posterior probabilities (rather than only a classification) is required. This is usually the case when some confidence measure in the classification is needed. In this article, we propose a new posterior probability estimator. The proposed estimator considers the K-nearest neighbors. It attaches a weight to each neighbor that contribut...

In this paper, a new camera based lumber strength classification solution using solely knot features is presented. Two alternative classifiers, k-NN and SVM, are applied to classifing pinewood boards based on their breaking strength. Features for classification are formed using knot properties, which are extracted from four sides of the board using machine vision algorithms. Extracted propertie...

Disc herniation is a major reason for lower back pain (LBP), it cost the United Kingdom (UK) government over £1.3 million per day. In fact a very high proportion of the UK population will complain from their back pain. Furthermore, Magnetic Resonance Imaging (MRI) is one of the main diagnosing procedure for LBP. Automatic disc labeling in the MRI to detect the herniation area will reduce the re...

Telecommunications Research Center Vienna (ftw.), Donaucitystr. 1, A-1220 Vienna, Austria IBK, Vienna University of Technology, Favoritenstr. 9/388, A-1040 Vienna, Austria INTHFT, Vienna University of Technology, Gusshausstr. 25-29/389, A-1040 Vienna, Austria Mobilkom Austria AG & Co KG, Obere Donaustr. 29, A-1020 Vienna, Austria Alcatel Austria AG, Scheydgasse 41, A-1210 Vienna, Austria Kapsch...

The task of automated searching for interesting text documents frequently suffers from a very poor balance among documents representing both positive and negative examples or from one completely missing class. This paper suggests the ranking approach based on the k-NN algorithm adapted for determining the similarity degree of new documents just to the representative positive collection. From th...

The main purpose of this paper is to use off-the-shelf devices to develop a fall detection system. In human body identification, human body silhouette is adopted to improve privacy protection, and vertical projection histograms of the silhouette image and statistical scheme are used to reduce the effect of human body upper limb activities. The kNN classification algorithm is used to classify th...

This paper presents an interval type-2 fuzzy Knearest neighbor (NN) algorithm that is an extension of the type1 fuzzy K-NN algorithm proposed in [l]. In our proposed method, the membership values for each vector are 'extended as interval type-2 fuzzy memberships by assigning uncertainty to the type-1 memberships. By doing so, the classification result obtained by the interval type-2 fuzzy K-NN ...

This paper presents a novel approach to the analysis of the overtraining phenomenon in pattern classiiers. A nonlinearity measure N is introduced which relates the shape of the classiication function to the generalization capability of a classiier. Experiments using the k-nearest neighbour rule, a neural classiier and the quadratic classiier show that the introduced measure N can be used to stu...

In this paper, we propose an analytical and methodological comparison between two of the most known distance-based methods in the evaluation of the geographic concentration of economic activity. These two methods are Ripley’s K function, a cumulative function popularised by Marcon and Puech (2003) that counts the average number of neighbours of each point within a circle of a given radius, and ...