Sharif Digital Repository

Most email users have encountered with spam problems, which have been addressed as a text classification or categorization problem. In this paper, we propose a novel spam detection method that uses ensemble of classifiers based on clustering and selection techniques. There is diversity in genre of e-mail's content and this method can find different topics in emails by clustering. It first computes disjoint clusters of emails, and then a classifier is trained on each cluster. When new email arrives, its cluster is identified. The classifier of the identified cluster is selected to classify the new email. Our method can extract many kinds of topics in emails. The evaluation shows that the... 

In the last decade spam detection has been addressed as a text classification or categorization problem. In this paper we propose a new dynamic weighted voting method based on the combination of clustering and weighted voting, and apply it to the task of spam filtering. In order to classify a new sample, it first compares with all cluster centroids and its similarity to each cluster is identified; Classifiers in the vicinity of the input sample obtain greater weight for the final decision of the ensemble. The evaluation shows that the algorithm outperforms pure SVM. © 2008 IEEE  

We study the structure of a minimum dominating set of C2n+1 n, the Cartesian product of n copies of the cycle of size 2n + 1, where 2n + 1 is a prime  

In this study, natural convection heat transfer in a vertical flat-plate solar air heater of 2.5 m height and 1 m width, with one- and two-glass covers was studied experimentally. Totally six cases of airflow (two for air heater with one glass cover and four for air heater with two-glass covers) were considered. These cases included states that air could flow within spaces between absorber plate and glass covers or air was enclosed in such spaces. Absorber plate temperature, back-plate temperature, glass cover temperatures, mass flow rates of air within channels and the solar radiation were measured. The following relations are suggested:. For channels in which air could flow:Nu = 0.7362... 

Suppose we have a family F of sets. For every S∈F, a set D⊆S is a defining set for (F,S) if S is the only element of F that contains D as a subset. This concept has been studied in numerous cases, such as vertex colorings, perfect matchings, dominating sets, block designs, geodetics, orientations, and Latin squares. In this paper, first, we propose the concept of a defining set of a logical formula, and we prove that the computational complexity of such a problem is Σ2-complete. We also show that the computational complexity of the following problem about the defining set of vertex colorings of graphs is Σ2-complete: INSTANCE: A graph G with a vertex coloring c and an integer k.QUESTION: If... 

Circular chromatic number, χc is a natural generalization of chromatic number. It is known that it is NP-hard to determine whether or not an arbitrary graph G satisfies χ(G)= χc(G). In this paper we prove that this problem is NP-hard even if the chromatic number of the graph is known. This answers a question of Xuding Zhu. Also we prove that for all positive integers k ≥ 2 and n ≥ 3, for a given graph G with χ(G)= n, it is NP-complete to verify if χc(G)≤ n -1/k. © 2004 Wiley Periodicals, Inc  

Adapting natural phenomena rendering for realtime applications has become a common practice in computer graphics.We propose a GPU-based multi-band method for optimized synthesis of "far from coast" ocean waves using an empirical Fourier domain model. Instead of performing two independent syntheses for low- and high-band frequencies of ocean waves, we perform only low-band synthesis and employ results to reproduce high frequency details of ocean surface by an optimized iterative up-sampling stage. Our experimental results show that this approach greatly improves the performance of original multi-band synthesis while maintaining image quality. © Springer-Verlag 2009  

We study the 2-center problem with outliers in high-dimensional data streams. Given a stream of points in arbitrary d dimensions, the goal is to find two congruent balls of minimum radius covering all but at most z points. We present a (1.8+ε)-approximation streaming algorithm, improving over the previous (4+ε)-approximation algorithm available for the problem. The space complexity and update time of our algorithm are poly(d,z,1/ε), independent of the size of the stream. © 2016 Elsevier B.V  

It is proved that every n × n Latin square has a partial transversal of length at least n - O (log2 n). The previous papers proving these results (including one by the second author) not only contained an error, but were sloppily written and quite difficult to understand. We have corrected the error and improved the clarity. © 2008  

A simple method for the formation of inclusion complex of [60]fullerene with two O2N2-donor aza-crown macrocyclic ligands was introduced. The products were characterized using UV-vis and IR spectroscopies as well as HPLC, and ESI mass spectrometry. The ESI mass and elemental analysis data revealed that the ratio of macroring:[60]fullerene were not the same. The binding capability of the macrorings as well as the number of the macroring addends to [60]fullerene were investigated using thermogravimetric analysis (TG) and differential scanning calorimetry (DSC). Considering percentage mass loss in different steps of TG and the enthalpy changes in DSC found for each of these products, the number... 

In this paper an approach based on system identification is used for fault detection in a nuclear reactor. A continuous-time Extended Kalman Filter (EKF) is presented, which allows the parameters of the nonlinear system to be estimated. Also a fault tolerant control system is designed for the nuclear reactor during power changes operation. The proposed controller is an adaptive critic-based neuro-fuzzy controller. Performance of the controller in terms of transient response and robustness against failures is very good and considerable