Alanoud A. Alqahtani

Online meal delivery services have grown in popularity as they allow individuals to order food from the comfort of their own homes. COVID has limited the number of offline businesses, closing many physical stores. The online food delivery market is flooded with plenty of options, making it difficult for consumers to decide which one to use. B2B platforms have become increasingly popular, solving the lockout problem and presenting other challenges. The Takeaway Menu system is a web app that allows family-run businesses to advertise their services on the platform and expose them to a large audience. Similarly, consumers have a rich platform from which they can purchase without worrying about the kitchen's quality and hygiene because the system ensures that only licensed suppliers are registered. The project features a micro-service design, with the back-end written in Java Spring-boot and Node.js and the client-side written in React.js. Asynchronous messaging is handled by RabbitMQ, while Zipkin handles service tracking. This paper begins by explaining the project's necessity and motivation. Then it goes over the needs and specifications, which are prioritised. Next is the planning section, where the technology stack chosen for the project is briefly detailed, along with specifics of the process model used, database design, and software modeling. After that, the application's development and testing are discussed. Finally, the paper concludes with information regarding the assessment procedure, performance metrics, and recommendations based on the evaluation results.

Michelo Mtolo, Grayson Himunzowa

This study investigates and analyses the prevailing energy consumption challenge and practices at the University of Zambia. For this reason, Data Science was proposed in this study. An orange toolkit which is a data Visualisation, Machine learning and data mining toolkit from Anaconda open-source distribution software and Microsoft power BI were used to achieve the objective and answer the research questions of this study. The process involved data collection of energy consumption from a specified target area which was the University of Zambia through the office of the Residence Engineer, raw energy consumption datasets were collected for 10 years from 2009 to 2019 in form of Microsoft Excel and stored using Microsoft OneDrive software, the data was thoroughly cleaned, understood and analysed, then based on that, a model was developed in Orange toolkit using Linear regression as a machine learning tool to give us the predictions, with the main data split into test and training data, the model was able to analyse the data and do the prediction of energy consumption for the year 2020, later data visualisation was achieved through Microsoft power BI software, hence completing the Data Science cycle and meeting the research objectives. Results show total monthly energy consumption for 2020 based on the historical data and other changing factors, with an accuracy of 92.4% test on training data and 77.4% test on test data against the actual 2020 data. The research has identified expected energy consumption through predictive analysis at the University of Zambia, proper energy management is a matter of concern at the University of Zambia.

Damith Wickramanayake, Danae Reid, Jovaughn Bennett, Matthew Tingling, Roshaun Walker

The study’s purpose was to use quantitative methodology to investigate and study the willingness of Jamaican citizens to accept cryptocurrency as a major alternate currency, using a quantitative methodology. The Unified Theory of Acceptance and Use of Technology (UTAUT) theoretical framework was used to identify the factors that would prevent the adoption of cryptocurrency as an alternative currency. Key findings revealed that 58.8% of respondents are willing to use cryptocurrency as an alternative currency and 40.1% of respondents were educated about cryptocurrency. This study also revealed a significant correlation between performance, effort, social influence, and the behavioural intention of Jamaicans to use cryptocurrency as an alternate currency. Also, there is a significant correlation between behavioural intention, facilitating conditions and the use behaviour of Jamaicans to use cryptocurrencies. Further there is a significant correlation between sociodemographic factors, and the level of interest in cryptocurrency among Jamaican citizens. The researchers concentrated on user opinions rather than the economics of cryptocurrency and also highlighted their technological and technical characteristics. We found that not all participants are very well educated about the technology and that is one of the biggest factors affecting adoption so that is the stage which the researchers think is the next step into full adoption - education and awareness.

Tingfang Wang, Sikha Bagui, Subhash Bagui

In conventional binary classification algorithms, all features are treated as having equal weight and classification models are built without taking into consideration the fact that different attributes can have different levels of influence on a class. Attribute weighting adjustments are used in machine learning models to improve performance. In this paper, we propose a novel attribute weighting method based on mutual information and apply this method to four classical machine learning models for classification. We study the performance of our weighting method by conducting experiments on the Wisconsin Breast Cancer database and Blood transfusion service center dataset. In three of the four machine learning models, our weighted attribute model outperformed the corresponding conventional machine learning models in binary classification.

Vasanth K. Makam

Continuous Integration (CI) aids in ensuring the software development process is reliable and provides fast feedback on tests and builds. It helps to detect errors quickly and corrects them before the system gets complicated. The CI process comprises of tools that help in identifying the correctness and provision of feedback for the software under process. The CI tools cleanse the data of errors through constant build and tests. The CI tools assist in transforming and mapping of the data before Integration and quick detection of errors. As cloud computing grows and gets embraced by businesses, the choice to host CI tools on premise or on cloud becomes a critical management decision. To make this decision, businesses ought to get acquainted with various CI tools and models they wish to implement, the security of their data, the capital to invest, and the reliability of the providers. Analyzing these variables will give business management an understanding of 1) the right CI tools for their businesses and 2) the best platform to host the integration tools. This review of CI on the cloud versus on premise revealed that it is imperative when determining the right CI tool for your need, to consider the right platform to host it to ensure synchronization.

