Neural networks

Open Access Article

1 - A Method Based on Learning Automata for Adaptation of the Vigilance factor in Fuzzy ARTMAP Network
M. Anjidani M. R. Meybodi

20.1001.1.16823745.1384.3.6.3.8

In this paper, a method based on learning automata for adaptation of the vigilance factor in Fuzzy ARTMAP network when used for classification problems is proposed. The performance of the proposed algorithm is independent of the initial value for vigilance factor. Fuzz More

In this paper, a method based on learning automata for adaptation of the vigilance factor in Fuzzy ARTMAP network when used for classification problems is proposed. The performance of the proposed algorithm is independent of the initial value for vigilance factor. Fuzzy ARTMAP network in which the vigilance factor adapted using learning automata generates smaller structure with higher recognition rate. To study the performance of the proposed method it has been applied to several problems: circle in square, spirals and square in square problems. The results of experiments show the effectiveness of the proposed method. Manuscript profile

Open Access Article

2 - A New Ensemble Learning Method for Improvement of Classification Performance
S. H. Nabavi-Kerizi E. Kabir

20.1001.1.16823745.1385.4.8.7.3

The combination of multiple classifiers is shown to be suitable for improving the performance of pattern recognition systems. Combining multiple classifiers is only effective if the individual classifiers are accurate and diverse. The methods have been proposed for dive More

The combination of multiple classifiers is shown to be suitable for improving the performance of pattern recognition systems. Combining multiple classifiers is only effective if the individual classifiers are accurate and diverse. The methods have been proposed for diversity creation can be classified into implicit and explicit methods. In this paper, we propose a new explicit method for diversity creation. Our method adds a new penalty term in learning algorithm of neural network ensembles. This term for each network is the product of its error and the sum of other networks errors. Experimental results on different data sets show that proposed method outperforms the independent training and the negative correlation learning methods. Manuscript profile

Open Access Article

3 - Determining of Classifiers Behavior Using Hidden Markov Model Based Decision Template
H. Sadoghi Yazdi

20.1001.1.16823745.1387.6.14.10.2

Studying of classifier behavior is interested from viewpoint of error checking and presentation of suitable solution for decreasing error rates and decreasing performance. Weakness operation of recognition system is because of small number of training samples, noisy sam More

Studying of classifier behavior is interested from viewpoint of error checking and presentation of suitable solution for decreasing error rates and decreasing performance. Weakness operation of recognition system is because of small number of training samples, noisy samples, unsuitable extracted features, method of determining of system response. Presentation of suitable model for behavior or response of recognition system, we can improve operation of recognition system. In this paper, a new hidden Markov model based decision template is generated for modeling of neurons behavior in neural network. In existing methods, relation of neurons and interaction between them is not studied whereas; response of neural network includes response value of all neurons. So, relations of neurons are modeled using new hidden Markov decision templates. This method is used into three applications include recognition of Farsi number images, normal traffic in internet network, and recognition of types of vehicles. Increasing performance of neural network indicates to superiority of the proposed system. Manuscript profile

Open Access Article

4 - A Simple and Effective Scheme for Hand Gesture Recognition in Finger Spelling of Farsi Alphabet
M. J. Barzegar Sakhvidi A. R. Sharafat

20.1001.1.16823745.1388.7.4.7.6

In recent years, automated recognition of gestures in the finger spelling paradigm has become an active research area. Gesture is a combination of hand postures, hand movements, and face gestures; and finger spelling is a way of presenting alphabets of a word that does More

In recent years, automated recognition of gestures in the finger spelling paradigm has become an active research area. Gesture is a combination of hand postures, hand movements, and face gestures; and finger spelling is a way of presenting alphabets of a word that does not exist in the sign language dictionary. In this paper, we present a scheme for hand gesture recognition in finger spelling of Farsi alphabets, where a different shape for hand and fingers denote a different letter in the alphabet. Our scheme has five stages, namely, visual data gathering, preprocessing of the image, detection and extraction of hand’s features, feature reduction and consolidation, and finally, hand gesture recognition. For the last stage (hand gesture recognition), we employ three techniques, namely, the nearest neighbor using the Euclidian distance, the nearest neighbor using the normalized Euclidian distance, and neural networks. For reducing the feature space, we use the discrete cosine transform (DCT), which yields better results as compared to the discrete Fourier transform and Fourier coefficients. We achieved 99.1% correct recognition using neural networks, which is superior to existing schemes. Manuscript profile

