Search

In data mining, mining high utility itemset (HUI) is one among the recent thrust area that receives several approaches for solving it in an effective manner. In the past decade, addressing optimization problems using evolutionary algorithms are an unavoidable strategy due to its convergence towards optimal solution within the stipulated time. The results of evolutionary algorithms on various optimization problems are far effective when compared to the exhaustive approaches with respect to computational time. The problem with HUI is discovering a set of items from a transactional database that possess high level of utility when compared with other distinctive sets. This problem becomes harder while addressing the count of items in the database while its higher and computational time to solve this problem using exhaustive search becomes exponential as proposition of items in transaction database increases. In this paper, an optimization model based on the biological behaviour of grey wolf is proposed; the model namely grey wolf optimization algorithm is used to solve HUI using five different Boolean operations. The proposed model is evaluated using standard performance metrics over synthetic datasets and real-world datasets. The proposed model results are then compared with recent HUIM models to show the significance.

Face swapping has been a thriving genre of work which primarily is associated with the replacement or substitution of one referral face on the face of the other person, be it by means of still images or in real time. This paper exhibits the research on Real-Time Face Swapping algorithm, using the reference image of the user and henceforth considering it as our input image instead of using a data set thereby working on a real time model by the use of the python open-source model of Computer Vision i.e., OpenCV. The input image facial features and attributes are extracted and are replaced to make the final resulting output through our model. The training of the model starts with learning the facial alignments and features to be able to recognize a face and extract its features before swapping it. Now, heading over to the process of image warping by the dissociation and partitioning of the user's face and its background. To increase the precision of the model, in the second part of parsing the face is performed to cancel out other features of the live image, hence the output is obtained. In this research, it was tried to bring accuracy and precision to the model by improving the condition of the image, editing out borders, and training it to a bunch of data for face recognition and correction. On analyzing it delivers the most subtle outputs alongside performing the tasks with the process including the analysis of our image and its output too. The model delivered is to improve the conditions of security, privacy, image capture, and entertainment purposes.

Big Data Technologies are well-accepted in the recent years in bio-medical and genome informatics. They are capable to process gigantic and heterogeneous genome information with good precision and recall. With the quick advancements in computation and storage technologies, the cost of acquiring and processing the genomic data has decreased significantly. The upcoming sequencing platforms will produce vast amount of data, which will imperatively require high-performance systems for on-demand analysis with time-bound efficiency. Recent bio-informatics tools are capable of utilizing the novel features of Hadoop in a more flexible way. In particular, big data technologies such as MapReduce and Hive are able to provide high-speed computational environment for the analysis of petabyte scale datasets. This has attracted the focus of bio-scientists to use the big data applications to automate the entire genome analysis. The proposed framework is designed over MapReduce and Java on extended Hadoop platform to achieve the parallelism of Big Data Analysis. It will assist the bioinformatics community by providing a comprehensive solution for Descriptive, Comparative, Exploratory, Inferential, Predictive and Causal Analysis on Genome data. The proposed framework is user-friendly, fully-customizable, scalable and fit for comprehensive real-time genome analysis from data acquisition till predictive sequence analysis.

The growth of cancer in India is growing hastily in recent years. Efficient monitoring and medication procedures are needed in high demand. Recent research states diagnose of cancer during its early break through will prevent mortality. The evolution of smart mobile devices paves its mutual focus in healthcare sectors. In this paper, an Intellectual model of disease diagnosis using the advantage of smart mobile devices has been proposed. This mobile based cancer diagnosis model uses a cloud environment for disease prediction and analysis. The Principal Component Analysis (PCA) technique is utilized to confiscate the superfluous features and choose the most appropriate features. Using the optimized features, cancer disease classification is accomplished using Support Vector Machines with sigmoid kernel function. SVM classifies the patients as normal and abnormal and the evaluated results are conveyed to the patients as well as the respective medical practitioners. The accuracy achieved through proposed model is satisfiable in comparison with other existing methods. Proposed Model incorporates with big data technologies to address the current issues of cancer system.

In today’s Business world, big data projects play an important role in increasing the profit of the business. Success of the big data project lies on the complete understanding of the customer requirement. A big data project not only gives the solution for day to day functionality of the system but also gives a solution to improve the business. Hence understanding the requirement is very important. Most of the big data projects are based on the object oriented approach. Most of the software industry relies on UML use case model to specify, analyze and design the big data project. The difference between other project and big data project is data characteristics. The three V characters, volume, velocity and variety play major role in the design decision of big data project. The existing UML use case diagram does not have a facility to represent the big data V characteristics. In this paper we are proposing big data use case diagram, suitable for specifying big data characteristics. The research result shows that big data use case diagram improves the understandability for the requirements of the big data project.

