University of Kashan Soft Computing Journal 2322-3707 1 2 2021 05 23 Wireless Sensor Network Fault Tolerance Using Spare Node in the Low Density Wireless Sensor Network Fault Tolerance Using Spare Node in the Low Density 2 13 111358 FA Salman Goli Journal Article 2021 05 23 To increase reliability and fault tolerance in wireless sensor network, using spare nodes is a useful method. In this article, we survey the influence of using single-type and multi-type spare nodes on the fault tolerance in the low density. To achieve this, we use single-type and multi-type spare nodes and construct the network reliability graph for one, two and three-node densities. Then, we consider the mentioned spares from spare-less to three spares and construct Markov models. Afterwards, reliability function, the average failure rate and the entire network MTTF are calculated by solving Markov equations. The simulation results showed that the use of shared spare nodes leads to improving MTTF in the higher density. Moreover, we found that there is a limited number of spares to improve the MTTF so that we will no longer have more improvement by applying more spares. To increase reliability and fault tolerance in wireless sensor network, using spare nodes is a useful method. In this article, we survey the influence of using single-type and multi-type spare nodes on the fault tolerance in the low density. To achieve this, we use single-type and multi-type spare nodes and construct the network reliability graph for one, two and three-node densities. Then, we consider the mentioned spares from spare-less to three spares and construct Markov models. Afterwards, reliability function, the average failure rate and the entire network MTTF are calculated by solving Markov equations. The simulation results showed that the use of shared spare nodes leads to improving MTTF in the higher density. Moreover, we found that there is a limited number of spares to improve the MTTF so that we will no longer have more improvement by applying more spares. Wireless sensor network Reliability Spare node Fault tolerance MTTF Markov model https://scj.kashanu.ac.ir/article_111358_5941e6df59a1c58c807ecbec79aef40f.pdf
University of Kashan Soft Computing Journal 2322-3707 1 2 2021 05 23 Improvement of Semantic Labeling Method by KBO for Automated Termination Proof Improvement of Semantic Labeling Method by KBO for Automated Termination Proof 14 25 111359 FA Mohammad Kadkhoda Saeed Jalili Mohammad Izadi Journal Article 2021 05 23  The term rewriting systems (TRSs) is an abstract model of functional languages. The termination proving of TRSs is necessary for confirming accuracy of functional languages. The semantic labeling (SL) is a complete method for proving termination. The semantic part of SL is given by a quasi-model of the rewrite rules. The most power of SL is related to infinite models that is difficult for support by tools of automated termination proving. In this paper, we combined the SL method with natural numbers and Knuth-Bendix order (KBO) so that one can automatically prove termination using infinite models. We: (1) made a generalization based on KBO, called labeling Knuth-Bendix order (ℓKBO), (2) showed its ability in proving termination of TRSs, (3) we introduced an algorithm to automatically search a ℓKBO for a given TRS and (4) successfully tested the algorithm functionality on TPDB 3.1, a data set of TRSs.   The term rewriting systems (TRSs) is an abstract model of functional languages. The termination proving of TRSs is necessary for confirming accuracy of functional languages. The semantic labeling (SL) is a complete method for proving termination. The semantic part of SL is given by a quasi-model of the rewrite rules. The most power of SL is related to infinite models that is difficult for support by tools of automated termination proving. In this paper, we combined the SL method with natural numbers and Knuth-Bendix order (KBO) so that one can automatically prove termination using infinite models. We: (1) made a generalization based on KBO, called labeling Knuth-Bendix order (ℓKBO), (2) showed its ability in proving termination of TRSs, (3) we introduced an algorithm to automatically search a ℓKBO for a given TRS and (4) successfully tested the algorithm functionality on TPDB 3.1, a data set of TRSs.  Termination proving Semantic labeling Knuth-Bendix order Term rewriting system https://scj.kashanu.ac.ir/article_111359_bdbeb6d88d15fefb533a97c35810810e.pdf