Open Access Article

5 - Phrase Segmentation on Persian Texts Using Neural Networks
M. M. Mirdamadi A. M. Zareh Bidoki M. Rezaeian

20.1001.1.16823745.1392.11.4.2.9

Word and phrase segmentation is one of the main activities in natural languages processing (NLP). Many programs in NLP need to be preprocessed for extraction of text’s words and distinction phrases. Getting meaningful words with their prefix and suffix is the main and t More

Word and phrase segmentation is one of the main activities in natural languages processing (NLP). Many programs in NLP need to be preprocessed for extraction of text’s words and distinction phrases. Getting meaningful words with their prefix and suffix is the main and the final goal of segmentation. This activity depends on various natural languages can be easy or hard. Persian is among the languages with complex preprocessing tasks. One of the complexity sources is handling different writing scripts. In written Persian texts, we have two kinds of spaces: short space and white space. Also there are various scripts for writing Persian texts, differing in the style of writing words, using or elimination of spaces within or between words, using various forms of characters and so on. In this paper, we want to suggest a statistical method for phrase segmentation on Persian texts using neural networks due to using in search engines. For this purpose, we use occurrence likelihood of uniwords and biwords in corpus. The suggested algorithm includes four steps and could detect about 89.6% of correct tokens. Experimental results show this method can improve the performance of the usual methods Manuscript profile

Open Access Article

6 - Proposing a Robust Method Against Adversarial Attacks Using Scalable Gaussian Process and Voting
Mehran Safayani Pooyan Shalbafan Seyed Hashem Ahmadi Mahdieh Falah aliabadi Abdolreza Mirzaei

20.1001.1.16823745.1400.19.4.9.2

In recent years, the issue of vulnerability of machine learning-based models has been raised, which shows that learning models do not have high robustness in the face of vulnerabilities. One of the most well-known defects, or in other words attacks, is the injection of More

In recent years, the issue of vulnerability of machine learning-based models has been raised, which shows that learning models do not have high robustness in the face of vulnerabilities. One of the most well-known defects, or in other words attacks, is the injection of adversarial examples into the model, in which case, neural networks, especially deep neural networks, are the most vulnerable. Adversarial examples are generated by adding a little purposeful noise to the original examples so that from the human user's point of view there is no noticeable change in the data, but machine learning models make mistakes in categorizing the data. One of the most successful methods for modeling data uncertainty is Gaussian processes, which have not received much attention in the field of adversarial examples. One reason for this could be the high computational volume of these methods, which limits their used in the real issues. In this paper, a scalable Gaussian process model based on random features has been used. This model, in addition to having the capabilities of Gaussian processes for proper modeling of data uncertainty, is also a desirable model in terms of computational cost. A voting-based process is then presented to deal with adversarial examples. Also, a method called automatic relevant determination is proposed to weight the important points of the images and apply them to the kernel function of the Gaussian process. In the results section, it is shown that the proposed model has a very good performance against fast gradient sign attack compared to competing methods. Manuscript profile

Open Access Article

7 - Numeric Polarity Detection based on Employing Recursive Deep Neural Networks and Supervised Learning on Persian Reviews of E-Commerce Users in Opinion Mining Domain
Sepideh Jamshidinejad Fatemeh Ahmadi-Abkenari Peiman Bayat

20.1001.1.16823745.1400.19.3.15.6

Opinion mining as a sub domain of data mining is highly dependent on natural language processing filed. Due to the emerging role of e-commerce, opinion mining becomes one of the interesting fields of study in information retrieval scope. This domain focuses on various s More

Opinion mining as a sub domain of data mining is highly dependent on natural language processing filed. Due to the emerging role of e-commerce, opinion mining becomes one of the interesting fields of study in information retrieval scope. This domain focuses on various sub areas such as polarity detection, aspect elicitation and spam opinion detection. Although there is an internal dependency among these sub sets, but designing a thorough framework including all of the mentioned areas is a highly demanding and challenging task. Most of the literatures in this area have been conducted on English language and focused on one orbit with a binary outcome for polarity detection. Although the employment of supervised learning approaches is among the common utilizations in this area, but the application of deep neural networks has been concentrated with various objectives in recent years so far. Since the absence of a trustworthy and a complete framework with special focuses on each impacting sub domains is highly observed in opinion mining, hence this paper concentrates on this matter. So, through the usage of opinion mining and natural language processing approaches on Persian language, the deep neural network-based framework called RSAD that was previously suggested and developed by the authors of this paper is optimized here to include the binary and numeric polarity detection output of sentences on aspect level. Our evaluation on RSAD performance in comparison with other approaches proves its robustness. Manuscript profile

