The overall effectiveness of a website as an e-commerce platform is influenced by how usable it is. This study aimed to find out if advanced web metrics, derived from Google Analytics software, could be used to evaluate the overall usability of e-commerce sites and identify potential usability issues. It is simple to gather web indicators, but processing and interpretation take time. This data is produced through several digital channels, including mobile. Big data has proven to be very helpful in a variety of online platforms, including social networking and e-commerce websites, etc. The sheer amount of data that needs to be processed and assessed to be useful is one of the main issues with e-commerce today as a result of the digital revolution. Additionally, on social media a crucial growth strategy for e-commerce is the usage of BDA capabilities as a guideline to boost sales and draw clients for suppliers. In this paper, we have used the KMP algorithm-based multivariate pruning method for web-based web index searching and different web analytics algorithm with machine learning classifiers to achieve patterns from transactional data gathered from e-commerce websites. Moreover, through the use of log-based transactional data, the research presented in this paper suggests a new machine learning-based evaluation method for evaluating the usability of e-commerce websites. To identify the underlying relationship between the overall usability of the eLearning system and its predictor factors, three machine learning techniques and multiple linear regressions are used to create prediction models. This strategy will lead the e-commerce industry to an economically profitable stage. This capability can assist a vendor in keeping track of customers and items they have viewed, as well as categorizing how customers use their e-commerce emporium so the vendor can cater to their specific needs. It has been proposed that machine learning models, by offering trustworthy prognoses, can aid in excellent usability. Such models might be incorporated into an online prognostic calculator or tool to help with treatment selection and possibly increase visibility. However, none of these models have been recommended for use in reusability because of concerns about the deployment of machine learning in e-commerce and technical issues. One problem with machine learning science that needs to be solved is explainability. For instance, let us say B is 10 and all the people in our population are even. The hash function’s behavior is not random since only buckets 0, 2, 4, 6, and 8 can be the value of h(x). However, if B = 11, we would find that 1/11th of the even integers is transmitted to each of the 11 buckets. The hash function would work well in this situation.

Due in large part to the Internet of Things’ (IoT) anticipated enormous scope and extensive implementation, achieving safe and private communications on the IoT is difficult. Recent initiatives have investigated the use of blockchain technology to enable decentralized protection and privacy. Such methods, however, are prohibitive for the bulk of IoT applications due to their high computational and time requirements. We specifically offer a resource-efficient, blockchain-based IoT security and privacy solution in this study. The approach is made achievable by utilizing Deep Extreme Learning Machine in combination with unique computational resource exploitation in a typical IoT context (such as smart houses) (DELM). In the proposed method, the privacy, integrity, and accessibility of the Blockchain based Architecture of Smart Homes are prudently considered while assessing the reliability of the system. The overheads caused this strategy are negligible with respect to the security and privacy benefits, we further underline by presenting simulated findings.

Programming codes are of variable length. When the size of codes becomes greater than that of primary memory, the concept of virtual memory comes into play. As the name suggests, virtual memory allows to outstretch the use of primary memory by using storage devices such as disks. The implementation of virtual memory can be done by using the paging approach. Allocation of memory frames to each and every program is done by the operating system while loading them into the memory. Each program is segregated into pages as per the size of frames. Equal size of pages and frames enhance the usability of memory. As, the process or program which is being executed is provided with a certain amount of memory frames; therefore, swap out technique is necessary for the execution of each and every page. This swap out technique is termed as Page Replacement. There are many algorithms proposed to decide which page needs to be replaced from the frames when new pages come. In this paper, we have proposed a new page replacement technique. This new technique is based on the approach of reading and counting of the pages from secondary storage. Whenever the page fault is detected, the needed page is fetched from the secondary storage. This process of accessing the disc is slow as compared to the process in which the required page is retrieved from the primary storage. In the proposed technique, the pages having least occurrence will be replaced by the new page and the pages having same count will be replaced on the basis of LRU page replacement algorithm. In this method, the paged are retrieved from the secondary storage hence, possibility of page hit will be increased and as a result, the execution time of the processes will be decreased as the possibility of page miss will be decreased.

The wireless sensor networks have become a very hot topic of research in the recent years because of their wide range of applications such as industrial and agricultural monitoring, military surveillance, smart homes etc. These sensors can also be used in locations that are potentially hazardous to the human beings or are just out of our reach at this moment of time. A wireless sensor network is collection of large number of individual sensor nodes. A sensor usually comprises of components for sensing the environment, processing, communicating and a power supply. The major limiting factor of the wireless sensor networks is the limited amount of power that each of the sensors can carry and also the energy in these sensors cannot be replenished easily. So, the major design challenge in a wireless sensor network has always been about reduction in the consumption of energy by the sensors. Clustering is one of the popular methods to reduce energy consumption in wireless sensor networks. Here, we propose a scheme to decrease the energy consumption and prolong the lifetime of wireless sensor network. The main idea behind the scheme is that we try to minimise the communication distance between the individual sensor nodes using the clustering technique.