117
Archived Theses
0
DOIs Assigned
0%
DOI Rate
Archived Theses
Yazılım tanımlı ağ ile kablosuz sensör ağı tasarlayın
This paper present design to wireless sensor networks based on Software Define networks .in this paper we provide some examples that show how architecture can use for networking .We also identify how to design WSN network with SDN controller and how to evaluate the performance of this network. Our design of WSN network consist of one sink node and twentyfour WSN node. The network is first tested without using SDN controller. Then the WSN network is operated with SDN controller. Both scenarios are evaluated by studding and analysis the main performance parameter (Received packet, received power, Throughput) and a comparisons is made. The received packet of WSN with SDN controller is increased by (53%) compared by received packet of WSN network without SDN controller. The throughput of WSN network with SDN controller is increased by (53.2%) compared by throughput of WSN network with SDN controller. The received power of WSN network with SDN controller is increased by (3%) compared by received power of WSN network without SDN. Also in this paper we implement networks of WSN and SDN using Real Node (Z1 mote) . From the last network we can see that the we can see small different in result between Real Network using (Z1 mote ) and simulated node Throughput is increased due to the increase in receive packet
Genişletilmiş derin evrişimsel sinir ağı kullanarak göğüs kanseri tespiti ve görüntü değerlendirmesi
Breast malignancy is one of the primary driver of disease demise around the world. Early diagnostics essentially builds the odds of right treatment and survival, however this procedure is dull and regularly prompts a contradiction between pathologists. PC supported conclusion frameworks indicated potential for enhancing the demonstrative precision. In this work, we build up the computational methodology dependent on augmented deep convolution neural systems for bosom malignant growth histology picture characterization. Our methodology uses a few deep neural system structures and inclination helped trees classifier. For 3-class grouping undertaking to recognize benign, malignant and normal/invasive. We report 88.3% exactness, 86.2%, and affectability at the high-affectability working point. As far as anyone is concerned, this methodology performs other basic techniques in computerized image grouping. After testing different network architectures and training configurations, we showed that deep convolutional networks are able to segment breast cancer lesions with promising results. Furthermore, this performance will only improve as richer data sets become available. We highly encourage research in this direction. The techniques we utilized in this work are ground-breaking and our outcomes can be enhanced just by the methods for applying more computational assets without fundamentally changing the strategy. In this report, we suggest a straightforward and powerful strategy for the order of recolored histological bosom malignant growth images in the circumstance of little preparing for detection and classification of cancerous tumors
Güç spektral yoğunluğuna (PSD) dayalı veri madenciliği teknikleri kullanarak otomatik kötü amaçlı yazılım tespiti
A malware is a software that furtively achieves its process below the appearance of genuine software. classic methods apply signatures to distinguish these software's denote tiny risk to new and hidden examples whose signatures are not offered. The emphasis of malware investigation is unstable from applying signature designs to classifying the malicious conduct showed by this malware. Numerous data mining methods proposed to notice malware mechanically in the effectual face. In this thesis, Power Spectral Density (PSD) applied to extract the features of malware dataset and the yield of PSD confidential applying several of data mining methods: Support Vector Machine (SVM), Radial Basis Network (RBF) and multi-layer perceptron (MLP). These techniques presented remarkable results when compared with common researches in this field. Keywords: Malware, Data mining, Power spectral density, computer security.
Kablosuz algılayıcı ağ (KAA) öngörülebilir hava istasyonu
The object of this research is to build a wireless weather station that able to make a decision about playing in the right weather condıtıons. The weather station is designed to collect data from the environment. There are four sensors used to collect the required data. These sensors are temperature and humidity sensor, Light sensor, Rain detection sensor and dc motor as a wind sensor. The weather station contains two part. The first part placed outdoor for collecting data and sending them with nrf24l01 transceiver module as a sender. The second part placed indoor to receiving data from the outdoor station using the nrf24l01 transceiver module as a receiver. The electronic circuit of the indoor station designed using Arduino Uno. The electronic circuit of the indoor station designed using Arduino Nano. Each station provided with a Lipo battery with 350 mAh capacity and 7.4 volts. The nrf24l01 module has the ability to transmitting data with 100 meters. The decision tree algorithm used to make the decision with famous dataset knowing as play golf dataset. The indoor station has an LCD screen to display the weather conditions and after that showing the decision by printing yes or no. In conclusion; a predictable weather station was designed. The Tasks of the station examined. It can be concluded that the weather station is ready to accumulate data from every environment. Also, it is able to make a decision about playing the game or not playing. Keywords: Weather station, Arduino, Sensors, Decision Tree.
HESSDS analizlerinin twitter verilerinde Keullehiler makensi algoritme lerenin òğrenimi
Massive amounts of data are generated by social media users for each second, such as posts, tweets, images, and videos. Getting valuable information from this big data is a significant, challenging and interesting issue in the text mining area. Twitter data are analyzed with text mining techniques to discover society agenda, trends, user behaviors, and feelings. We proposed a text analysis method to determine sentiments from tweets. Natural language processing techniques are carried out to put the data into meaningful context. After that classification model is trained with data mining methods on the processed data. It carries out the classification label as people's opinion, such as positive, negative, and neutral sentiments, using Twitters streaming data. We select imdb and collect tweets with hashtags about these brands by using twitter API.