Satish Kumar Boguda

Innovation is disruptive, all of this influences the way we communicate, it streamlines the way we do business and also socializes our way of interacting with people and it is much more connected these days than it has ever been. Throughout the last century, it has taken numerous shapes and sizes, for example: being able to run the entire business on the very first large mainframe computer to finding ways to connect using the client server architecture, as well as from emergence of personal computers to contemporary-day most advanced digital business models, influencing pretty much every division of the economy essentially supporting individuals and organizations to maximize their brand recognition on the value chain. Anymore these days, with such an overwhelming majority of people, processes and intelligent product lines getting completely connected to the Internet, an enormous amount of data is now being accumulated, attempting to make the last decades IT infrastructures and applications fundamentally outdated. Therefore, it creates a burgeoning stumbling block among business executives and IT managers across numerous organizations making them to remodel their enterprise IT strategies that will really sustain the ever-increasing market standards and as such organizations are making every effort pushing themselves to explore numbers of ways to meet such demands. Well into the age of the digital revolution, data is indeed the primary source of fuel for several organizations that already is pushing the horizons of most of the 21st century technologies. Furthermore, with a mission to serve customers in real time, it is quite valuable for businesses to always have greater visibility of their daily operations through accelerating data to insight procedures. In order to overcome all of these challenges, organizations really do need to optimize their enterprise landscape through implementing modern, technology solutions, for example: Edge Computing Platforms designed and built to run Artificial Intelligence and Machine Learning models that effectively facilitate the massive amount of data to be processed in real time at just the edge of the device where it has been generated. This introduces operations closer to the edge of a network, which really minimizes the physical distance and time the data needs to be moved between things (device, smart product or any enterprise application) and cloud storage that enables real-time sensing and improves decision making processes. This brings in enterprises to benefit actionable insights and, at the same time, tends to increase operational efficiency, minimize cost, bring in real-time intelligence and helps to reduce the response time to milliseconds, something that eventually generates a sustainable value proposition for the designed to operate business process, creating the right edge to make each and every decision instantaneously without relying heavily on cloud systems in order to get the instructions.

Komolafe E. O., Awoyelu I. O., Ojetade J. O.

The use of machine learning algorithms to discover knowledge and important relationships in a dataset is a promising trend in recent technological advancement. Machine learning is has been gaining ground in different fields, one of which is Agriculture. Thus, this study presents a predictive model for analysing and extracting important information from available data about how suitable a land is for cultivating cassava. Secondary data that contains result of several land mapping activities, were obtained from the Institute of Agriculture, Research and Training (IAR&T), Ibadan. The collected data was preprocessed and converted to .csv format that can be recognised by WEKA simulator. 10-fold cross validation was used for training and testing. The classification mining rule was generated using J48 algorithm in Decision tree. The decision tree were generated using ‘If-then’ construct models that were used to classify a land to be suitable or not suitable for cultivating cassava.. The classification model had an accuracy of 87.5% and 12.5% for correctly and incorrectly classified instances respectively. The classification model is a potential tool that can be used by extension workers to identify and recommend to farmers, the appropriate land to cultivate cassava.

Salah H. Abid, Saad S. Mahmood, Yaseen A. Oraibi

The aim of this paper is to design a feed forward artificial neural network (Ann) to estimate three dimensional Loranz dynamical map by selecting an appropriate network, transfer function and node weights to get Loranz dynamical map estimation. The proposed network side by side with using Fast Fourier Transform (FFT) as transfer function is used. For different cases of the system, Deterministic, chaotic and noisy, the experimental results of proposed algorithm will compared empirically, by means of the mean square error (MSE) with the results of the same network but with traditional transfer functions, Logsig and Tagsig. The performance of proposed algorithm is best from others in all cases from Both sides, speed and accuracy.

Keisuke Katsura, Fumiko Harada, Hiromitsu Simakawa

The decline of traditional culture is regarded as a problem in Japan. In addition, Japanese with sufficient knowledge of traditional culture are required in order to disseminate Japanese traditional culture to the world. In order to increase Japanese who have sufficient knowledge of traditional culture, it is necessary to cause young people interested. In this research, we propose a method that stimulates user's interest by using a guide avatar in a VR space to provide background knowledge of Japanese culture. The guide avatar is a famous historical figure. The topic that the guide avatar introduces to users is the events related to both the historical figures and traditional crafts. Utilizing story marketing, the guide avatar explains the knowledge for understanding the charm of the traditional crafts to users, thereby stimulating the user's interest. As a result of examining the usefulness of the method, it became clear that compared to the audio guidance provided at a museum, the guide avatar introducing traditional culture can more strongly stimulate the subject's interest in traditional culture. It became clear that, by allowing the well-known person speak the episodes related to both himself and traditional culture, the user’s interest in traditional culture is more strongly evoked. Since young people are interested in VR technology, the method can make them learn traditional culture through story marketing.

Ali Obied

The design of systems, or agents, situated in dynamic environments is of considerable practical and theoretical importance. This paper describes experiments examining the efficacy of dynamic sensing policy when the time cost of processing sensor information is significant. Extending the TILEWORLD experiments performed earlier by Obied, et al. in [1] this article produce interesting development of EBDI-POMDP agent by integrating among knowledge base models, decision theory and Self-Organizing system. This article clear distinguishes between sensing policy and sensing cost, since; it described experiments examining the efficacy of dynamic sensing policy when the time cost of processing sensor information is significant. It is demonstrates that several expected features of sensing cost and planning cost do arise in empirical tests. In particular, it is trying to answer the question is how would scalability of agent improve? The observations that for a given sensing cost and degree of world dynamism, an optimal sensing rate exists and, it is shows how this optimal rate is affected by changes in these parameters.