A B C D E F G H I J K L M N O P Q R S T U V W X Y Z All
Toghraee, Maysam
- Evaluation Neural Networks on Selected Feature by Meta Heuristic Algorithms
Authors
1 Department of Computer Engineering, Yasouj Branch, Islamic Azad University, Yasouj, IR
2 Department of Computer Engineering, Ardabil Branch, Islamic Azad University, Ardabil, IR
3 Department of computer engineering, Iran University of Science and Technology, Tehran, IR
Source
Artificial Intelligent Systems and Machine Learning, Vol 8, No 3 (2016), Pagination: 108-115Abstract
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.Keywords
Feature Selection, Data Mining, Algorithm Cluster, Heuristic Methods.
- The Role of Bee Colony Algorithmics on Learning Machine by Data Mining Method
Authors
1 Department of Computer Engineering, Yasouj Branch, science and research, Islamic Azad University, Yasouj, Tehran, IR
2 Department of Electrical and Computer Engineering, Islamic Azad University, Qazvin Branch, Qazvin, IR
Source
Artificial Intelligent Systems and Machine Learning, Vol 11, No 9 (2019), Pagination: 157-162Abstract
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.
Keywords
Bee Colony Algorithm, Data Mining, Learning Machine, Sampling.- Heart Attack Detection in Internet of Things: Modeling with Health Markov Game Theory
Authors
1 Department of Electrical and Computer Engineering, Islamic Azad University, Qazvin Branch, Qazvin, IR
2 Department of Computer Engineering, Yasouj Branch, science and research, Islamic Azad University, Yasouj, IR
Source
Biometrics and Bioinformatics, Vol 12, No 2 (2020), Pagination: 25-30Abstract
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 into the dominant and weak dominant strategies, and combine different heart diseases with the combined strategy. Results: Based on smart systems based on hybrid strategy technology combined with Markov model and game theory, the processes of diagnosing different diseases, especially myocardial infarction, blood disease design, as well as predicting all cardiovascular diseases It is simulated in software environment.Keywords
Internet of Things, Heart Attack Detection, Game Theory, Markov Model.- Identification of Appropriate Features for Classification Using Clustering Algorithm
Authors
1 Maysam. Toghraee is with the Department of Computer Engineering,Science & Research, Yasouj Branch, Islamic Azad University, Yasouj, IR
Source
Biometrics and Bioinformatics, Vol 11, No 4 (2019), Pagination: 53-59Abstract
Now a day, 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 cannot 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, Meta heuristic methods are used to choose the best and the stable algorithms among a set of algorithms.Keywords
Feature Selection, Data Mining, Cluster Algorithm, Heuristic Methods.- Calculation of Mean Data on Gini Relationship by Data Mining Method
Authors
1 Department of Computer Engineering, Yasouj Branch, Islamic Azad University, Yasouj, IR
Source
Data Mining and Knowledge Engineering, Vol 11, No 8 (2019), Pagination: 129-133Abstract
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.