Search

In recent times, the healthcare industry has been generating a significant amount of data in distinct formats, such as electronic health records (EHR), clinical trials, genetic data, payments, scientific articles, wearables, and care management databases. Data science is useful for analysis (pattern recognition, hypothesis testing, risk valuation) and prediction. The major, primary usage of data science in the healthcare domain is in medical imaging. At the same time, lung cancer diagnosis has become a hot research topic, as automated disease detection poses numerous benefits. Although numerous approaches have existed in the literature for lung cancer diagnosis, the design of a novel model to automatically identify lung cancer is a challenging task. In this view, this paper designs an automated machine learning (ML) with data science-enabled lung cancer diagnosis and classification (MLDS-LCDC) using computed tomography (CT) images. The presented model initially employs Gaussian filtering (GF)-based pre-processing technique on the CT images collected from the lung cancer database. Besides, they are fed into the normalized cuts (Ncuts) technique where the nodule in the pre-processed image can be determined. Moreover, the oriented FAST and rotated BRIEF (ORB) technique is applied as a feature extractor. At last, sunflower optimization-based wavelet neural network (SFO-WNN) model is employed for the classification of lung cancer. In order to examine the diagnostic outcome of the MLDS-LCDC model, a set of experiments were carried out and the results are investigated in terms of different aspects. The resultant values demonstrated the effectiveness of the MLDS-LCDC model over the other state-of-the-art methods with the maximum sensitivity of 97.01%, specificity of 98.64%, and accuracy of 98.11%.

Increase in the development of Intranet usage take part a key role in monitoring and controlling Internet traffic within an organization. For getting good cyber security solutions, it is necessary to have proper network monitoring in any organizations. Network can be monitored with variety of tools available in the market. This paper discusses about network monitoring approaches in CPU and GPU along with that compares some of the network monitoring tools available in the market.

Rainfall prediction is the one of the important technique to predict the climatic conditions in any country. This paper proposes a rainfall prediction model using Multiple Linear Regression (MLR) for Indian dataset. The input data is having multiple meteorological parameters and to predict the rainfall in more precise. The Mean Square Error (MSE), accuracy, correlation are the parameters used to validate the proposed model. From the results, the proposed machine learning model provides better results than the other algorithms in the literature.

Since industrial revolution, the rate of industrialization and urbanization has increased dramatically. Most of the industry applications create pollution in the air and the vehicle emissions are also dangerous to the health of the people. In the developing countries, air pollution is severe in most of the areas. Air quality is the important factor to measure the quality of air. Most of the air quality measuring systems uses air quality index to tell the people about the air quality of their location. The primary objective of the system is to analyze and visualize air quality from the real time sensor data. The proposed system analyses six critical air pollutants which are, ozone (O 3 ), Particulate Matter (PM 2.5 ), Carbon monoxide (CO), Nitrogen dioxide (NO 2 ) and Sulphur dioxide (SO 2 ) are the most widespread health threats. The Fuzzy c-Means clustering is used to process the polluted air data from the sensors. From the results it is clear that the Fuzzy c-Means algorithm provides better results for the parameter accuracy while evaluating with the other algorithms in the literature.

Testing in a cloud environment involves a range of tests to analyze various aspects of the system in various cloud adoption scenarios such as public, private, or hybrid. Performance testing is a key aspect of the cloud testing strategy and it evaluates the application/software for various issues such as speed, stability, and scalability under varying load conditions. The key issues that must be considered in a cloud set-up as a part of the performance testing strategy are collection of statistics on the load, conducting stress test, monitoring the memory, checking for elasticity and scalability. This paper focuses on testing the resource utilization in cloud infrastructure in general and memory utilization in particular. The memory utilization is evaluated in public and private cloud environment with simulation of 10 users to 10,000 users.

Data analytics is used to obtain the useful insights from small or large data set to conclude some useful information and also used for future recommendation and decision making. Predictive Analytics uses the data mining, machine learning techniques to make predictions about future. It involves analysis of available data. The predictive analytics in health care is primarily used to determine patients having initial stages of diabetes, asthma, heart disease and other critical lifetime disease. The proposed method PDD uses data mining algorithms to predict the type2 diabetes. The data mining algorithms used in the proposed system are K-Means Clustering and Random Forest. The predictive model, PDD provides better results in terms of accuracy when compared to hierarchical clustering and Bayesian network clustering with random forest prediction.