Recently machine learning have been a field of interest for many researchers. Machine learning holds applications in tremendous sectors like healthcare, e-commerce and so on. It plays a vital role in the field of healthcare by their significant contribution through its ability to read from the historic/ past data especially in the diagnosis/predicting of diseases. Many classification algorithms have been proposed for the prediction process. The problem with classifiers is their vast variety and difficulty in choosing an appropriate method for the particular problem. To overcome the issue of selecting the appropriate method a comprehensive study on various classifiers should be done. This paper focuses on the working of various classifiers for prediction of prostate Cancer in calculating the level of efficiency in prediction and this helps in selecting the best method.

Honey bee is one of the charming insect that utilizes a collective behavioral nature to achieve the powerful action. Protecting honey bees is one of the important jobs of every human in the world to preserve the ecological balance. Tracking and determining the several species of the bees over their life span electronically is a tedious work. Automated classification of species is important to preserve the various species of honey bees from danger. The diseases that affect the honey bees during their life span have to be detected autonomously and the spread of the diseases to other healthy honey bees has to be preserved. The proposed technique aims in classifying the several species of honey bees and identifying the diseases that are prone to honey bees. Convolution neural network with two dimensional layers are used as a classifier in the proposed model. Data augmentation using Synthetic Minority Over-sampling Technique (SMOTE) is utilized. More than 5000 images of honey bees with lot of features are used for learning purpose. The proposed methodology attained an accuracy of 86% for subspecies classification and 84% for bee health identification.

In this chapter, we aim to scale-up the operation of GAN using recurrent neural network (RNN). The GAN is designed with two deep neural network models, where the first network is called a generator that generates perceptually synthetic samples drawn through real data distribution. The first network recursively learns the network and transforms the noise vectors into samples. The second network is called a discriminator that accesses both real and synthetic samples and provides classification between these two data. The discriminator uses the same RNN to classify the generated and real data samples. The aim of using RNN in the GAN structure is to delimit the error rate using its time series prediction based on its past inputs. It further uses its neighborhood relationship between the samples to generate the target output and error generated. The errors are then propagated in the backward direction over the GAN to update the network weights to estimate the output. The RNN generator, on the other hand, reduces the probability of the RNN discriminator in identifying the false generated samples and increasing the probability of discriminator in identifying correctly the real samples. The objective function in RNN is designed in such a way that its gradient operator for the false samples is quite far from the decision boundary of RNN discriminator, thereby producing the increasing true classification rate. The experiments carried out on the real-world time series dataset show the results of accurate classification (97.01%) with increased false detection rate than benchmark GAN method.

Initial cases of COVID-19 trace back to the end of 2019 which has laid foundations for the extensive spread of the disease risking lives worldwide. In response to the global coronavirus pandemic, early diagnosis of the disease is vital to prevent the virus from being spread to a larger population. Because of the unavailability of precise diagnostic toolkits, there arises a crying need to find efficient techniques that can be implemented for faster disease prediction while ensuring the accuracy of the prediction. Artificial intelligence (AI)–based solutions have the potential to help diagnose COVID-19 pandemic in an effective way. Automated image analysis with AI techniques can support clinical decision-making, improve workflow efficiency, and allow accurate and fast diagnosis of infection in a large number of patients. In the present study, existing convolutional neural network (CNN) models such as ResNeXt, Channel Boosted CNN, DenseNet, AlexNet, and VGG 16 were repurposed to assist in identifying the presence of COVID-19 before they reach mass scale. The dataset used in the study comprises of computed tomography (CT) images taken from 275 healthy individuals and 195 COVID-19 samples collected from 216 affected individuals. The proposed AI-based approach using deep learning models classifies COVID-19 affected cases by analyzing CT images and provides rapid detection of COVID-19 in a shorter time span. Ensemble classifier employed in this study proved to predict the presence of infection with a greater accuracy of 90.67% compared to the other similar works in the literature.

Lung cancer is one of the most prevalent cancer-related diseases with a high mortality rate, and this is largely due to the lateness in detecting the presence of malignancy. Again, the conventional methods used in the diagnosis of lung cancer have had their shortfalls. While the effectiveness of computerized tomography in detecting this malignancy, the large volumes of data that radiologists have to process not only present an arduous task but may also slow down the process of detecting lung cancer early enough for treatment to take its course. It is against this backdrop that computer-aided diagnostic (CAD) systems have been designed. One of such is the convolutional neural network, a method that best describes a group of deep learning models featuring filters that can be trained with local pooling operations being incorporated on input CT images in an alternating manner to create an array of hierarchical complex features. The need to have this type of data-driven technique is further informed by the attempt to ensure successful segmentation of lung nodules, a step that cannot be overruled when striving for a good model of detection or diagnosis. There are variations and models of the convolutional neural networks that have been effectively put to use in the lung nodule detection. The 2D CNN model has been utilized in the medical field for quite a while now, and as it has displayed its many strengths, so could the limitations not be hidden. It is in addressing these limitations and improving on the detection prowess of the convolutional neural network that the 3D model is now fast gaining traction. The 3D models have been reported to return pronounced sensitivity and specificity in detection of lung nodules, but the issues of time-consumption, training complexities and hardware memory usage could make it difficult to implement the 3D model in the medical field. In this paper, review the advances that have been made in the area of adopting 3D CNN model in the diagnosis of lung cancer.

