Feature ranking for Persian Review Spam detection
Neshat
Safarian
Department of Computer، Safahan Institute of Higher Educations، Janbazan St.، Isfahan، Iran
author
Mohammad Ehsan
Basiri
Department of computer, Faculty of engineering, Shahrekord university, Sharekord, Iran
author
Hadi
Khosravi
Computer Department، Shahrekord University، Rahbar Blvd.، Shahrekord، Iran
author
text
article
2019
per
Using online reviews is one of the main factors in customers’ decision making for buying a product or using a service. These reviews are valuable sources of information which can be used for detecting public opinion about products or services. Although online reviews are useful, trusting them blindly is dangerous for both costumers and sellers as they may be manipulated to earn profit; such reviews are called spam reviews. The current study addresses Persian reviews about cell-phone extracted from Digikala.com and investigates spam type 1 and type 2 which are fake reviews and reviews describing brands’ names only, respectively. Features used in this study, due to their efficiency, are review-based and metadata features. These features and their combinations in detecting Persian spam reviews, also their effect on the accuracy of classifier are assessed. Spam classification is performed using decision tree, support vector machines, and naïve Bayes classifiers and their accuracy are compared using different features’ combinations. The highest accuracy is obtained using the decision tree classifier which achieves 0.778 in terms of F-measure. In ranking features, again the decision tree outperforms the other two classifiers by achieving 0.824 F-measure by combining the positive feedback, overall score, and review polarity features.
Journal of Soft Computing and Information Technology
Babol Noshirvani University of Technology
2383-1006
8
v.
2
no.
2019
1
16
https://jscit.nit.ac.ir/article_87279_df20f941945a01babc0de419ea3e34af.pdf
A cost-aware method for cloud services composition using a hybrid algorithm
Saied
Asghari
Department of Computer Engineering, Tabriz Branch, Islamic Azad University, Tabriz, Iran
author
Nima
Jafari Navimipour
Department of Computer Engineering, Islamic Azad University, Tabriz Branch, Tabriz, Iran
author
text
article
2019
per
Cloud computing is considered as a new methods of computations where resources can be scaled and provides services in virtualized format using the Internet. In some cases, the user’s need is in a way that the underlying service cannot meet user’s need individually and it is needed to combine services in order to meet the requirements. The previously presented methods had some problems such as not checking the cost, energy consumption, and not providing a framework for using the lowest number of clouds to respond to user requests. Therefore, the proposed method in this paper combines an ant colony algorithm with a based cloud algorithm (ACOBC). In this method, first, the cloud compositions that can respond to user requests are arranged in ascending order based on the number of clouds in the cloud composition, then the ant colony algorithm selects the appropriate services from each category, respectively. The obtained results have shown that the proposed method can act better in terms of energy consumption and cost.
Journal of Soft Computing and Information Technology
Babol Noshirvani University of Technology
2383-1006
8
v.
2
no.
2019
17
26
https://jscit.nit.ac.ir/article_87833_1b6fe4fb16b39352f24ea72f30916584.pdf
A new model for selecting information flow pattern in build-to-order Supply Chains
Saber
Dinparast
computer and information technology engineering department, Amirkabir University of technology, Tehran, Iran
author
S. Alireza
Hashemi Golpayegany
Assistant professor, computer and information technology engineering department, Amirkabir University of technology, Tehran, Iran
author
text
article
2019
per
Build-to-order supply chains are categorized as agile supply chains, therefore reshaping their physical structure is inevitable. The reshape affects chains material flow pattern, therefore revising the chains information flow pattern becomes a necessity. The revision should create the most coordinated information flow pattern with the new physical structure. Hence, we have tried to study and survey the way material flows in supply chain affects its information flow and vice versa. We have thought up a model in which mathematical modeling establishes coordinated information and material flow patterns. To achieve this, the parameters which build the two flow patterns were studied and considered. Each parameter’s effects on others has been studied and relations were extracted. Using the capabilities of mathematical modeling the studied system converted to a model in which, some parameters as inputs give away the most coordinated information and material flow with chains physical structure considering minimum cost as objective.
Journal of Soft Computing and Information Technology
Babol Noshirvani University of Technology
2383-1006
8
v.
2
no.
2019
27
43
https://jscit.nit.ac.ir/article_85773_a72c079b9ee2318210ef8be05dd2d8f3.pdf
A New Approach for Optimal Placement of Virtual Machines in Cloud Datacenters Using Discrete Gravitational Search Algorithm and Chaotic Functions
sasan
gharehpasha
Department of Computer, Islamic Azad University, Urmia Branch, Urmia, Iran
author
mohammad
masdari
Department of Computer, Islamic Azad University, Urmia Branch, Urmia, Iran
author
ahmad
jafarian
Department of Mathematics, Islamic Azad University, Urmia Branch, Urmia, Iran
author
text
article
2019
per
Placement of virtual machines on physical machines in cloud computing infrastructure is an important issue. Our approach for placement of virtual machines includes a mapping process of this machines on physical machines in cloud datacenters. Optimal placement results in lower power consumption, optimal usage of resources, traffic reduce in datacenters, decrease in costs and also increase in functionality of datacenters in cloud datacenters. In this paper we propose a discrete gravitational search algorithm and chaotic function for placement of virtual machines on physical machines in cloud datacenters. Our primary goal for proposing the approach is minimizing resource wastage, power consumption and network links. At the end of this paper we also compare our results with some other metaheuristic algorithms. Our results show that this approach is more effective than previous algorithms. by optimal placement in cloud datacenter, we can get best performance of our devices. Also we can do it by chaotic functions.
