WoS İndeksli Yayınlar Koleksiyonu
Permanent URI for this collectionhttps://hdl.handle.net/20.500.12416/8653
Browse
9 results
Search Results
Conference Object Covariance Features for Trajectory Analysis(IEEE, 2016) Karadeniz, Talha; Maras, Hadi HakanIn this work, we aimed to demonstrate that covariance estimation methods can be used for trajectory classification. We have shown that, features obtained via shrunk covariance estimation are suitable for describing trajectories. We have arrived to the conclusion that, when compared to Dynamic Time Warping, the explained technique is faster and may yield more accurate results.Conference Object Covariance Features for Trajectory Analysis(Institute of Electrical and Electronics Engineers Inc., 2016) Karadeniz, T.; Maras, H.H.In this work, we aimed to demonstrate that covariance estimation methods can be used for trajectory classification. We have shown that, features obtained via shrunk covariance estimation are suitable for describing trajectories. We have arrived to the conclusion that, when compared to Dynamic Time Warping, the explained technique is faster and may yield more accurate results. © 2017 Elsevier B.V., All rights reserved.Article Citation - WoS: 1Citation - Scopus: 1Computational Structural Analysis for Substituted Nickel Phthalocyanines Preparation and Characterization(Revista Chimie Srl, 2016) Youssef, Tamer E.; Baleanu, Dumitru; Al Turaif, Hamad; Aftab, Wasim; Baleanu, Dumitru; MatematikHay synthesis of a novel series of symmetrically substituted nickel phthalocyanine derivatives, [(heterothio)(8)NiPcs] 4(a-e) bearing N-heterocycle moieties, i.e. Pyridine, Thiazoline, Imidazole, Tetrazole rings, was reported. Their novel heterocycle-thio phthalonitrile precursors 3(a-e) were synthesized by the aromatic nucleophilic substitution reaction of 4-nitrorophthalonitrile with N-heterocycle substituted thiols 2(a-e). The structure of the compounds was analyzed by the spectroscopic analysis tools. Besides, the powerful techniques of maximal wavelet discrete wavelet transformation together with boxplot clustering pointed out new aspects of the signals which cannot be detected by the spectroscopic analysis.Conference Object Machine Learning-Based Silence Detection in Call Center Telephone Conversations(Ieee, 2019) Iheme, Leonardo O.; Ozan, Sukru; Akagunduz, ErdemThis study presents the development of a voice activity detection (VAD) system tested on call center telephony data obtained from our local site. The concept of bag of audio words (BoAW) combined with a naive Bayes classifier was applied to achieve the task. It was formulated as a binary classification problem with speech as the positive class and silence/background noise as the negative class. All the processing was performed on the Mel-frequency cepstral coefficients (MFCCs) extracted from the audio recordings. The results which are presented as accuracy score and receiver operating characteristics (ROC) indicate an excellent performance of the developed model. The system is to be deployed within our call center to aid data analysis and improve overall efficiency of the center.Article Citation - WoS: 21Citation - Scopus: 27Reporting and Analyzing Alternative Clustering Solutions by Employing Multi-Objective Genetic Algorithm and Conducting Experiments on Cancer Data(Elsevier, 2014) Peng, Peter; Addam, Omer; Ozyer, Sibel T.; Elzohbi, Mohamad; Elhajj, Ahmad; Gao, Shang; Alhajj, RedaClustering is an essential research problem which has received considerable attention in the research community for decades. It is a challenge because there is no unique solution that fits all problems and satisfies all applications. We target to get the most appropriate clustering solution for a given application domain. In other words, clustering algorithms in general need prior specification of the number of clusters, and this is hard even for domain experts to estimate especially in a dynamic environment where the data changes and/or become available incrementally. In this paper, we described and analyze the effectiveness of a robust clustering algorithm which integrates multi-objective genetic algorithm into a framework capable of producing alternative clustering solutions; it is called Multi-objective K-Means Genetic Algorithm (MOKGA). We investigate its application for clustering a variety of datasets, including microarray gene expression data. The reported results are promising. Though we concentrate on gene expression and mostly cancer data, the proposed approach is general enough and works equally to cluster other datasets as demonstrated by the two datasets Iris and Ruspini. After running MOKGA, a pareto-optimal front is obtained, and gives the optimal number of clusters as a solution set. The achieved clustering results are then analyzed and validated under several cluster validity techniques proposed in the literature. As a result, the optimal clusters are ranked for each validity index. We apply majority voting to decide on the most appropriate set of validity indexes applicable to every tested dataset. The proposed clustering approach is tested by conducting