Sharif Digital Repository

Voluminous amount of videos in today’s world has made the subject of objective (or semi-objective) classification of videos to be very popular. Among the various descriptors used for video classification, SIFT and LIFT can lead to highly accurate classifiers. But, SIFT descriptor does not consider video motion and LIFT is time-consuming. In this paper, a robust descriptor for semi-supervised classification based on video content is proposed. It holds the benefits of LIFT and SIFT descriptors and overcomes their shortcomings to some extent. For extracting this descriptor, the SIFT descriptor is first used and the motion of the extracted keypoints are then employed to improve the accuracy of... 

Adaptive application detection in today's high-bandwidth networks is resource consuming and inaccurate due to the high volume, velocity, and variety characteristics of the networks traffic. To generate a robust classifier for identifying applications over encrypted traffic, we proposed DSCA as a DPI-based Stream Classification Algorithm. DSCA utilizes applications detected by the DPI, Deep Packet Inspection technique, as ground truth data and updates the classification model accordingly. To reduce the classification algorithms overhead without accuracy reduction, a feature selection method, named CfsSubsetEval, is deployed in DSCA. The proposed approach is implemented via the MOA tool and... 

Machine vision systems are among the novel tools proven to be useful in different applications, among which monitoring and controlling manufacturing processes is one of the most important ones. However, due to the complexity resulted from high-dimensional image data and their inherent correlations, the acquisition of traditional statistical process control tools seems inapplicable. To overcome the shortcomings of the traditional methods in this regard, a statistical model is proposed in this paper which utilizes the concepts of both the PCA-based T2 control chart and the classification methods to develop a tool capable of controlling an image-based process. By defining the warning zones,... 

Fuzzy rule-based classifiers are one of the famous forms of the classification systems particularly in the data mining field. Genetic algorithm is a useful technique for discovering this kind of classifiers and it has been used for this purpose in some studies. In this paper, we propose a new symbiotic evolutionary approach to find desired fuzzy rulebased classifiers. For this purpose, a symbiotic combination operator has been designed as an alternative to the recombination operator (crossover) in the genetic algorithms. In the proposed approach, the evolution starts from simple chromosomes and the structure of chromosomes gets complex gradually during the evolutionary process. Experimental... 

Segmentation of aerial images has been a challenging task in recent years. This paper introduces a simple and efficient method for segmentation and classification of aerial images based on a pixel-level classification. To this end, we use the Gabor texture features in HSV color space as our best experienced features for aerial images segmentation and classification. We test different classifiers including KNN, SVM and a classifier based on sparse representation. Comparison of our proposed method with a sample of segmentation pre-process based classification methods shows that our pixel-wise approach results in higher accuracy results with lower computation time  

Classification of data streams has become an important area of data mining, as the number of applications facing these challenges increases. In this paper, we propose a new ensemble learning method for data stream classification in presence of concept drift. Our method is capable of detecting changes and adapting to new concepts which appears in the stream  

This paper presents a new interpretable fuzzy classification system. Simulated annealing heuristic is employed to effectively investigate the large search space usually associated with classification problem. Here, two criteria are used to evaluate the proposed method. The first criterion is accuracy of extracted fuzzy if-then rules, and the other is comprehensibility of obtained rules. Experiments are performed with some data sets from UCI machine learning repository. Results are compared with several well-known classification algorithms, and show that the proposed approach provides more accurate and interpretable classification system. © 2009 IEEE  

In this paper, a new feature set is presented and evaluated based on sinusoidal modeling of audio signals. Duration of the longest sinusoidal model frequency track, as a measure of the harmony, is used and compared to typical features as input into an audio classifier. The performance of this sinusoidal model feature is evaluated through classification of audio to speech and music using both the GMM and the SVM classifiers. Classification results show the proposed feature, which could be used for the first time in such an audio classification, is quite successful in speech/music classification. Experimental comparisons with popular features for audio classification, such as HZCRR and LSTER,... 

The purpose of this paper is to study and review some major impacts of the disasters during the past 105 years and develop a new theoretical classification of disasters. Design/methodology/approach– A detailed study of disasters in the world during the period (1900‐2005) has been obtained from the recent published sources. In that period more than 40 lessons have been reported based on statistical data analysis of disasters. Furthermore, a two‐dimensional probability density function is developed to categorize the different types of disasters. This paper studies and reviews some major impacts of disasters during the past 105 years and summarizes some major lessons for the future.... 

Efficient storage and handling of data stored in a peer-to-peer (P2P) network, proves vital for various applications such as query processing and data mining. This paper presents a distributed, scalable and robust layered overlay (L-overlay) to index and manage multidimensional data in a dynamic P2P network. The proposed method distinguishes between the data and peer layers, with efficient mapping between the two. The data is organized such that semantically similar data objects are accessed hastily. Grid and tree structures are proposed for the peer layer. As application examples of L-overlay in query processing and data mining, k-nearest neighbors query processing and distributed Naïve... 