Journal of Soft Computing and Information Technology
Babol Noshirvani University of Technology
2383-1006
8
v.
2
no.
2019
44
54
https://jscit.nit.ac.ir/article_85845_019e28b52f42d88b7ada668e9a79f858.pdf
A Novel Approach for being Completely Anonymous in Cloud Computing Environment
Fatemeh
Raji
Department of Computer Engineering,
Faculty of Software Engineering,
University of Isfahan,
Hezar Jerib Ave.,
Isfahan, Iran.
author
text
article
2019
per
Cloud computing technology has attracted the attention of researchers in recent years. Providing user security in terms of anonymity is one of the most important subject in the domain of cloud computing. Users desire to conceal their identity while using cloud computing services. Although there are researches for providing anonymity in the networks, there are limited works on embedding the anonymity feature in the cloud computing context. In this paper, we propose an anonymity approach to provide the anonymity of cloud users against the cloud provider and make the user to be resistant against traffic analysis attacks. In this way, all the communication messages between users and the provider has been passed through a set of intermediate hosts. Therefore, not only the users' request messages but also the provider response messages are resistant against traffic analysis attackers. Moreover, the users desired anonymity and computation is adjustable in the proposed technique.
Journal of Soft Computing and Information Technology
Babol Noshirvani University of Technology
2383-1006
8
v.
2
no.
2019
55
75
https://jscit.nit.ac.ir/article_87396_35667f712da5a3579013489d4ef523c7.pdf
A Novel Approach in Computer Vision and Photogrammetry to Recover the Relative Position and Orientation of Cameras in Stereo Images Using the SVD Decomposition of the Essential Matrix
Masood
Varshosaz
Department of Photogrammetry and Remote Sensing, Geomatics Engineering Faculty, K.N.Toosi University of Technology, Tehran, Iran
author
Alireza
Afary
Department of Photogrammetry and Remote Sensing, Geomatics Engineering Faculty, K.N.Toosi University of Technology, Tehran, Iran.
author
Mohammad
Saadatseresht
School of Surveying and Geospatial Engineering, University of Tehran, Tehran, Iran
author
Barat
Mojaradi
School of Civil Engineering, Iran University of Science and Technology, Tehran, Iran
author
text
article
2019
per
The relative position and orientation between two cameras in a stereo pair are included within the Essential matrix, E. The decomposition of this matrix into a rotation matrix, R, and a skew-symmetric matrix, S, is an efficient tool for retrieving the relative position and orientation of the cameras. In this paper, a new method is proposed to recover the relative position and orientation of the cameras in a stereo pair using the singular value decomposition (SVD) of the Essential matrix. First, the existing formulas in the decomposition of the Essential matrix into a rotation matrix and a skew-symmetric matrix using the SVD decomposition are directly proved using the SVD properties. Then, based on these results, a new method in the decomposition of the Essential matrix using SVD will be presented. The Essential matrix decomposition in this method is accomplished by extracting the base vector of the left null space of the Essential matrix and then by SVD decomposition of the skew-symmetric matrix corresponding to this base vector. In this method, the initial mapping of the Essential matrix, recovered from the erroneous coordinates of the corresponding image points in two images, into the space of Essential matrices does not require. This mapping is performed by determining the skew-symmetric matrix, S. The proposed numerical analysis shows that the results of the new presented method are correct and identical with the results of the existing formulas.
Journal of Soft Computing and Information Technology
Babol Noshirvani University of Technology
2383-1006
8
v.
2
no.
2019
76
88
https://jscit.nit.ac.ir/article_88158_b54aab6cf8ff5763680fff30f952eb5c.pdf
Sentiment Classification of Opinions based on Multi-source Transfer Learning Using Structural Correspondence Learning
Saeed
Dehghani Ashkezari
Computer Engineering Department, Faculty of Engineering, Yazd University, Safayieh, Yazd, Iran
author
Vali
Derhami
Computer Engineering Department, Faculty of Engineering, Yazd University, Safayieh, Yazd, Iran
author
ALI
Zareh Bidoki
Computer Engineering Department, Faculty of Engineering, Yazd University, Safayieh, Yazd, Iran
author
Ehsan
Basiri
Computer Engineering Department, Faculty of Engineering, Shahrekord University, Rahbar Blvd., Shahrekord, Iran
author
text
article
2019
per
Abstract :Sentiment classification of opinions is a field of Natural Language Processing which has been considered in recent years by researchers due to popularity of Internet stores and the possibility of expressing opinions about sold goods or services. To train classifier models, we need labeled datasets, but as there are not rich labeled samples and as labeling is a difficult and time-consuming process, we must employ labeled samples of other domains. In this article, a new method for binary classification of opinions is proposed based on multi-domain transfer learning. The proposed method tries to adapt different domains by using Structural Correspondence Learning; and based on repetitive procedure of the boosting algorithm, a weight is assigned to classified samples of different domains and the class of each opinion is specified by merging these classifiers. Weighting the dataset samples to boost the process of classification based on the Adaboost algorithm and combining it with the Structural Corresponding Learning is the most important innovation of the current research. The Amazon dataset of four different domains, each one containing 1000 positive and 1000 negative opinions is used for training the proposed model. Accuracy measures of %89.64, %93.97, %92.39 and %90.17 are obtained for Electronics, DVD, Books and Kitchen domains, respectively. It illustrates that the proposed method is very effective compared with the similar methods.
Journal of Soft Computing and Information Technology
Babol Noshirvani University of Technology
2383-1006
8
v.
2
no.
2019
89
101
https://jscit.nit.ac.ir/article_88271_170f0b72da66e7a400a799586643f219.pdf