Shahrood University of TechnologyJournal of AI and Data Mining2322-52111120130128Outlier Detection in Wireless Sensor Networks Using Distributed Principal Component Analysis1119810.22044/jadm.2013.98ENMohammadAhmadi LivaniFaculty of Electrical and Computer Engineering, Tarbiat Modares University, Tehran, IranMahdiAbadiFaculty of Electrical and Computer Engineering, Tarbiat Modares University, Tehran, IranMeysamAlikhanyFaculty of Electrical and Computer Engineering, Tarbiat Modares University, Tehran, IranMeisamYadollahzadeh TabariDepartment of Computer Eng. & IT, Shahrood University of Technology,Shahrood,IranJournal Article20130109Detecting anomalies is an important challenge for intrusion detection and fault diagnosis in wireless sensor networks (WSNs). To address the problem of outlier detection in wireless sensor networks, in this paper we present a PCA-based centralized approach and a DPCA-based distributed energy-efficient approach for detecting outliers in sensed data in a WSN. The outliers in sensed data can be caused due to compromised or malfunctioning nodes. In the distributed approach, we use distributed principal component analysis (DPCA) and fixed-width clustering (FWC) in order to establish a global normal pattern and to detect outlier. The process of establishing the global normal pattern is distributed among all sensor nodes. We also use weighted coefficients and a forgetting curve to periodically update the established normal profile. We demonstrate that the proposed distributed approach achieves comparable accuracy compared to the centralized approach, while the communication overhead in the network and energy consumption is significantly reduced.https://jad.shahroodut.ac.ir/article_98_581145b2adcc49bf156eee44c1df7703.pdfShahrood University of TechnologyJournal of AI and Data Mining2322-52111120130209An Enhanced Median Filter for Removing Noise from MR Images131711010.22044/jadm.2013.110ENS.ArastehfarNational University of Singapore (NUS)Ali A.PouyanSchool of Computer Engineering, Shahrood University of TechnologyA.JalalianNational University of Singapore (NUS)Journal Article20120801<span>In this paper, a novel decision based median (DBM) filter for enhancing MR images has been proposed. The method is based on eliminating impulse noise from MR images. A median-based method to remove impulse noise from digital MR images has been developed. Each pixel is leveled from black to white like gray-level. The method is adjusted in order to decide whether the median operation can be applied on a pixel. The main deficiency in conventional median filter approaches is that all pixels are filtered with no concern about healthy pixels. In this research, to suppress this deficiency, noisy pixels are initially detected, and then the filtering operation is applied on them. The proposed decision method (DM) is simple and leads to fast filtering. The results are more accurate than other conventional filters. Moreover, DM adjusts itself based on the conditions of local detections. In other words, DM operation on detecting a pixel as a noise depends on the previous decision. As a considerable advantage, some unnecessary median operations are eliminated and the number of median operations reduces drastically by using DM. Decision method leads to more acceptable results in scenarios with high noise density. Furthermore, the proposed method reduces the probability of detecting noise-free pixels as noisy pixels and <em>vice versa</em>. </span>https://jad.shahroodut.ac.ir/article_110_4b3efdd1562a00a00833a9a15ffbd83e.pdfShahrood University of TechnologyJournal of AI and Data Mining2322-52111120130211An Improved Algorithm for Network Reliability Evaluation192511110.22044/jadm.2013.111ENMohammadGhasemzadehYazd University0000-0002-6805-4852Journal Article20130119Binary Decision Diagram (BDD) is a data structure proved to be compact in representation and efficient in manipulation of Boolean formulas. Using Binary decision diagram in network reliability analysis has already been investigated by some researchers. In this paper we show how an exact algorithm for network reliability can be improved and implemented efficiently by using CUDD - Colorado University Decision Diagram.https://jad.shahroodut.ac.ir/article_111_c4062d67d678eb9bb832bba521229f18.pdfShahrood University of TechnologyJournal of AI and Data Mining2322-52111120130220Image retrieval using the combination of text-based and content-based algorithms273411310.22044/jadm.2013.113ENHaniehMohamadiIslamic Azad University of ArakAsadollahShahbahramiUniversity of GuilanJavadAkbariIslamic Azad University of ArakJournal Article20130111Image retrieval is an important