Serkan Gunal

A novel probabilistic feature selection method for text classification

High dimensionality of the feature space is one of the most important concerns in text classification problems due to processing time and accuracy considerations. Selection of distinctive features is therefore essential for text classification. This study proposes a novel filter based probabilistic feature selection method, namely distinguishing feature selector (DFS), for text classification. The proposed method is compared with well-known filter approaches including chi square, information gain, Gini index and deviation from Poisson distribution. The comparison is carried out for different datasets, classification algorithms, and success measures. Experimental results explicitly indicate that DFS offers a competitive performance with respect to the abovementioned approaches in terms of classification accuracy, dimension reduction rate and processing time.

The impact of preprocessing on text classification

Preprocessing is one of the key components in a typical text classification framework. This paper aims to extensively examine the impact of preprocessing on text classification in terms of various aspects such as classification accuracy, text domain, text language, and dimension reduction. For this purpose, all possible combinations of widely used preprocessing tasks are comparatively evaluated on two different domains, namely e-mail and news, and in two different languages, namely Turkish and English. In this way, contribution of the preprocessing tasks to classification success at various feature dimensions, possible interactions among these tasks, and also dependency of these tasks to the respective languages and domains are comprehensively assessed. Experimental analysis on benchmark datasets reveals that choosing appropriate combinations of preprocessing tasks, rather than enabling or disabling them all, may provide significant improvement on classification accuracy depending on the domain and language studied on.

Subspace based feature selection for pattern recognition

Feature selection is an essential topic in the field of pattern recognition. The feature selection strategy has a direct influence on the accuracy and processing time of pattern recognition applications. Features can be evaluated with either univariate approaches, which examine features individually, or multivariate approaches, which consider possible feature correlations and examine features as a group. Although univariate approaches do not take the correlation among features into consideration, they can provide the individual discriminatory power of the features, and they are also much faster than multivariate approaches. Since it is crucial to know which features are more or less informative in certain pattern recognition applications, univariate approaches are more useful in these cases. This paper therefore proposes subspace based separability measures to determine the individual discriminatory power of the features. These measures are then employed to sort and select features in a multi-class manner. The feature selection performances of the proposed measures are evaluated and compared with the univariate forms of classic separability measures (Divergence, Bhattacharyya, Transformed Divergence, and Jeffries-Matusita) on several datasets. The experimental results clearly indicate that the new measures yield comparable or even better performance than the classic ones in terms of classification accuracy and dimension reduction rate.

Text classification using genetic algorithm oriented latent semantic features

Abstract In this paper, genetic algorithm oriented latent semantic features (GALSF) are proposed to obtain better representation of documents in text classification. The proposed approach consists of feature selection and feature transformation stages. The first stage is carried out using the state-of-the-art filter-based methods. The second stage employs latent semantic indexing (LSI) empowered by genetic algorithm such that a better projection is attained using appropriate singular vectors, which are not limited to the ones corresponding to the largest singular values, unlike standard LSI approach. In this way, the singular vectors with small singular values may also be used for projection whereas the vectors with large singular values may be eliminated as well to obtain better discrimination. Experimental results demonstrate that GALSF outperforms both LSI and filter-based feature selection methods on benchmark datasets for various feature dimensions.

A comparative study on machine learning algorithms for indoor positioning

Fingerprinting based positioning is commonly used for indoor positioning. In this method, initially a radio map is created using Received Signal Strength (RSS) values that are measured from predefined reference points. During the positioning, the best match between the observed RSS values and existing RSS values in the radio map is established as the predicted position. In the positioning literature, machine learning algorithms have widespread usage in estimating positions. One of the main problems in indoor positioning systems is to find out appropriate machine learning algorithm. In this paper, selected machine learning algorithms are compared in terms of positioning accuracy and computation time. In the experiments, UJIIndoorLoc indoor positioning database is used. Experimental results reveal that k-Nearest Neighbor (k-NN) algorithm is the most suitable one during the positioning. Additionally, ensemble algorithms such as AdaBoost and Bagging are applied to improve the decision tree classifier performance nearly same as k-NN that is resulted as the best classifier for indoor positioning.

