جستجوی مقالات مرتبط با کلیدواژه "Neural Network" در نشریات گروه "فناوری اطلاعات"

In power systems, the transmission and distribution networks of electrical energy rely heavily on the performance of various equipment. Any malfunction within these systems can lead to network interruptions, short circuits, and power failures. Arresters are critical devices used to limit transient overvoltages caused by lightning strikes and switching events in transmission and distribution networks. These arresters protect equipment from transient overvoltages while ensuring that they do not react to temporary overloads. Their effectiveness is influenced by environmental conditions, such as humidity and pollution. This research aims to analyze the factors affecting voltage and energy absorption during lightning strikes on power systems. Additionally, we focus on designing an artificial neural network (ANN) to estimate the energy absorbed by the arrester, minimizing the error of this neural network. The results demonstrate that the ANN can effectively estimate the power of the arrester within the power system, providing a valuable tool for enhancing system reliability and performance. This study contributes to the understanding of arrester behavior under transient conditions and offers a novel approach to estimating their energy absorption capabilities using advanced computational techniques.

The energy sector encompasses essential processes such as the production, distribution, and consumption of energy. Traditionally, these processes have been managed through conventional networks, which often lead to issues such as process fluctuations, increased costs, and inefficiencies. However, the advent of Internet of Energy technology facilitates a transition from traditional to smart networks. In the Internet of Energy, the use of sensors results in the generation of large volumes of data. By employing machine learning to analyze this data, it becomes possible to make accurate predictions in the energy sector, which in turn supports effective decision-making for energy production and distribution. The objective of this study is to analyze data within the Internet of Energy using machine learning techniques, ultimately leading to the development of an artificial intelligence model capable of predicting energy consumption. Initially, previous models will be reviewed, and their outcomes will be compared and analyzed based on scores and evaluation metrics. Finally, a deep neural network model will be introduced, demonstrating an error rate of 0.3. The mean absolute error is reported as 0.4, and the mean square error is 0.3. Despite these advantages, there are also limitations to consider. The data involved in the analysis and prediction process must meet appropriate standards. The significant variability present in industrial processes adds complexity to the environment.

Nowdays, organizations generate a significant volume of log files that require processing for condition checking, debugging, and anomaly resolution. Outsourcing such processing is not suitable due to the need for real-time processing and security maintenance. Given the multitude of different software and services, organizations face a substantial volume of production logs that should be processed rather than deleted or ignored. In the traditional approach, experts manually check the logs daily. This, on one hand, slows down the process, increases the time and inaccuracy, and, on the other hand, results in a high hiring cost due to the need for an expert force. This article introduces a solution that employs generative neural networks to establish a local structure for log analysis within the organization. The process involves retrieving and parsing text files from various sectors, segmenting them into manageable portions, embedding them, and storing them in a vector database. In this structure, a trained individual without special expertise can quickly access necessary information using appropriate prompts from a local language model available at any time. As a result, three overarching goals are achieved: maintaining security, increasing the speed of analysis, and reducing human resource costs.

Contemporary machine learning models, like deep neural networks, require substantial labeled datasets for proper training. However, in areas such as natural language processing, a shortage of labeled data can lead to overfitting. To address this challenge, data augmentation, which involves transforming data points to maintain class labels and provide additional valuable information, has become an effective strategy. In this paper, a deep reinforcement learning-based text augmentation method for sentiment analysis was introduced, combining reinforcement learning with deep learning. The technique uses Deep Q-Network (DQN) as the reinforcement learning method to search for an efficient augmentation strategy, employing four text augmentation transformations: random deletion, synonym replacement, random swapping, and random insertion. Additionally, various deep learning networks, including CNN, Bi-LSTM, Transformer, BERT, and XLNet, were evaluated for the training phase. Experimental findings show that the proposed technique can achieve an accuracy of 65.1% with only 20% of the dataset and 69.3% with 40% of the dataset. Furthermore, with just 10% of the dataset, the method yields an F1-score of 62.1%, rising to 69.1% with 40% of the dataset, outperforming previous approaches. Evaluation on the SemEval dataset demonstrates that reinforcement learning can efficiently augment text datasets for improved sentiment analysis results.

Air pollution is the biggest environmental hazard that cannot be ignored. Due to increase in number of industries and urbanization increases air pollutants concentrations in many areas because of this different changes are been happening in human life like health issues and as well as other living organisms. We have some pollutant emission monitoring systems, like Opsis, Codel, Urac and TAS-Air metrics which are expensive. As well as these systems have limitations to be installed on chimney due to their principle of operation. In this work I like to propose a function that is easy to use and causes less cost compared to the other ones. That is an industrial air pollution monitoring system based on the technology of Wireless Sensor Networks (WSNs). This system is integrated with the Global System for Mobile (GSM) communications and the protocol it uses is zigbee. The system consists of sensor nodes, a control center and data base through which sensing data can be stored for history and future plans. It is used to monitor Carbon Monoxide (CO), Sulfur Dioxide (SO2) and dust concentration caused by industrial emissions due to process.