University of Kashan Soft Computing Journal 2322-3707 1 2 2021 05 23 A Novel Intelligent Persian Authorship System based on Writing Style A Novel Intelligent Persian Authorship System based on Writing Style 26 35 111360 FA Zeinab Farahmandpoor Hooman Nikmehr Moharram Mansoorizade Omid Tabibzadeh Ghamsary Journal Article 2021 05 23 The rapid development of communication by the Internet and the misuse of the anonymity embedded in the nature of online written documents have led to serious security issues. Anonymous identity of the Internet tools such as emails, blogs, and Web sites have made them target methods of interest for criminal activities. On the other hand, world social and political relations have made a great interest in Persian language leading to the spread of Persian manuscripts in the Internet. In this paper, an intelligent writeprint technique is introduced to demonstrate a Persian authorship based on his/her writing style. In this research, we used specific features of: (1) lexical, syntactic and semantic and (2) the application for identifying the Persian writer. Moreover, we reviewed: (1) the impact of the features performance and (2) KNN and Delta classification methods combined with the genetic algorithm on a database. To make implementation of the proposed approach possible, we designed a pos-tagger to detect Persian nouns, adjectives and adverbs using the word structure. The experimental results showed that, among others, the KNN and genetic algorithm combination method is more accurate in the Persian authorship recognition. The rapid development of communication by the Internet and the misuse of the anonymity embedded in the nature of online written documents have led to serious security issues. Anonymous identity of the Internet tools such as emails, blogs, and Web sites have made them target methods of interest for criminal activities. On the other hand, world social and political relations have made a great interest in Persian language leading to the spread of Persian manuscripts in the Internet. In this paper, an intelligent writeprint technique is introduced to demonstrate a Persian authorship based on his/her writing style. In this research, we used specific features of: (1) lexical, syntactic and semantic and (2) the application for identifying the Persian writer. Moreover, we reviewed: (1) the impact of the features performance and (2) KNN and Delta classification methods combined with the genetic algorithm on a database. To make implementation of the proposed approach possible, we designed a pos-tagger to detect Persian nouns, adjectives and adverbs using the word structure. The experimental results showed that, among others, the KNN and genetic algorithm combination method is more accurate in the Persian authorship recognition. Authorship Delta classification KNN classification Writing style writeprint https://scj.kashanu.ac.ir/article_111360_3df4cbfbb90d025ccdb6e58b2808d996.pdf
University of Kashan Soft Computing Journal 2322-3707 1 2 2021 05 23 Using Semantic Relations to Improve Quality of a Citation Recommendation System Using Semantic Relations to Improve Quality of a Citation Recommendation System 36 45 111361 FA Fattane Zarrinkalam Mohsen Kahani Journal Article 2021 05 23 With the increasingly growth of scientific documents in the Web, it is difficult to select a concerned document. A citation recommendation system receives a text and recommends documents to be cited by the text. Such recommendation helps a researcher in hitting his/her concerned texts. Based on sematic relations, this paper presents a new indicator to measure the similarity between documents and presents a citation recommendation system exploiting the indicator along with other document features. The experimental results showed that the indicator succeeds in the document similarity recognition and leads to improvement in the recommendation. With the increasingly growth of scientific documents in the Web, it is difficult to select a concerned document. A citation recommendation system receives a text and recommends documents to be cited by the text. Such recommendation helps a researcher in hitting his/her concerned texts. Based on sematic relations, this paper presents a new indicator to measure the similarity between documents and presents a citation recommendation system exploiting the indicator along with other document features. The experimental results showed that the indicator succeeds in the document similarity recognition and leads to improvement in the recommendation. Citation recommendation Semantic similarity Textual similarity https://scj.kashanu.ac.ir/article_111361_f18f79b74db517163821dde643cfdc5f.pdf