research field which has received great attention in the last decades. In this paper, we present an approach for the image retrieval based on the combination of text-based and content-based features. For text-based features, keywords and for content-based features, color and texture features have been used. Query in this system contains some keywords and an input image. At first, the images are retrieved based on the input keywords. Then, visual features are extracted to retrieve ideal output images. For extraction of color features we have used color moments and for texture we have used color co-occurrence matrix. The COREL image database have been used for our experimental results. The experimental results show that the performance of the combination of both text- and content- based features is much higher than each of them which is applied separately.https://jad.shahroodut.ac.ir/article_113_93818ee658bb0f2f0ef087911590431e.pdfShahrood University of TechnologyJournal of AI and Data Mining2322-52111120130220Expert Discovery: A web mining approach354711610.22044/jadm.2013.116ENMuhammadNaeemMohammad Ali Jinnah University Islamabad PakistanMuhammad BilalKhanMohammad Ali Jinnah University Islamabad PakistanMuhammad TanvirAfzalMohammad Ali Jinnah University Islamabad, PakistanJournal Article20130203Expert discovery is a quest in search of finding an answer to a question: “Who is the best expert of a specific subject in a particular domain within peculiar array of parameters?” Expert with domain knowledge in any field is crucial for consulting in industry, academia and scientific community. Aim of this study is to address the issues for expert-finding task in real-world community. Collaboration with expertise is critical requirement in business corporate such as in fields of engineering, geographies, bio-informatics, medical domain etc. We have proposed multifaceted web mining heuristic that resulted into the design and development of a tool using data from Growbag, dblpXML with Authors home pages resource to find people of desired expertise. We mined more than 2,500 Author's web pages based on the credibility of 12 key parameters while parsing on each page for a large number of co-occurred keyword and all available general terms. It presents evidence to validate this quantification as a measure of expertise. The prototype enables users to distinguish easily someone, who has briefly worked in a particular area with more extensive experience, resulting in the capability to locate people with broader expertise throughout large parts of the product. Through this extension to the web enabling methodology, we have shown that the implemented tool delivers a novel web mining idea with improved results.https://jad.shahroodut.ac.ir/article_116_a4340c56079eae6e7078c17de68a6460.pdfShahrood University of TechnologyJournal of AI and Data Mining2322-52111120130226Protection Scheme of Power Transformer Based on Time–Frequency Analysis and KSIR-SSVM496112210.22044/jadm.2013.122ENMehdiHajiansemnan universityAsgharAkbari ForoudElectrical & Computer Faculty, Semnan UniversityJournal Article20130129The aim of this paper is to extend a hybrid protection plan for Power Transformer (PT) based on MRA-KSIR-SSVM. This paper offers a new scheme for protection of power transformers to distinguish internal faults from inrush currents. Some significant characteristics of differential currents in the real PT operating circumstances are extracted. In this paper, Multi Resolution Analysis (MRA) is used as Time–Frequency Analysis (TFA) for decomposition of Contingency Transient Signals (CTSs), and feature reduction is done by Kernel Sliced Inverse Regression (KSIR). Smooth Supported Vector Machine (SSVM) is utilized for classification. Integration KSIR and SSVM is tackled as most effective and fast technique for accurate differentiation of the faulted and unfaulted conditions. The Particle Swarm Optimization (PSO) is used to obtain optimal parameters of the classifier. The proposed structure for Power Transformer Protection (PTP) provides a high operating accuracy for internal faults and inrush currents even in noisy conditions. The efficacy of the proposed scheme is tested by means of numerous inrush and internal fault currents. The achieved results are utilized to verify the suitability and the ability of the proposed scheme to make a distinction inrush current from internal fault. The assessment results illustrate that proposed scheme presents an enhancement of distinguish inrush current from internal fault over the method to be compared without Dimension Reduction (DR).https://jad.shahroodut.ac.ir/article_122_2b1ec9b936799ad947a4c9b02dfa7cd9.pdf