Classifying streams of data, for instance financial transactions or emails, is an essential element in applications such as online advertising and spam or fraud detection. The data stream is often large or even unbounded; furthermore, the stream is in many instances non-stationary. Therefore, an adaptive approach is required that can manage concept drift in an online fashion. This paper presents a probabilistic non-parametric generative model for stream classification that can handle concept drift efficiently and adjust its complexity over time. Unlike recent methods, the proposed model handles concept drift by adapting data-concept association without unnecessary i.i.d. assumption among the... 

Incremental learning has been used extensively for data stream classification. Most attention on the data stream classification paid on non-evolutionary methods. In this paper, we introduce new incremental learning algorithms based on harmony search. We first propose a new classification algorithm for the classification of batch data called harmony-based classifier and then give its incremental version for classification of data streams called incremental harmony-based classif ier. Finally, we improve it to reduce its computational overhead in absence of drifts and increase its robustness in presence of noise. This improved version is called improved incremental harmony-based classifier. The... 

With the extensive growth in information, text classification as one of the text mining methods, plays a vital role in organizing and management information. Most text classification methods represent a documents collection as a Bag of Words (BOW) model and then use the histogram of words as the classification features. But in this way, the number of features is very large; therefore performing text classification faces serious computational cost problems. Moreover, the BOW representation is unable to recognize semantic relations between words. Recently, topic-model approaches have been successfully applied for text classification to overcome the problems of BOW. Our main goal in this paper... 

The widespread expansion of the semiconductor industry and various production phases have led to the increased importance of fabricating highly secure chips. Both in factories manufacturing and later at actual operation, digital integrated circuits (IC) might encounter a variety of hardware attacks, one type of which involves Hardware Trojans (HT). Due to their diversity, it has become a major hardware security challenge to prevent, detect and track down HTs. In this regard, the first step is to understand the taxonomy of Trojans and the current ways in which they can be encountered. For that purpose, certain classifications are required. With their downsized dimensions, the Trojans have... 

This paper proposes a new method based on Particle Swarm Optimization (PSO) for feature selection in phonemes sound classification. Inspired of biologist's studies, each particle is represented by filterbank which is motivated by human hearing. Thus, we propose a technique in which PSO is used to extract audio features similar to human's ear in order to achieve better classification. We use PSO technique for optimizing particle's filterbank in order to classify sound signals accurately. Then, feature extraction is done by using particle's information. Moreover, a classification method based on nearest neighbor is used. Furthermore, by using a defined fitness function in this paper, the... 

Background and objective: One of the important applications of non-invasive respiration monitoring using ECG signal is the detection of obstructive sleep apnea (OSA). ECG-derived respiratory (EDR) signals, contribute to useful information about apnea occurrence. In this paper, two EDR extraction methods are proposed, and their application in automatic OSA detection using single-lead ECG is investigated. Methods: EDR signals are extracted based on new respiration-related features in ECG beats morphology, such as ECG variance (EDRVar) and phase space reconstruction area (EDRPSR). After evaluating the EDRs by comparing them to a reference respiratory signal, they are used in an automatic OSA... 

Common similarity measures of time domain signals such as cross-correlation and Symbolic Aggregate approximation (SAX) are not appropriate for nonlinear signal analysis. This is because of the high sensitivity of nonlinear systems to initial points. Therefore, a similarity measure for nonlinear signal analysis must be invariant to initial points and quantify the similarity by considering the main dynamics of signals. The statistical behavior of local extrema (SBLE) method was previously proposed to address this problem. The SBLE similarity index uses quantized amplitudes of local extrema to quantify the dynamical similarity of signals by considering patterns of sequential local extrema. By... 

Purpose - The purpose of this paper is to study and review some major impacts of the disasters during the past 105 years and develop a new theoretical classification of disasters. Design/methodology/approach - A detailed study of disasters in the world during the period (1900-2005) has been obtained from the recent published sources. In that period more than 40 lessons have been reported based on statistical data analysis of disasters. Furthermore, a two-dimensional probability density function is developed to categorize the different types of disasters. This paper studies and reviews some major impacts of disasters during the past 105 years and summarizes some major lessons for the future.... 

In this paper, a new feature set for audio classification is presented and evaluated based on sinusoidal modeling of audio signals. Variance of the birth-death frequencies in sinusoidal model of signal, as a measure of harmony, is used and compared to typical features as the input into an audio classifier. The performance of this sinusoidal model feature is evaluated through classification of audio to speech and music using both the GMM and the SVM classifiers. Classification results show that the proposed feature is quite successful in speech/music classification. Experimental comparisons with popular features for audio classification, such as HZCRR and LSTER, are presented and discussed.... 

This paper proposes a noninvasive, acoustic-based method to differentiate between individuals with and without dysphagia or swallowing dysfunction. Swallowing sound signals, both normal and abnormal (i.e., at risk of some degree of dysphagia) were recorded with accelerometers over the trachea. Segmentation based on waveform dimension trajectory (a distance-based technique) was developed to segment the nonstationary swallowing sound signals. Two characteristic sections emerged, Opening and Transmission, and 24 characteristic features were extracted and subsequently reduced via discriminant analysis. A discriminant algorithm was also employed for classification, with the system trained and...