Papers by maysam toghraee
This thesis proposes a deep learning approach to bone segmentation in abdominal CNN+PG. Segmentat... more This thesis proposes a deep learning approach to bone segmentation in abdominal CNN+PG. Segmentation is a common initial step in medical images analysis, often fundamental for computer-aided detection and diagnosis systems. The extraction of bones in PG is a challenging task, which if done manually by experts requires a time consuming process and that has not today a broadly recognized automatic solution. The method presented is based on a convolutional neural network, inspired by the U-Net and trained end-to-end, that performs a semantic segmentation of the data. The training dataset is made up of 21 abdominal PG+CNN, each one containing between 0 and 255 2D transversal images. Those images are in full resolution, 4*4*50 voxels, and each voxel is classified by the network into one of the following classes: background, femoral bones, hips, sacrum, sternum, spine and ribs. The output is therefore a bone mask where the bones are recognized and divided into six different classes. In th...
In live migration, if a process is moving from one place to another, we say live migration; Migra... more In live migration, if a process is moving from one place to another, we say live migration; Migrate to a virtual machine moving from one host server or storage to another server in the following cases: preventive failure, fault tolerance, energy management, server maintenance, load balance, resource scheduling, server consolidation, and high availability (which allows the virtual machine to restart automatically if the underlying hardware fails or fails). Virtual machine state migration is the transfer from one physical host to another and is one of the techniques used to store energy and load balance in data centers
The methods available for structuring the collections are: Classification methods and clustering... more The methods available for structuring the collections are: Classification methods and clustering methods. The following is a summary of the basic principles of the text mining process. Then some of the important methods for classifying the texts are evaluated together. Clustering is the process of organizing anarchy into groups whose components are similar. A cluster is an irregular set of similarities that are heterogeneous with other components of the cluster. The goal of clustering is to achieve a steady and reliable correlation, and to identify the logical connection between them. Therefore, clustering algorithms can be used in a wide range of subject areas. Since clustering results can be varied with the number of terms used, several empirical methods are proposed to diagnose the approximate number of terms that can be expected to provide an appropriate distribution of data among clusters and to define the upper and lower limits of the clustering algorithm. Our goal is to stud...
Intrusion detection system shave been around for quite some time, to protect systems from inside ... more Intrusion detection system shave been around for quite some time, to protect systems from inside and outside threats. Researchers and scientists are concerned on how to enhance the intrusion detection performance, to be able to deal with real-time attacks and detect them fast from quick response. One way to improve performance is to use minimal number of features to define a model in a way that it can be used to accurately discriminate normal from anomalous behaviour. Many feature selection techniques are out there to reduce feature sets or extract new features out of them. In this paper, we propose an anomaly detectors generation approach using Meta heuristic algorithm in conjunction with several features selection techniques, including principle components analysis, sequential floating, and correlation-based feature selection. Meta heuristic algorithm was applied with deterministic crowding niching technique, to generate a set of detectors from a single run. In this test, based on...
International Journal of Mathematical Sciences and Computing
Artificial Intelligent Systems and Machine Learning, 2016
Feature selection is one of the issues that have been raised in the discussion of machine learnin... more Feature selection is one of the issues that have been raised in the discussion of machine learning and statistical identification model. We have provided definitions for feature selection and definitions needed to understand this issue, we check. Then, different methods for this problem were based on the type of product, as well as how to evaluate candidate subsets of features, we classify the following categories. As in previous studies may not have understood that different methods of assessment data into consideration, we propose a new approach for assessing similarity of data to understand the relationship between diversity and stability of the data is selected. After review and meta-heuristic algorithms to implement the algorithm found that the cluster algorithm has better performance compared with other algorithms for feature selection sustained.
Biometrics and Bioinformatics, 2020
Introduction: With the advent of IoT, will see exciting developments in the 21st century. Althoug... more Introduction: With the advent of IoT, will see exciting developments in the 21st century. Although most IoT is related to future products, warehousing, and smart factories, it is not limited to these, and are seeing significant improvements in the health care sector, including Reducing lab room waiting times; Remote and surveillance; Ensure access and availability of critical hardware; Personnel, patients and inventory tracking; Advanced medical management; Chronic illness management. Objective: Design an intelligent system for IoT based technology. Proposed Methods: First: Combining Game Theory Model and Proposed Markov Model and Research Status by Reviewing Articles on IoT and Intelligent Systems for Myocardial Infarction and Finally, according to Research Results, Using Soft Engineering Method MATLAB software is used to complete the design of an intelligent system for cardiac diagnosis. The purpose of this simulation is to save time; reduce the problems, combine the strategies in...