University of Kashan Soft Computing Journal 2322-3707 1 2 2021 05 23 A Survey of Evolutionary Computations Usage in spectral Analysis of hyperspectral images A Survey of Evolutionary Computations Usage in spectral Analysis of hyperspectral images 46 59 111362 FA Hossein Fayyazi Hamid Dehghani Mojtaba Hosseini Journal Article 2021 05 23 In recent years, the spectral analysis has been one of the most important research areas in remote sensing, which has received different traditional solutions. Most of these methods have several special conditions to work well. However, they suffer from the problems such as sticking in local optimum and having sensibility to parameters initialization. Although the Evolutionary Computation (EC) techniques may also have such deficiencies, recently their combination with the traditional methods has been used increasingly to overcome the defects. Some other methods use EC exclusively to solve the problems. In the new scheme, the spectral analysis problem is modeled as an optimization problem demonstrating the EC techniques lead to optimum solutions. In this paper, we address the methods using the EC to solve the spectral analysis problem. Having introduce the EC technique briefly in each section of this paper, we explain methods use EC. For each method, the assumptions and limitations are discussed and the various components of the EC method such as individual representation, evolutionary operators and fitness function are assessed. In recent years, the spectral analysis has been one of the most important research areas in remote sensing, which has received different traditional solutions. Most of these methods have several special conditions to work well. However, they suffer from the problems such as sticking in local optimum and having sensibility to parameters initialization. Although the Evolutionary Computation (EC) techniques may also have such deficiencies, recently their combination with the traditional methods has been used increasingly to overcome the defects. Some other methods use EC exclusively to solve the problems. In the new scheme, the spectral analysis problem is modeled as an optimization problem demonstrating the EC techniques lead to optimum solutions. In this paper, we address the methods using the EC to solve the spectral analysis problem. Having introduce the EC technique briefly in each section of this paper, we explain methods use EC. For each method, the assumptions and limitations are discussed and the various components of the EC method such as individual representation, evolutionary operators and fitness function are assessed. Remote Sensing Hyperspectral images Spectral analysis Optimization problems Evolutionary computations https://scj.kashanu.ac.ir/article_111362_8ee236949465f951d5ff9ac98059ddf8.pdf
University of Kashan Soft Computing Journal 2322-3707 1 2 2021 05 23 Optimization of PID Controller Parameters for Load Frequency Controller Using Imperialist Competitive Algorithm Optimization of PID Controller Parameters for Load Frequency Controller Using Imperialist Competitive Algorithm 60 73 111363 FA Seyed Abbas Taher Mehdi Zeraati Journal Article 2021 05 23 In this paper, considering variant power system parameters and using Imperialist Competitive Algorithm (ICA) and ITAE (Integral Time Absolute Error) criterion we deal with tuning optimal parameter of load frequency PID controller in two-area power systems. To attain the desirable robust performance, selecting the appropriate objective function is important. The obtained simulation results indicate that despite frequency deviation and Area Control Error (ACE), applying ITAE criterion leads to attain optimal control of parameters for power system using ICA, which has high accuracy and convergence speed. Moreover, performance of the proposed method is illustrated on a two-area power system under load disturbance. The obtained results indicate that the proposed optimization method has a desired and robust performance in a wide range of system parameters and load variations. In this paper, considering variant power system parameters and using Imperialist Competitive Algorithm (ICA) and ITAE (Integral Time Absolute Error) criterion we deal with tuning optimal parameter of load frequency PID controller in two-area power systems. To attain the desirable robust performance, selecting the appropriate objective function is important. The obtained simulation results indicate that despite frequency deviation and Area Control Error (ACE), applying ITAE criterion leads to attain optimal control of parameters for power system using ICA, which has high accuracy and convergence speed. Moreover, performance of the proposed method is illustrated on a two-area power system under load disturbance. The obtained results indicate that the proposed optimization method has a desired and robust performance in a wide range of system parameters and load variations. Imperialist Competitive Algorithm Load Frequency Control Optimal Tuning PID Controller https://scj.kashanu.ac.ir/article_111363_5a0660fe146bef6f21cd033eb8161f07.pdf