Education is the most powerful weapon by which the world could be changed said Nelson Mandela. The history of spreading knowledge has its root from teaching Vedas in temples until eLearning and eTeaching. Both the traditional teaching methodology and online teaching provide students with knowledge. National Center for Education Statistics says that an online class is stated as “a formal education process in which the students and instructor are separated”. Alike traditional courses, online classes are educated by respective course instructors which includes several classmates. However, the communication medium between the both would be email, telephone or live chat. By considering the advantage of online courses we propose an online performance evaluation tool in which the student can be evaluated once when the course is completed and tests are done. From the test results the performance of the students is evaluated using fuzzy based technique and reports of the individuals are generated.

Software testing is the process of software engineering to free the software from bugs. Load testing is one of the techniques in software testing and is used to find the maximum load that software can handle without affecting its performance. Load testing is used to test the cloud services that are running in a cloud. All the resources in a cloud are used by the cloud users based on their demand. Using cloud, it is easy to gather the required load for a particular application by forming clusters. If the required load is coming from different clusters and it is not known quantitatively then the problem of load balancing is raised. The proposed load testing tool avoids the problem of getting unequal loads coming from different clusters by distributing the same amount of load to all the clusters. Also the proposed load testing tool for cloud is used to find the maximum number of simultaneous users for a particular cloud system is to handle.

Grid computing uses computers that are distributed across various geographical locations in order to provide enormous computing power and massive storage. Scientific applications produce large quantity of sharable data which requires efficient handling and management. Replica selection is one of the data management techniques in grid computing and is used for selecting data from large volumes of distributed data. Replica selection is an interesting data access problem in data grid. Genetic Algorithms (GA) and Simulated Annealing (SA) are two popularly used evolutionary algorithms which are different in nature. In this paper, a hybrid approach which combines Genetic Algorithm with Simulated Annealing, namely, HGASA, is proposed to solve replica selection problem in data grid. The proposed algorithm, HGASA, considers security, availability of file, load balance and response time to improve the performance of the grid. GridSim simulator is used for evaluating the performance of the proposed algorithm. The results show that the proposed algorithm, HGASA, outperforms Genetic Algorithms (GA) by 9 % and Simulated Annealing (SA) by 21 % and Ant Colony Optimization (ACO) by 50 %.

In the last few years there has been tremendous increase in the storage and processing of data, with significant speed and storage space requirements. Medical image storage and retrieval is one such application which requires huge storage space and not easily solved by computers. The researchers who work in the area of medical image processing deal with a high degree of storage and retrieval of medical images. Digitizing medical images and computer based image retrieval replaces the use of hard copies and provides efficient remote access. The proposed grid architecture implemented in this work uses Apache Hadoop framework. Hadoop is an open source grid architecture, complies with various image formats and can be established among various hospitals to store, share and retrieve images. Various performance metrics such as reliability, accuracy, interoperability, confidentiality and security are improved as with the use of Hadoop. The user authentication and patient confidentiality is also achieved. In this work, texture based Content Based Image Retrieval (CBIR) algorithm is used for efficient image retrieval. The performance of the proposed image retrieval system using Hadoop is tested with three current operative nodes. Experimental results obtained show that the retrieval time of the proposed system, Hadoop-CBIR, is considerably less with large number of images in the database.

Data grid is a hardware and software infrastructure built on top of Internet and provides sharing and management of distributed data resources. Dynamic data replication techniques are used in data grids to enhance data sharing with reduced access latency and bandwidth consumption. Creating and placing duplicate copies of popular files in data grid is replica placement, whereas, replica selection is selecting the best location to access data for executing job in data grid. Efficient techniques with load balancing and prediction based on SVM and AHL that improve and optimize the replica placement and selection are introduced in this paper. The proposed algorithms reduce the makespan i.e. total job execution time, storage consumption and effective network usage. The proposed techniques consist of two phases. In the first phase, all the nodes in the same region are grouped together and replica is placed in the highest degree and highest frequency node in the region. During second phase, SVM & AHL are used for predicting the behavior of user applications. The jobs are executed in the grid with predicted events. Simulation results reveal that the proposed methodologies perform better in terms of makespan, storage consumption and replication count.