Data mining and knowledge engineering, 2019
Data clustering is one of the main tasks of data mining which have to show the hidden patterns in... more Data clustering is one of the main tasks of data mining which have to show the hidden patterns in unlabeled data. Due to inherent complexity and weakness of basic clusterings, a considerable amount of research has nowadays turned to ensemble based clusterings. Because of effectiveness of weighting in classifier ensemble it is expected that the usage of weighting can be effective in clustering ensemble. In classifier ensemble, the vote of each classifier is related to its accuracy. There, the accuracy of each classifier is approximated by testing the classifier over a test data set, but the accuracy of clustering can't be approximated at all; because it lacks supervision and also a well-known measure for estimation of accuracy. Here, research test the leukemia data using the Gini relation on meta-hierarchical algorithms and use the data mining relation to evaluate the accuracy of the algorithms over other algorithms.
Tumor is an uncontrolled growth of tissues in any part of the body. Tumors are of different varie... more Tumor is an uncontrolled growth of tissues in any part of the body. Tumors are of different varieties and that they have totally different Characteristics and different treatment. As it is thought, brain tumor is inherently serious and serious due to its character within the restricted area of the intracranial cavity (space shaped within the skull).Most analysis in developed countries show that the number of individuals who have brain tumors were died because of the actual fact of inaccurate detection. Generally, CT scan or mri that's directed into intracranial cavity produces an entire image of brain. This image is visually examined by the physician for detection & diagnosis of brain tumour. But this methodology of detection resists the accurate determination of stage & size of tumor. To avoid that, this project uses computer aided methodology for segmentation (detection) of brain tumour supported the combination of two algorithms. This technique permits the segmentation of tum...
International Journal of Modern Education and Computer Science
International Journal of Information Technology and Computer Science, 2016
Artificial Intelligent Systems and Machine Learning, 2019
What the intellectual basis of the bee colony algorithm is based on can be easily stated in one s... more What the intellectual basis of the bee colony algorithm is based on can be easily stated in one sentence: Bees always choose the best way out of various obstacles in nature to access food. The bees choose the shortest route from the different paths to the food, with the bees secreting substances from the pheromone after finding the food, which is traced to white after rain. The bees find the path above when faced with a path that has more pheromones. In this paper, the studies show that the learning machine data set, which is a sampling of leukemia data, tested the accuracy of the error measurement on the bee colony algorithm, and the accuracy of the error before and after. It looks bad when performing data execution.
Journal of Computer Based Parallel Programming, 2021
In live migration, if a process is moving from one place to another, we say live migration; Migra... more In live migration, if a process is moving from one place to another, we say live migration; Migrate to a virtual machine moving from one host server or storage to another server in the following cases: preventive failure, fault tolerance, energy management, server maintenance, load balance, resource scheduling, server consolidation, and high availability (which allows the virtual machine to restart automatically if the underlying hardware fails or fails). Virtual machine state migration is the transfer from one physical host to another and is one of the techniques used to store energy and load balance in data centers .
VLSI Design and Signal Processing, 2021
Gas neural network or neural gas network is one of the types of competitive neural networks with ... more Gas neural network or neural gas network is one of the types of competitive neural networks with unsupervised learning pattern, whose main application is in solving clustering problems and learning topology. This type of neural network, in terms of classification, falls into the category of Vector Quantification algorithms (VQ) and is very closely related to the k-Means clustering algorithm, SOM neural network (selforganizing maps) and LVQ neural network. Purpose: The purpose of this article is that we should be able to convert very large and huge data into very small and useful data. has it. In addition to clustering and placing the center of the clusters in the right place, the neural gas network dynamically creates neighborhood connections between neurons (cluster centers), which ultimately enables this algorithm to learn topology.
Intrusion detection system shave been around for quite some time, to protect systems
from inside ... more Intrusion detection system shave been around for quite some time, to protect systems
from inside and outside threats. Researchers and scientists are concerned on how to
enhance the intrusion detection performance, to be able to deal with real-time attacks and
detect them fast from quick response. One way to improve performance is to use minimal
number of features to define a model in a way that it can be used to accurately
discriminate normal from anomalous behaviour. Many feature selection techniques are
out there to reduce feature sets or extract new features out of them. In this paper, we
propose an anomaly detectors generation approach using Meta heuristic algorithm in
conjunction with several features selection techniques, including principle components
analysis, sequential floating, and correlation-based feature selection. Meta heuristic
algorithm was applied with deterministic crowding niching technique, to generate a set of
detectors from a single run. In this test, based on various algorithms, we conclude that
NWINE data is low in accuracy and only in the clustering algorithm, which rises
precisely.
