UN Department of Economics and Social Affairs predicted that the world population will increase by 2 billion in 2050 with over 50% from the Sub-Saharan Africa (SSA). Considering the level of poverty and food insecurity in the region, there is an urgent need for a sustainable increase in agricultural produce. However, farming approach in the region is primarily traditional. Traditional farming is characterized by high labor costs, low production, and under/oversupply of farm inputs. All these factors make farming unappealing to many. The use of digital technologies such as broadband, Internet of Things (IoT), Cloud computing, and Big Data Analytics promise improved returns on agricultural investments and could make farming appealing even to the youth. However, initial cost of smart farming could be high. Therefore, development of a dedicated IoT cloud-based platform is imperative. Then farmers could subscribe and have their farms managed on the platform. It should be noted that majority of farmers in SSA are smallholders who are poor, uneducated, and live in rural areas but produce about 80% of the food. They majorly use 2G phones, which are not internet enabled. These peculiarities must be factored into the design of any functional IoT platform that would serve this group. This paper presents the development of such a platform, which was tested with smart irrigation of maize crops in a testbed. Besides the convenience provided by the smart system, it recorded irrigation water saving of over 36% compared to the control method which demonstrates how irrigation is done traditionally.

This paper refers to work evaluating the performance of PI controllers integrated with optimization techniques designed for Single Ended Primary Inductance Converters (SEPIC). With the SEPIC converter, a constant voltage output can be retained while switching a range of dc voltages. Performance of PI controller has been combined with Artificial Bee Colony (ABC) Algorithm, Particle swarm optimization (PSO) Algorithm, Whale optimization algorithm (WOA). In this research, a performance analysis of the SEPIC dc-dc converter controller constructed with the aforementioned optimization strategies is carried out. Statistics proves WOA provides best stability exhibited with fast response when compared to other optimization techniques.

In recent years, Machine learning is being used in various systems in wide variety of applications like Healthcare, Image processing, Computer Vision, Classifications, etc. Machine learning algorithms have shown that it can solve complex problem-solving capabilities close to humans or beyond humans as well. But recent studies show that Machine Learning Algorithms and models are vulnerable to various attacks which compromise security the systems. These attacks are hard to detect because they can hide in data at various stages of machine learning pipeline without being detected. This survey aims to analyse various security attacks on machine learning and categorize them depending on position of attacks in machine learning pipeline. This paper will focus on all aspects of machine learning security at various stages from training phase to testing phase instead of focusing on one type of security attack. Machine Learning pipeline, Attacker’s goals, Attacker’s knowledge, attacks on specified applications are considered in this paper. This paper also presented future scope of research of security attacks in machine learning. In this Survey paper, we concluded that Machine Learning Pipeline itself is vulnerable to different attacks so there is need to build a secure and robust Machine Learning Pipeline. Our survey has categorized these security attacks in details with respect to ML Pipeline stages.

One area that has seen rapid growth and differing perspectives from many developers in recent years is document management. This idea has advanced beyond some of the steps where developers have made it simple for anyone to access papers in a matter of seconds. It is impossible to overstate the importance of document management systems as a necessity in the workplace environment of an organization. Interviews, scenario creation using participants' and stakeholders' first-hand accounts, and examination of current procedures and structures were all used to collect data. The development approach followed a software development methodology called Object-Oriented Hypermedia Design Methodology. With the help of Unified Modeling Language (UML) tools, a web-based electronic document management system (WBEDMS) was created. Its database was created using MySQL, and the system was constructed using web technologies including XAMPP, HTML, and PHP Programming language. The results of the system evaluation showed a successful outcome. After using the system that was created, respondents' satisfaction with it was 96.60%. This shows that the document system was regarded as adequate and excellent enough to achieve or meet the specified requirement when users (secretaries and departmental personnel) used it. Result showed that the system developed yielded an accuracy of 95% and usability of 99.20%. The report came to the conclusion that a suggested electronic document management system would improve user happiness, boost productivity, and guarantee time and data efficiency. It follows that well-known document management systems undoubtedly assist in holding and managing a substantial portion of the knowledge assets, which include documents and other associated items, of Organizations.

An epileptic seizure is a type of seizure induced by aberrant brain activity caused by an epileptic condition, which is a brain Central Nervous System disorder (CNS). CNSs are relatively prevalent and include a wide range of symptoms, including loss of awareness, and strange behaviour. These symptoms frequently result in injuries as a result of walking imbalance, tongue biting, and hearing loss. For many researchers, detecting a prospective seizure in advance has been a difficult undertaking. In this research work we have used non-imaging data and applied supervised learning algorithms to determine the classification of epilepsy and try to improve the efficiency of the model, compared to the existing ones. Random Forest algorithm was found to have highest accuracy compared to other machine learning models. The paper can be helpful in diagnosing high-risk brain diseases and predicting diseases such as Alzheimer's with symptoms challenging to predict and diseases with overlapping symptoms and overlapping symptoms and attribute values. The scope of the research work can be further extended to determine at which stage the epilepsy is present in a patient, in order to provide a correct diagnosis and medical treatment.