Open Access Article

8 - An Intelligent Vision System for Automatic Forest Fire Surveillance
Mohammad Sadegh Kayhanpanah Behrooz Koohestani

20.1001.1.16823745.1401.20.1.3.7

Fighting forest fires to avoid their potential dangers as well as protect natural resources is a challenge for researchers. The goal of this research is to identify the features of fire and smoke from the unmanned aerial vehicle (UAV) visual images for classification, o More

Fighting forest fires to avoid their potential dangers as well as protect natural resources is a challenge for researchers. The goal of this research is to identify the features of fire and smoke from the unmanned aerial vehicle (UAV) visual images for classification, object detection, and image segmentation. Because forests are highly complex and nonstructured environments, the use of the vision system is still having problems such as the analogues of flame characteristics to sunlight, plants, and animals, or the smoke blocking the images of the fire, which causes false alarms. The proposed method in this research is the use of convolutional neural networks (CNNs) as a deep learning method that can automatically extract or generate features in different layers. First, we collect data and increase them according to data augmentation methods, and then, the use of a 12-layer network for classification as well as transfer learning method for segmentation of images is proposed. The results show that the data augmentation method used due to resizing and processing the input images to the network to prevent the drastic reduction of the features in the original images and also the CNNs used can extract the fire and smoke features in the images well and finally detect and localize them. Manuscript profile

Open Access Article

9 - A Step towards All-Optical Deep Neural Networks: Utilizing Nonlinear Optical Element
Aida Ebrahimi Dehghan Pour S. K.

20.1001.1.16823745.1401.20.3.2.0

In recent years, optical neural networks have received a lot of attention due to their high speed and low power consumption. However, these networks still have many limitations. One of these limitations is implementing their nonlinear layer. In this paper, the implement More

In recent years, optical neural networks have received a lot of attention due to their high speed and low power consumption. However, these networks still have many limitations. One of these limitations is implementing their nonlinear layer. In this paper, the implementation of nonlinear unit for an optical convolutional neural network is investigated, so that using this nonlinear unit, we can realize an all-optical convolutional neural network with the same accuracy as the electrical networks, while providing higher speed and lower power consumption. In this regard, first of all, different methods of implementing optical nonlinear unit are reviewed. Then, the impact of utilizing saturable absorber, as the nonlinear unit in different layers of CNN, on the network’s accuracy is investigated, and finally, a new and simple method is proposed to preserve the accuracy of the optical neural networks utilizing saturable absorber as the nonlinear activating function. Manuscript profile

Open Access Article

10 - Spam Detection in Twitter by Ensemble Learning Approach
Maryam Fasihi Mohammad Javad shayegan zahra hosieni zahra sejdeh

Today, social networks play a crucial role in disseminating information worldwide. Twitter is one of the most popular social networks, with 500 million tweets sent on a daily basis. The popularity of this network among users has led spammers to exploit it for distributi More

Today, social networks play a crucial role in disseminating information worldwide. Twitter is one of the most popular social networks, with 500 million tweets sent on a daily basis. The popularity of this network among users has led spammers to exploit it for distributing spam posts. This paper employs a combination of machine learning methods to identify spam at the tweet level. The proposed method utilizes a feature extraction framework in two stages. In the first stage, Stacked Autoencoder is used for feature extraction, and in the second stage, the extracted features from the last layer of Stacked Autoencoder are fed into the softmax layer for prediction. The proposed method is compared and evaluated against some popular methods on the Twitter Spam Detection corpus using accuracy, precision, recall, and F1-score metrics. The research results indicate that the proposed method achieves a detection of 78.1%. Overall, the proposed method, using the majority voting approach with a hard selection in ensemble learning, outperforms CNN, LSTM, and SCCL methods in identifying spam tweets with higher accuracy. Manuscript profile

Current Issue

Published Issues

Menu

Browse

List of Articles Neural networks