The methods available for structuring the collections are: Classification methods and clustering ... more The methods available for structuring the collections are: Classification methods and clustering methods. The following is a summary of the basic principles of the text mining process. Then some of the important methods for classifying the texts are evaluated together. Clustering is the process of organizing anarchy into groups whose components are similar. A cluster is an irregular set of similarities that are heterogeneous with other components of the cluster. The goal of clustering is to achieve a steady and reliable correlation, and to identify the logical connection between them. Therefore, clustering algorithms can be used in a wide range of subject areas. Since clustering results can be varied with the number of terms used, several empirical methods are proposed to diagnose the approximate number of terms that can be expected to provide an appropriate distribution of data among clusters and to define the upper and lower limits of the clustering algorithm. Our goal is to study data mining on different data, the results of this method show that the meta-heuristic method is suitable for the meta cluster algorithm compared to other clusters.
—Feature selection is one of the issues that have been raised in the discussion of machine learni... more —Feature selection is one of the issues that have been raised in the discussion of machine learning and statistical identification model. We have provided definitions for feature selection and definitions needed to understand this issue, we check. Then, different methods for this problem were based on the type of product, as well as how to evaluate candidate subsets of features, we classify the following categories. As in previous studies may not have understood that different methods of assessment data into consideration, We propose a new approach for assessing similarity of data to understand the relationship between diversity and stability of the data is selected. After review and meta-heuristic algorithms to implement the algorithm found that the cluster has better performance compared with other algorithms for feature selection sustained.
—Now a days, developing the science and technology and technology tools, the ability of reviewing... more —Now a days, developing the science and technology and technology tools, the ability of reviewing and saving the important data has been provided. It is needed to have knowledge for searching the data to reach the necessary useful results. Data mining is searching for big data sources automatically to find patterns and dependencies which are not done by simple statistical analysis. The scope is to study the predictive role and usage domain of data mining in medical science and suggesting a frame for creating, assessing and exploiting the data mining patterns in this field. As it has been found out from previous researches that assessing methods can not be used to specify the data discrepancies, our suggestion is a new approach for assessing the data similarities to find out the relations between the variation in data and stability in selection. Therefore we have chosen meta heuristic methods to be able to choose the best and the stable algorithms among a set of algorithms.
Conference Presentations by maysam toghraee
Feature selection is one of the issues that have been raised in the discussion of machine learnin... more Feature selection is one of the issues that have been raised in the discussion of machine learning and statistical identification model. We have provided definitions for feature selection and definitions needed to understand this issue, we check. Then, different methods for this problem were based on the type of product, as well as how to evaluate candidate subsets of features, we classify the following categories. As in previous studies may not have understood that different methods of assessment data into consideration, We propose a new approach for assessing similarity of data to understand the relationship between diversity and stability of the data is selected. After review and meta-heuristic algorithms to implement the algorithm found that the cluster algorithm has better performance compared with other algorithms for feature selection sustained.
Uploads
Papers by maysam toghraee
from inside and outside threats. Researchers and scientists are concerned on how to
enhance the intrusion detection performance, to be able to deal with real-time attacks and
detect them fast from quick response. One way to improve performance is to use minimal
number of features to define a model in a way that it can be used to accurately
discriminate normal from anomalous behaviour. Many feature selection techniques are
out there to reduce feature sets or extract new features out of them. In this paper, we
propose an anomaly detectors generation approach using Meta heuristic algorithm in
conjunction with several features selection techniques, including principle components
analysis, sequential floating, and correlation-based feature selection. Meta heuristic
algorithm was applied with deterministic crowding niching technique, to generate a set of
detectors from a single run. In this test, based on various algorithms, we conclude that
NWINE data is low in accuracy and only in the clustering algorithm, which rises
precisely.
Conference Presentations by maysam toghraee
from inside and outside threats. Researchers and scientists are concerned on how to
enhance the intrusion detection performance, to be able to deal with real-time attacks and
detect them fast from quick response. One way to improve performance is to use minimal
number of features to define a model in a way that it can be used to accurately
discriminate normal from anomalous behaviour. Many feature selection techniques are
out there to reduce feature sets or extract new features out of them. In this paper, we
propose an anomaly detectors generation approach using Meta heuristic algorithm in
conjunction with several features selection techniques, including principle components
analysis, sequential floating, and correlation-based feature selection. Meta heuristic
algorithm was applied with deterministic crowding niching technique, to generate a set of
detectors from a single run. In this test, based on various algorithms, we conclude that
NWINE data is low in accuracy and only in the clustering algorithm, which rises
precisely.