Fig 5 - uploaded by Mahendra Pratap Yadav
Content may be subject to copyright.
Similar publications
This paper discusses our vision of multirole-capable decision-making systems across a broad range of Data Science (DS) workflows working on graphs through disaggregated data centres. Our vision is that an alternative is possible to work on a disaggregated solution for the provision of computational services under the notion of a disaggregated data...
Zusammenfassung
Im Rahmen des vom BMBF geförderten Projekts Fahrerkabine 4.0 wird eine adaptive Mensch-Maschine-Schnittstelle für Landmaschinen entwickelt, die das aktuelle Beanspruchungslevel mit Hilfe physiologischer Daten detektiert. Zu diesem Zwecke wird in dieser Arbeit eine Experimentalaufgabe entwickelt und evaluiert, die ein psychisches Bel...
In the recent past, characterizing workloads has been attempted to gain a foothold in the emerging serverless cloud market, especially in the large production cloud clusters of Google, AWS, and so forth. While analyzing and characterizing real workloads from a large production cloud cluster benefits cloud providers, researchers, and daily users, an...
Citations
... Gian Hug et al., classified conception clustering method is proposed with highest accuracy index of 90% with respect to neural network optimization [6]. Clever Land Dataset can be used for testing the results [7]. Principle Component analysis method has good classification factor and cross validation approach is applied for predicting accuracy. ...
The clinical dataset consists of a huge amount of disease information and which marks negative effects with respect to features. Feature selection is a major key player removing redundant information. It also increases the decision making result in an effective manner. The important step in feature selections is classification and dimensionality reduction. Medical experts and researchers are using various machine learning techniques to predict and analyze the huge volume of data. The large scale volume of medical dataset is having number attributes and dimensions. Dimensions can affect the analytics and diagnose the results. In this paper we evaluate principle component analysis, support vector machine, factor analysis and ranking methods compared with our proposed CNN-based ensemble feature selection using K-nearest classifier. Our proposed method is followed as standard testing features with parameter optimization. Compared with the existing method our method has 94% of accuracy result by using TensorFlow.KeywordsDeep learningTensorFlowPredictionClassifierMedical dataset
... The prediction will be established based on the temporal locality principle and the dynamic assignment of weights to different data points in recent history. Yadav and Yadav [31], presents a predictive analysis of time series forecasting using deep learning method (LSTM) to predict the future load over servers. Based on that analysis it allocates the number of containers for running an application to manage the fluctuating workload. ...
Resource management is addressed using infrastructure as a service. On demand, the resource management module effectively manages available resources. Resource management in cloud resource provisioning is aided by the prediction of central processing unit (CPU) and memory utilization. Using a hybrid ARIMA–ANN model, this study forecasts future CPU and memory utilization. The range of values discovered is utilized to make predictions, which is useful for resource management. In the cloud traces, the ARIMA model detects linear components in the CPU and memory utilization patterns. For recognizing and magnifying nonlinear components in the traces, the artificial neural network (ANN) leverages the residuals derived from the ARIMA model. The resource utilization patterns are predicted using a combination of linear and nonlinear components. From the predicted and previous history values, the Savitzky–Golay filter finds a range of forecast values. Point value forecasting may not be the best method for predicting multi-step resource utilization in a cloud setting. The forecasting error can be decreased by introducing a range of values, and we employ as reported by Engelbrecht HA and van Greunen M (in: Network and Service Management (CNSM), 2015 11th International Conference, 2015) OER (over estimation rate) and UER (under estimation rate) to cope with the error produced by over or under estimation of CPU and memory utilization. The prediction accuracy is tested using statistical-based analysis using Google's 29-day trail and BitBrain (BB).
As economies continue to move forward credit card providers face novel emerging problems. A pressing economic issue that has become prominent in recent years is the phenomenon of credit card churn. A process in which malicious users sign up for new credit cards simply to profit from new user benefits, only to close the card once the perks expire. The application of novel artificial intelligence (AI) techniques is a promising approach for accurately handling the immense amounts of complex information present in the financial sector. Furthermore, the utilization of AI for this problem has not yet been sufficiently explored. This works attempts to address the prominent research gap by applying a support vector machine (SVM) to forecasting credit card churn. To attain the best possible performance a recently introduced reptile search algorithm (RSA) is tasked with selecting the best possible hyperparameters of an SVM. Furthermore, an improved version of the RSA is proposed in hopes of enhancing performance. The proposed optimized approach is applied to real-world data concerning credit card churn, and the results indicated the proposed approach outperformed all other tested algorithms applied to the same task, achieving an accuracy score of 99.56%, weighted average precision of 0.91, weighted average recall of 0.92, and weighted average f-1 score of 0.91. These empirical findings suggest the high predictive potential of the proposed method.
