Iranian Journal of Science and Technology Transactions of Electrical Engineering
Volume:36 Issue: 2, 2012

Ontology matching is one of the most important topics in many fields of semantic web research. The focus of ontology matching is on discovering semantic correspondence between entities of differently modelled ontologies of the same domain. Although there have been many efforts on developing ontology matching tools, the results of matching process is not appealing. In this paper, a new approach is proposed in order to get a better result from matching process. In this approach a pre-processing phase is added to matchers for analysing and modifying input ontologies. Refactoring rules are utilized on detected inappropriate patterns. Matching algorithms are then applied on the refactored ontologies. To evaluate our proposed approach two well-known matchers, RiMOM, and ASMOV, are used. Evaluation results show that our proposed approach improved the quality of the matching process with respect to standard evaluation measurements such as precision, recall, and F-measure.

In this paper, a novel cost-sensitive learning algorithm is proposed to improve the performance of the nearest neighbor for intrusion detection. The goal of the learning algorithm is to minimize the total cost in leave-one-out classification of the given training set. This is important since intrusion detection is a problem in which the costs of different misclassifications are not the same. To optimize the nearest neighbor for intrusion detection, the distance function is defined in a parametric form. The free parameters of the distance function (i.e., the weights of features and instances) are adjusted by our proposed feature-weighting and instance-weighting algorithms. The proposed feature-weighting algorithm can be viewed as general purpose wrapper approach for feature weighting. The instance-weighting algorithm is designed to remove noisy and redundant training instances from the training set. This, in turn improves the speed and performance of the nearest neighbor in the generalization phase, which is quite important in real-time applications such as intrusion detection. Using the KDD99 dataset, we show that the scheme is quite effective in designing a cost-sensitive nearest neighbor for intrusion detection.

Although several studies have been conducted toward quantitative measuring depth of anesthesia (DOA), the state of art DOA indexes sometimes fail in practice. Hence, specialists are looking to find a new source of information, rather than modifying the former indexes, to introduce an accurate DOA index. In this regard, here, a new horizon to this field has been unveiled by photic stimulating the anesthetized patients’ eyelashes during surgical operations. In this way, this paper presents a new recording protocol to produce the depth-related visual evoke potential (VEP). Another contribution of this paper is to introduce an efficient method to elicit the VEPs within short trials (10 seconds). The suggested VEP extraction method can explain and detect the deterioration of VEP waveform through the successive trials. Finally, a novel DOAmeasure based on features of the clean VEPs is presented. Specificity and sensitivity of the proposed DOA is assessed by measuring its statistical similarity to the gold-standard BIS index over six patients. The presented VEP-based DOA index can be considered as an alternative of BIS index in the light and moderate anesthetic depth.

Brain Computer Interface (BCI) systems still suffer from lack of accuracy in real-time applications. This problem emerges from isolated optimization, and in some occasions from mismatching of feature extraction and classification stages. To unify optimization of both stages, this paper presents a novel scheme to integrate them and simultaneously optimize under a unit criterion. The proposed method iteratively estimates both spatio-spectral filters and classifier weights under a non-linear form of Fisher criterion. In order to validate the introduced method, two standard EEG sets, one containing 118 EEG signals and the other 29, were employed to demonstrate its spatial resolution capability. Experimental results on both datasets reveal the superiority of the proposed scheme in terms of enhancing the classification performance simultaneously with speeding up the optimization process, compared to the conventional methods.

In this paper, two low noise amplifiers (LNAs), one without feedback and another one with active shunt partial feedback, are proposed for ultra wide band (UWB) applications. Both the proposed LNAs are designed using 90 nm CMOS technology and their performance parameters are analyzed by using post layout simulation. The proposed LNA without feedback achieves a power gain (S21) of 16.4 dB over the band of 3 – 10.4 GHz with NF (Noise Figure) in the range of 4.9 – 5.2 dB. This high NF has been reduced to 2.4 – 2.7 dB by employing active shunt partial feedback. The proposed LNA with active shunt partial feedback achieves a power gain of 15 dB over the band of 2 – 12 GHz. The input matching (S11) and output matching (S22) are less than – 10 dB while maintaining the reverse isolation (S12) is less than -60 dB for both of the proposed circuits. Both circuits, with and without active shunt partial feedback, maintain better linearity with in-band third order input intercept point (IIP3) of – 1 dBm and – 4.242 dBm, respectively and consume 3.643 mW and 2.862 mW of power while operating at 1 V power supply.

Most previous studies in estimation of the target position and velocity through Bearing Only Measurements (BOM) consider targets with constant velocity moving along a straight line. In this paper, state and measurement equations are presented for moving targets with constant acceleration by using the previously presented state vector in the Extended Modified Polar Coordinates (EMPC) system. In the BOM systems, by increasing the distance between target and observer (Own ship) the estimation accuracy of the target kinematic parameters degrades noticeably. In order to solve this problem, here the idea of hybrid data measurements is presented. In this approach both low rate range information, from active sensor, and high rate BOM are exploited. The improvement in the performance of the hybrid system compared to BOM system is represented through computer simulations.

with heavy loading. In this paper, Modal Series (MS) as a method to analyze modal interactions is extended for a power system installed with a static synchronous series compensator (SSSC) stabilizer. The - based and m-based stabilizers as two stabilizers in different control channels are presented for a SSSC. The parameters of the stabilizers are calculated by a quadratic mathematical programming method. In this procedure, the gain and the phase of a stabilizer are calculated simultaneously. A particular measure of stabilizer gain is considered as an objective function. The effects of SSSC based stabilizers on damping inter-area oscillations for a small disturbance are studied and compared. Modal interactions between an inter-area mode and control modes related to SSSC stabilizers are studied by a proposed index based on MS method in a 4-machine stressed power system. Oscillatory instability caused by modal interactions is investigated and compared in the system for both SSSC stabilizers.