Global Journal of Computer Science and Technology (GJCST)
Not a member yet
1830 research outputs found
Sort by
Recognition and Classification of Fast Food Images
Get PDFImage processing is widely used for food recognition. A lot of different algorithms regarding food identification and classification have been proposed in recent research works. In this paper, we have use an easy and one of the most powerful machine learning technique from the field of deep learning to recognize and classify different categories of fast food images. We have used a pre trained Convolutional Neural Network (CNN) as a feature extractor to train an image category classifier. CNN2019;s can learn rich feature representations which often perform much better than other handcrafted features such as histogram of oriented gradients (HOG), Local binary patterns (LBP), or speeded up robust features (SURF). A multiclass linear Support Vector Machine (SVM) classifier trained with extracted CNN features is used to classify fast food images to ten different classes. After working on two different benchmark databases, we got the success rate of 99.5% which is higher than the accuracy achieved using bag of features (BoF) and SURF
Accuracy Analysis of Continuance by using Classification and Regression Algorithms in Python
Get PDF- Reinforcement rate of technics and appositeness towards the convenience of the human being is a perennial mechanism. Mathematics has always been in the root towards the implementation of any algorithm or analysis regarding statistics or language. Extracting more about the data and analyzing them to solve a particular problem is the reason behind any analysis. Scrutiny itself has the different number of outcome which can be predictive or descriptive. Now prediction is how far accurate is tested by using various techniques. The enhancement in problem-solving capability leads to come up with a new aptitude concerning machine learning algorithms. But before prediction of data set collection, exploration, feature extraction, model building, accuracy testing are primarily required to invent. So for explaining all these processes, concept learning is essential. In this paper different algorithms like SVM, Linear and Logistic Regression, Decision tree, and Random forest algorithms will be used to demonstrate the accuracy in titanic data from Kaggle Website with all the required steps by using Python language
The Method of Normalizing OWL 2 DL Ontologies
Get PDFThe paper proposes a method of normalizing OWL 2 DL ontologies. The method introduces rules aimed at refactoring OWL 2 constructs. The proposed transformations only use a subset of OWL 2 constructs and enable to present an input OWL 2 ontology in a new but semantically equivalent form. The normalization is motivated by the fact that normalized OWL 2 DL ontologies have a unified structure of axioms so that they can be compared in an algorithmic way
Novel System and Method For Telephone Network Planing Based on Neutrosophic Graph
Get PDFTelephony is gaining momentum in the daily lives of individuals and in the activities of all companies. With the great trend towards telephony networks, whether analogue or digital known as Voice over IP (VoIP), the number of calls an individual can receive becomes considerably high. However, effective management of incoming calls to subscribers becomes a necessity. Recently, much attention has been paid towards applications of single-valued neutrosophic graphs in various research fields. One of the suitable reason is it provides a generalized representation of fuzzy graphs (FGs) for dealing with human nature more effectively when compared to existing models i.e. intuitionistic fuzzy graphs (IFGs), inter-valued fuzzy graphs (IVFGs) and bipolar-valued fuzzy graphs (BPVFGs) etc. In this paper we focused on precise analysis of useful information extracted by calls received, not received due to some reasons using the properties of SVNGs. Hence the proposed method introduced one of the first kind of mathematical model for precise analysis of instantaneous traffic beyond the Erlang unit. To achieve this goal an algorithm is proposed for a neutrosophic mobile network model (NMNM) based on a hypothetical data set. In addition, the drawback and further improvement of proposed method with a mathematical proposition is established for it precise applications
From Service-Oriented Architecture To Cloud Computing
Get PDFCloud computing resembles a new paradigm of technology It suggests deploring technology services without owning the infrastructure behind them It also releases the burden of maintaining an adequate environment and quality and focusing on the business competency Service-Oriented Architecture SOA is a technology outlook that enables approaching cloud computing In this paper we reviewed the main feature of SOA The main migration from SOA to cloud computing is discussed Main features and characteristics of cloud computing are presente
Survey on Precision Farming using Mobile Applications
No full textThe need to provide a farmer with the right knowledge of the crops that would grow and increase the yield in respective farms is the main aim of the project The information technologies used to accomplish the farmers problem are Mobile Application Development Cloud Computing Internet of Things IoT and Databases The farmer will be able to choose the best and the right crops for their farm under different weather conditions seasonal conditions and also farmer will be able know the marketing condition such as Demand and Supply condition of the market before sowing the seeds in a farm By meeting the Demand and Supply of the Market a farmer will be able to increase the income which is achieved by farming on a farm The farmer will be able to access the condition of a farm through historic records like previous 10 years conditions of the farm The farmer will know whether the Crops are ready to be harvested or not The farmer will also be able to know the details of all crops All crops details will be provided to the farmer Farmers will also be able to monitor the farm remotely using the Internet of Things IoT technolog
Arabic Question Answering with Dialogue Support
No full textQuestion Answering QA system is a combination of Information Retrieval IR and Natural Language Processing NLP techniques It returns a specific answer in response to user question However a system that can interact with the user to clarify and refine the answer is required We propose QA system that adopts a user model for adaptation and a dialogue interface for interaction with the user combined with information retrieval and natural language techniques for Arabic Language Our system will be able to handle users questions in natural language and to present answers in in respect to the user s preferences and expected needs The system achieved a precision of 82 05 and a dialogue success rate of 71 6 The result is highly promising As an extension for the present work we need to make the system more adaptive and capable to learn and evolve with every new interactive scenari
Prediction and Judgmental Adjustments of Supply-Chain Planning in Festive Season
Get PDFFor a robust performance Shipping costs planning in festive seasons is given the input data as free from trends season-of-year effects etc Seasonal forecasting for supplychain planning with past few years of similar data impact shipping costs Additionally during a festive season of the year unbiased and accurate prediction of shipment load plays a major role in bringing up sales Time-series forecasting methods can be useful to remove traditional fluctuations due to gap in months-of-year of festivals We describe exponential smoothing techniques and trend fitting methods and compare the predictive accuracy The accuracy is compared using rootmean square error and median absolute deviation The exponential smoothing shows changing behavior with increased data size and data item values The data is compared with and without tuning the seasonal effects due to festive seaso
Intrusion Detection System based on Ant
Get PDFChallenge of designing and building of the current Network Intrusion Detection System not only improves the ability for discriminating the improper internet behaviors but also considers the plenty of computer resources which will be cost during the analysis of the network packets and behaviors During the establishment of a fast intrusion detection system the major purpose of the research is to intensify the data handling capacity when the network management system faces the mass network behavior In the research the modules stored in the network packets of the intrusion detection system are analyzed and then the network flow data by applying the clustering algorithm based on the ant colony system is classified Finally a kind of algorithm which can remove repetitive computation is designed so that the flow can accelerated and the velocity can be distinguished Experimental results show that the proposed fast clustering algorithm can significantly reduce the original computation time while sacrificing or promoting a very small accuracy and then the computation speed of the intrusion detection system can be accelerate
Classification of HRS using SVM
Get PDFThe kidney diseases are one of the main causes of death around the world. Automatic detection and classification of kidney related diseases are important for diagnosis of kidney irregularities. Hepatorenal Syndrome (HRS) is a lifethreatening medical condition when kidney fails due to liver failure. The treatment to such cases is liver transplant, or dialysis for temporary basis. This paper proposed to apply the Support Vector Machine (SVM) classification for diagnosis of HRS. The results were evaluated using realistic data from hospitals. RBF kernel function is used along with SVM. The results show a significant accuracy of 95%