The continuous progress of photography technologies as well as the increase in the number of images and their applications requires the emergence of new algorithms with new and different capabilities. Among the various processes on medical images, the segmentation of medical images has a special place and has always been considered and investigated as one of the important issues in the processing of medical images. Based on this, in this research, a solution to diagnose the tumor through the use of a combined method based on watershed algorithm, co-occurrence matrix and neural networks has been presented, so that through the use of this combined solution, the tumor can be detected with high accuracy. Medical images diagnosed. According to the method used in this research, as well as the implementation of the solution in the Python environment and through the use of CV2 and SimpleITK modules, it is possible to set parameters such as accuracy, correctness, recall and Fscore criteria. which are always important parameters that are investigated in researches, improved compared to the past and achieved favorable results. This will increase the improvement of tumor detection in the brain compared to Thersholding and TKMeans methods.

We all know forest is very important resource of oxygen. Saving our environmental resources is human beings responsibility. One of the techniques to save forests is forest fire detection. This is a technique used to detect the fire and prevent them in less time. Forest fire leads to death of wild life and trees. There are other techniques used to detect fire in forests like cameras, satellite system, manual monitoring but they take time to detect the fire whereas Forest fire detection system detects the fire within seconds and triggers the alarms. In this way we can save tress and wildlife in very less time.

Household object detection is a brand-new computer technique that combines image processing and computer vision to recognize objects in the home. All objects stored in the kitchen, room, and other areas will be detected by the camera. Low-end device techniques for detecting people in video or images are known as object detection. With picture and video analysis, we've lost our way.

Integration and diversity of IOT terminals and their applicable programs make them more vulnerable to many intrusive attacks. Thus, designing an intrusion detection model that ensures the security, integrity, and reliability of IOT is vital. Traditional intrusion detection technology has the disadvantages of low detection rates and weak scalability that cannot adapt to the complicated and changing environment of the Internet of Things. Hence, one of the most widely used traditional methods is the use of neural networks and also the use of evolutionary optimization algorithms to train neural networks can be an efficient and interesting method. Therefore, in this paper, we use the PSO algorithm to train the neural network and detect attacks and abnormalities of the IOT system. Although the PSO algorithm has many benefits, in some cases it may reduce population diversity, resulting in early convergence. Therefore,in order to solve this problem, we use the modified PSO algorithm with a new mutation operator, fuzzy systems and comparative equations. The proposed method was tested with CUP-KDD data set. The simulation results of the proposed model of this article show better performance and 99% detection accuracy in detecting different malicious attacks, such as DOS, R2L, U2R, and PROB.

Household object detection is a brand-new computer technique that combines image processing and computer vision to recognise objects in the home. All objects stored in the kitchen, room, and other areas will be detected by the camera. Low-end device techniques for detecting people in video or images are known as object detection. With picture and video analysis, we've lost our way.

With the accelerated development of Internet finance, electronic funds transfer, and the rapid growth of credit card activity, credit cards play a very important role in every area of ​​life today. There are some risks in this regard that are considered serious threats to both issuers and cardholders. The increasing number of fraudulent credit card transactions forged credit cards and fraudulent use of expired credit cards have led to increased losses. Therefore, finding fraud detection techniques accurately and quickly has become an important topic in current investigations. In this study, after normalizing and reducing the dimensionality of the data using the PCA algorithm, we used the modified perceptron neural network and the grasshopper algorithm to classify the data. In this study, we use the grasshopper algorithm to adjust the weights and biases of neural networks. In the end, we were able to achieve 99.20% accuracy.

The growing number of information on the web and the addition of different web pages and websites to this space has made users face problems. These problems appear to users when users are trying to obtain information on a particular topic, and finding all the pages that are suggested to them is a difficult and time consuming process. In the current research, a profile is first created based on the behavioral characteristics of users at different sessions that result from web server logs. These include things like the frequency of user page views, the length of time the user has been on different pages, and the date the page was viewed. We then group them using the clustering method, then fuzzy inference system, extract the fuzzy rules according to the interests of the users and their clusters, and after obtaining the users’ movement patterns, they Insertneural network into vector format Other tools such as bio-algorithms can be useful by obtaining optimal parameters in optimizing predictions and increasing accuracy in fuzzy neural network. The evaluation criteria in this study is accuracy.

In this study, an alternative method is proposed based on recursive deep learning with limited steps and prepossessing, in which the data is divided into A unit classes in order to change a long short term memory and solve the existing challenges. The goal is to obtain predictive results that are closer to real world in COVID-19 patients. To achieve this goal, four existing challenges including the heterogeneous data, the imbalanced data distribution in predicted classes, the low allocation rate of data to a class and the existence of many features in a process have been resolved. The proposed method is simulated using the real data of COVID-19 patients hospitalized in treatment centers of Tehran treatment management affiliated to the Social Security Organization of Iran in 2020, which has led to recovery or death. The obtained results are compared against three valid advanced methods, and are showed that the amount of memory resources usage and CPU usage time are slightly increased compared to similar methods  and the accuracy is increased by an average of 12%.

Providing efficient and powerful approach for liquidity management of bank branches has always been one of the most important and challenging issues for researchers and scholars in the banking field. In other words, estimating the amount of required cash in different branches of the bank is one of the basic and important questions for managers of the banking system. Because on the one hand, if the amount of cash is less than the required amount, the bank runs the default risk, and on the other hand, if the amount of cash is more than the required amount, the bank incurs opportunity costs. Therefore, the purpose of this study is to provide a practical approach to predict the optimal amount of required cash in bank branches. For this purpose, the concepts of time series, neural network approach and vector autoregressive model are used. The effectiveness of the proposed approach is also examined using real data.