Search results for "feature selection"

Unstable feature relevance in classification tasks

2011

Music mood annotation using semantic computing and machine learning

2015

Recognition of Cardiac Arrhythmia by Means of Beat Clustering on ECG-Holter Recordings

2012

The follow-up of some cardiac diseases may be achieved by ECG-holter record analysis. A heartbeat clustering method can be used to reduce the usually high computational cost of such Holter analysis. This study describes a method aimed at cardiac arrhythmia recognition based on this approach, by means of unsupervised inspection of morphologically similar heartbeat groups. Singular Value Decomposition (SVD) is used as the feature selection method since the complexity increases exponentially with the number of features. A modification of the k-means algorithm was developed for centroid computation, taking into account heartbeat length changes. Experimental set consisted of ECG records from the…

Feature selection for classification of music according to expressed emotion

2009

Signal processing techniques for robust sound event recognition

2019

The computational analysis of acoustic scenes is today a topic of major interest, with a growing community focused on designing machines capable of identifying and understanding the sounds produced in our environment, similar to how humans perform this task. Although these domains have not reached the industrial popularity of other related audio domains, such as speech recognition or music analysis, applications designed to identify the occurrence of sounds in a given scenario are rapidly increasing. These applications are usually limited to a set of sound classes, which must be defined beforehand. In order to train sound classification models, representative sets of sound events are record…

A comparison between two feature selection algorithms

2017

This article provides a comparison of two feature selection algorithms, Information Gain Thresholding and Koller and Sahami's algorithm in the context of text document classification on the Reuters Corpus Volume 1 dataset. The algorithms were evaluated by testing the performance of classifiers trained on the features they select from a given dataset. Results show that Koller and Sahami's algorithm consistently outperforms Information Gain Thresholding by capturing interactions between features and avoiding redundancy among features, although it achieves its gains through increased complexity and longer running time.

Generalizability and Simplicity as Criteria in Feature Selection: Application to Mood Classification in Music

2011

Classification of musical audio signals according to expressed mood or emotion has evident applications to content-based music retrieval in large databases. Wrapper selection is a dimension reduction method that has been proposed for improving classification performance. However, the technique is prone to lead to overfitting of the training data, which decreases the generalizability of the obtained results. We claim that previous attempts to apply wrapper selection in the field of music information retrieval (MIR) have led to disputable conclusions about the used methods due to inadequate analysis frameworks, indicative of overfitting, and biased results. This paper presents a framework bas…

Coupling high-resolution satellite imagery with ALS-based canopy height model and digital elevation model in object-based boreal forest habitat type …

2014

We developed a classification workflow for boreal forest habitat type mapping. In object-based image analysis framework, Fractal Net Evolution Approach segmentation was combined with random forest classification. High-resolution WorldView-2 imagery was coupled with ALS based canopy height model and digital terrain model. We calculated several features (e.g. spectral, textural and topographic) per image object from the used datasets. We tested different feature set alternatives; a classification accuracy of 78.0 % was obtained when all features were used. The highest classification accuracy (79.1 %) was obtained when the amount of features was reduced from the initial 328 to the 100 most imp…

Anomaly detection using one-class SVM with wavelet packet decomposition

2011

Anomaly detection has become a popular research topic in the field of machine learning. Support vector machine is one anomaly detection technique and it is coming one the most widely used. In this research, anomaly detection is applied to road condition monitoring, especially pothole detection, using accelerometer data. The proposed concept includes data preprocessing, feature extraction, feature selection and classification. Accelerometer data was first filtered and segmented, after which features were extracted with frequency- and time-domain functions, with genetic programming and with wavelet packet decomposition. A classification model was built using support vector machine and the cal…