experiments using seven well cited benchmark data sets. The obtained results are compared with those reported in the literature to demonstrate the applicability and effectiveness of the proposed approach. (C) 2013 Elsevier B.V. All rights reserved.Conference Object Evaluation of Clustering Performance of Hyperspectral Bands(Ieee, 2015) Sakarya, Ufuk; Toreyin, Behcet Ugur; Haliloglu, Onur; Haliloʇlu, OnurHyperspectral images have huge data volume that contains spectral and spatial information. This high data volume leads to processing, storage, and transmission problems. Moreover, insufficient training data results in Hughes phenomenon. It is possible to solve these problems with the help of feature selection. In this paper, a method that evaluates the clustering performance of spectral bands is proposed as a pre-processing operation in order to realize feature selection. This method is clustering each spectral band based on "dominant sets" technique and it evaluates the clustering performance of each band. The proposed method is time efficient since it works on a small set of training data instead of the whole hyperspectral data. In this study, "dominant sets" technique is first applied to hyperspectral image processing as a clustering method.Conference Object Citation - WoS: 2Citation - Scopus: 3Classification of Fmri Data by Using Clustering(Ieee, 2015) Mogultay, Hazal; Alkan, Sarper; Yarman-Vural, Fatos T.; Moʇultay, HazalRecognition of the the cognitive states by using functional Magnetic Rezonans Imaging (fMRI) data is a challenging problem that has been a focus of scientific research for a long time. In this study the effectiveness of clustering and the ensemble learning techniques on fMRI dataset is investigated and different paramaters are compared. Moreover, the performance of these techniques are tested on both raw voxel intensity values and meshes formed by multiple voxels. Clusters are compared to the functional brain regions, however higher performances are obtained when the number of clusters is higher than the number of functional brain regions.Article Citation - WoS: 8Citation - Scopus: 9Power Aware Routing Protocols in Wireless Sensor Network(Ieice-inst Electronics information Communications Eng, 2016) Oztoprak, Kasim; Hassanpour, Reza; Alsultan, MohammedWireless Sensor Networks (WSNs) have gained importance with a rapid growth in their applications during the past decades. There has also been a rise in the need for energy-efficient and scalable routing along with the data aggregation protocols for the large scale deployments of sensor networks. The traditional routing algorithms suffer from drawbacks such as the presence of one hop long distance data transmissions, very large or very small clusters within a network at the same moment, over-accumulated energy consumption within the cluster-heads (CHs) etc. The lifetime of WSNs is also decreased due to these drawbacks. To overcome them, we have proposed a new method for the Multi Hop, Far-Zone and Load-Balancing Hierarchical-Based Routing Algorithm for Wireless Sensor Network (MFLHA). Various improvements have been brought forward by MFLHA. The first contribution of the proposed method is the existence of a large probability for the nodes with higher energy to become the CH through the introduction of the energy decision condition and energy-weighted factor within the electing threshold of the CH. Secondly, MFLHA forms a Far-Zone, which is defined as a locus where the sensors can reach the CH with an energy less than a threshold. Finally, the energy consumption by CHs is reduced by the introduction of a minimum energy cost method called the Multi-Hop Inter-Cluster routing algorithm. Our experimental results indicate that MFLHA has the ability to balance the network energy consumption effectively as well as extend the lifetime of the networks. The proposed method outperforms the competitors especially in the middle range distances.Conference Object Citation - WoS: 35Citation - Scopus: 58Weather Data Analysis and Sensor Fault Detection Using an Extended Iot Framework With Semantics, Big Data, and Machine Learning(Ieee, 2017) Sezer, Omer Berat; Ozbayoglu, Murat; Dogdu, Erdogan; Onal, Aras Can; Berat Sezer, OmerIn recent years, big data and Internet of Things (IoT) implementations started getting more attention. Researchers focused on developing big data analytics solutions using machine learning models. Machine learning is a rising trend in this field due to its ability to extract hidden features and patterns even in highly complex datasets. In this study, we used our Big Data IoT Framework in a weather data analysis use case. We implemented weather clustering and sensor anomaly detection using a publicly available dataset. We provided the implementation details of each framework layer (acquisition, ETL, data processing, learning and decision) for this particular use case. Our chosen learning model within the library is Scikit-Learn based k-means clustering. The data analysis results indicate that it is possible to extract meaningful information from a relatively complex dataset using our framework.