A novel framework for SMS spam filtering

A novel framework for SMS spam filtering is introduced in this paper to prevent mobile phone users from unsolicited SMS messages. The framework makes use of two distinct feature selection approaches based on information gain and chi-square metrics to find out discriminative features representing SMS messages. The discriminative feature subsets are then employed in two different Bayesian-based classifiers, so that SMS messages are categorized as either spam or legitimate. Moreover, the paper introduces a real-time mobile application for Android™ based mobile phones utilizing the proposed spam filtering scheme, as well. Hence, SMS spam messages are silently filtered out without disturbing phone users. Effectiveness of the filtering framework is evaluated on a large SMS message collection including legitimate and spam messages. Following the evaluation, remarkably accurate classification results are obtained for both spam and legitimate messages.

On circular traffic sign detection and recognition

New methods are proposed for circular traffic sign detection and recognition.Comparable performances are attained with respect to the best performing methods.Compatibility to real-time operation is validated. Automatic traffic sign detection and recognition play crucial roles in several expert systems such as driver assistance and autonomous driving systems. In this work, novel approaches for circular traffic sign detection and recognition on color images are proposed. In traffic sign detection, a new approach, which utilizes a recently developed circle detection algorithm and an RGB-based color thresholding technique, is proposed. In traffic sign recognition, an ensemble of features including histogram of oriented gradients, local binary patterns and Gabor features are employed within a support vector machine classification framework. Performances of the proposed detection and recognition approaches are evaluated on German Traffic Sign Detection and Recognition Benchmark datasets, respectively. The results of the experimental work reveal that both approaches offer comparable or even better performances with respect to the best ones reported in the literature and are compatible to real-time operation as well.

A Low-Computational Approach on Gaze Estimation With Eye Touch System

Among various approaches to eye tracking systems, light-reflection based systems with non-imaging sensors, e.g., photodiodes or phototransistors, are known to have relatively low complexity; yet, they provide moderately accurate estimation of the point of gaze. In this paper, a low-computational approach on gaze estimation is proposed using the Eye Touch system, which is a light-reflection based eye tracking system, previously introduced by the authors. Based on the physical implementation of Eye Touch, the sensor measurements are now utilized in low-computational least-squares algorithms to estimate arbitrary gaze directions, unlike the existing light reflection-based systems, including the initial Eye Touch implementation, where only limited predefined regions were distinguished. The system also utilizes an effective pattern classification algorithm to be able to perform left, right, and double clicks based on respective eye winks with significantly high accuracy. In order to avoid accuracy problems for sensitive sensor biasing hardware, a robust custom microcontroller-based data acquisition system is developed. Consequently, the physical size and cost of the overall Eye Touch system are considerably reduced while the power efficiency is improved. The results of the experimental analysis over numerous subjects clearly indicate that the proposed eye tracking system can classify eye winks with 98% accuracy, and attain an accurate gaze direction with an average angular error of about 0.93 °. Due to its lightweight structure, competitive accuracy and low-computational requirements relative to video-based eye tracking systems, the proposed system is a promising human-computer interface for both stationary and mobile eye tracking applications.

Use of Novel Feature Extraction Technique with Subspace Classifiers for Speech Recognition

Speech recognition is one of the fast moving research areas in pervasive services requiring human interaction. Like any type of pattern recognition system, selection of the feature extraction method and the classifier play a crucial role for speech recognition in terms of accuracy and speed. In this paper, an efficient wavelet based feature extraction method for speech data is presented. The feature vectors are then fed into three widely used linear subspace classifiers for recognition analysis. These classifiers are Class Featuring Information Compression (CLAFIC), Multiple Similarity Method (MSM) and Common Vector Approach (CVA). TI-DIGIT database is used to evaluate the performance of speaker independent isolated word recognition system designed. Experimental results indicate that the proposed feature extraction method together with the CLAFIC and CVA classifiers give considerably high recognition rates.

ECG based biometric authentication using ensemble of features

In this work, the efficacy of various features on electrocardiogram (ECG) based biometric authentication process is thoroughly examined. In particular, the features acquired from temporal analysis, wavelet transformation, power spectral density estimation and QRS-complex detection over ECG signals are considered. These features are employed with two distinct classification algorithms, namely decision tree and Bayes network, specifically for gender, age and identity recognition problems. The biometric authentication framework is evaluated on a benchmark dataset that contains ECG records of 18 healthy people including 5 men, aged 26 to 45, and 13 women, aged 20 to 50. The results of the experimental analysis reveal that if all those features are used in combination rather than individually, better performance is attained for all classifiers in each recognition problem.