Makinenin örenmesi ile şeker hastalığını taşhisi
Artificial neural networks have been in the position of producing complex dynamics in control applications over the last decade, especially when they are linked to feedback. Although ANNs are strong for network design, the harder the design of the network, the more complex the desired dynamic is. Many researchers tried to automate the design process of ANN using computer programs. Search and optimization problems can be considered as the problem of finding the best parameter set for a network to solve a problem. Recently, the problem of optimizing ANN parameters to train different research datasets has been targeted by two commonly used stochastic genetic algorithms (GA) and particle swarm optimization (PSO). The process based on the neural network is optimized with GA and PSO to enable the robot to perform complex tasks. However, using such optimization algorithms to optimize the ANN training process cannot always be balanced or successful. These algorithms simultaneously aim to develop three main components of an ANN: synaptic weight, connections, architecture and transfer functions set for each neuron. Developed with the proposed approach, ANN is also compared with hand-designed Levenberg-Marquardt and Back Propagation algorithms.
Çoklu platform genel dinamik web patterleri
It is complex and expensive to develop generic web-based software applications. Software solutions must understand each platform, technology and architecture's behavior and environment. The objective of the work is to produce a generic web pattern generator for three platforms of dynamic web developers: the Java EE, PHP Net. We have used the generic design architecture of the MVC (Model-View-Controller). Based on their specific technologies, the analysis of each layer within MVC can be analyzed. The lists of technologies being explored are as follows: Layers: Java EE: JSF and JSP .NET: Razor and Blade .NET; PHP: Twig and Blade; controllers layer: Java EE: Spring MVC .NET: .NET Framework. PHP: Laravel; Models Layers: Java EE: Hibernate and EclipseLink. We have developed a software application for student enrolment in courses to validate our proposal. The application has been implemented on the basis of all previously described platforms, technologies and architectures. A functional test to test the dynamic web functionality has been conducted. We have also conducted compliance tests to verify that the rules, specifications and structures have been followed by the specific platforms defined.
ı
In this document we have shown the obstacles that Iraqi governmental institutions faced to apply the electronic management. We have covered the most issues in this study that related works has been published. By the literature review and meeting with some professionals in this field of study we designed our questions. We conducted this research in Baghdad city because the key informative were in Baghdad and because it is the capital of the Iraq. The survey was distributed among governmental employees through the internet. We found a number of issues and real challenges that institutions have, most of them were organizational and technical issues.
Gradyan arttırma makinesini kullanarak meme kanseri tahmini
Breast Cancer is the most fatal diseases with high mortality rates, such as this one, survival prediction assumes an important role, since it aids clinicians to better define each patient's prognosis and the corresponding treatments to be attempted. In particular for breast cancer, prognosis is related to the patterns of prediction. Cancer Prediction describes cancer that reappears after treatment, and in the specific case of breast cancer, prediction is very common, being experienced by about one third of patients after initial diagnosis. Therefore, establishing the patterns of prediction is a crucial task to accurately predict the clinical behavior of this pathology. This enables a more personalized treatment for the patients, avoiding undesired overtreatment and adverse complications. Gradient Boosting is a powerful machine learning algorithm founded on the idea that combining the labels of many 'weak' classifiers or learners translates to a strong robust one to predict the breast cancer. Boosting is a greedy algorithm that fits adaptive models by sequentially adding these base learners to weighted data where difficult to classify points are weighted more heavily. Experts claim that gradient boosting is the best off-the-shelf classifier developed so far to detect and predict the Breast Cancer. As we can see from the above versions of boosting, a unique boosting algorithm can be derived for each loss function and its performance can vary depending on which base learner. We can derive a generic version of boosting called gradient boosting for the identification, detection, recognition and prediction of breast cancer. Keywords: breast cancer, classification, machine learning, data mining, gradient boosting machine, prediction based system
Kablosuz sensör ağlarında optimal enerji tüketimi için basenode pozisyonu optimizasyonu
Breast Cancer is the most fatal diseases with high mortality rates, such as this one, survival prediction assumes an important role, since it aids clinicians to better define each patient's prognosis and the corresponding treatments to be attempted. In particular for breast cancer, prognosis is related to the patterns of prediction. Cancer Prediction describes cancer that reappears after treatment, and in the specific case of breast cancer, prediction is very common, being experienced by about one third of patients after initial diagnosis. Therefore, establishing the patterns of prediction is a crucial task to accurately predict the clinical behavior of this pathology. This enables a more personalized treatment for the patients, avoiding undesired overtreatment and adverse complications. Gradient Boosting is a powerful machine learning algorithm founded on the idea that combining the labels of many 'weak' classifiers or learners translates to a strong robust one to predict the breast cancer. Boosting is a greedy algorithm that fits adaptive models by sequentially adding these base learners to weighted data where difficult to classify points are weighted more heavily. Experts claim that gradient boosting is the best off-the-shelf classifier developed so far to detect and predict the Breast Cancer. As we can see from the above versions of boosting, a unique boosting algorithm can be derived for each loss function and its performance can vary depending on which base learner. We can derive a generic version of boosting called gradient boosting for the identification, detection, recognition and prediction of breast cancer. Keywords: breast cancer, classification, machine learning, data mining, gradient boosting machine, prediction based system