Dataset plays a major role in evaluating Machine Learning (ML), Artificial Intelligence (AI), Business Intelligence (BI) and enterprise software. To prove the efficacy of any algorithm / software and also its limitation with respect to the size of the input data, it has to be exercised on a large dataset. Creating such a large dataset is a challenging task especially with near real time data. In this paper, a layered architecture to construct Virtual DataStack (VDS) is proposed. The VDS contains near real time synthetic virtual datasets (specific to a particular domain) along with other components that comprises the generation of datasets. Virtual DataStack for CampusStack (academic campus management) domain has been generated and results prove that Query Inclusive Computational Logic reduces lines of code as well as the time taken significantly. This paper focuses mainly on efficient generation of near real time data for enterprise domains with a specific attention to Education Domain and CampusStack.

Alzheimer’s disease is the most dreadful. Despite the various Alzheimer’s treatments available nowadays, the survival rate of Alzheimer’s patients is very much low. Detection of Alzheimer’s disease in an earlier stage is one of the most important things to reduce the death rate. Magnetic Resonance Imaging (MRI) is an important tool in medical informatics and clinical diagnosis. This MRI image helps to diagnose and detect Alzheimer’s disease at advanced stages. There are several approaches implemented to discover Alzheimer’s by MRI data. Features are extracted discriminately and combinations of different classification techniques for classification is implemented in the proposed work. In this paper, the curvelet-based transform technique is utilized for extracting the features. The AdaBoost classifier is utilized for combining multiple weak classifiers into one strong classifier to improve the accuracy of the result. To improve the efficiency of the result, AdaBoost classifier with SVM was used to obtain better results than existing ones. The proposed model with 95.66% accuracy predicts the AD from the brain MRI images at an earlier stage. This accuracy value is far better than other classifiers like Decision Tree and Random Forest.

As a result of the recent developments that swept through the petroleum industry, the Internet of things (IoT) became a part of the main applied technologies in the past few years. This chapter illustrates how the future depends on IoT in building intelligent oil fields and oil wells. Major benefits and challenges of using IoT as discussed. It also includes literature reviews and comparisons of different studies. The contents of the research are supported by using analysis and diagrams of design and architects of the system.

This chapter rolls on about the infrastructure capabilities of Internet of Things in managing the huge amount of big data and the future research context in developing a dynamic environment where IoT devices can connect and manage their resources on their own. New services are needed to progress the performance and service quality provided by the old services. Self-adaptation is essential for the IoT devices in a dynamic environment. These devices could publish/subscribe/notify/search/retrieve the data autonomously in a dynamic environment. In opportunistic networks, mobile collaboration could be utilized in case of direct communication failure and in cases of handling large data. These networks are mainly used during short-range communication when nodes are closer to each other. During communication, the routes are constructed dynamically according to the availability of the nodes learned based on machine learning algorithms. Data distribution is made on the basis of publishing/subscribing method. For continuous connectivity of devices in wired and wireless links, standard communication protocols are essential to manage the high traffic in the network as well. Added to it, new solutions are required for efficient storage, fetching, and searching of data in these complex environments.

Phishing is a critical issue that faces the digital security. The straightforwardness of the web and Internet uncovered open doors for offenders to transfer malevolent substance at the same time with the upgrade of online business trades, for example, phishing – the demonstration of taking individual data which ascends in number. Internet clients’ costs have been increased to billions of dollars for each year due to phishing. Phishers use parodied email, Uniform Resource Locator (URL) locations of phony sites, and phishing programming to take individual data and monetary record subtleties, for example, usernames and passwords. The boycott system is definitely not a sufficient method to remain safe from the cybercriminals. Hence, phishing site pointers must be considered for this reason, with the presence and utilization of machine learning calculations. The current techniques make utilization of all separated attributes in the phishing URL location, prompting high false positive rate.

Smart farming based on recent Internet of Things (IoT) technologies is the most advanced method to grow food cleanly while making it sustainable. This method applies modern information and communication technologies to agriculture, targeted not only to reduce waste but also to increase agricultural productivity to an optimum level. Agriculture is considered the backbone of the world’s economy. The end-product’s quantity and quality both play a vital role. Using the proposed IoT technology, the identification of diseases in plants with focus on the affected area is more accurate. It also reduces the number of incorrect conclusions that may be drawn, which can lead to incorrect actions taken for sustaining the plants in farms. The proposed technique will also be able to predict the damage level by pests on plants to take appropriate actions to improve productivity. Digital images taken from the plants will be processed and examined using pattern recognition and digital image processing techniques. Proposed image analysis techniques will segment these images to identify diseases and the affected level. This will help to prevent plant disease and increase the nutrient level of the plant, with the help of IoT. The automatic detection of plant disease will be beneficial in monitoring a large field of crops.