International Journal of Computer (IJC)

Influence of Internet Usage on Academic Performance of College of Education Students: Rhetoric or Reality?

Benjamin Baiden, Albert Ato-Jackson — Sun, 20 Aug 2023 00:00:00 +0000

The study examined the influence of internet usage on the academic performance of College of Education students in Ghana as being rhetoric or reality. The study adopted descriptive survey design. All year groups (levels 100-400) of St. Joseph’s College of Education were considered as the main population for the study while stratified random sampling technique was used to select 132 respondents. Researchers’ designed questionnaire was used for data collection where Statistical Package for Social Sciences (SPSS version 25) was used for data analysis. The findings of the study revealed that Internet’s influence on the academic performance of the respondents used for the study is a clear-cut reality other than lip service. Internet provides opportunity to acquire special skills; improves their performance during examination; enhances students to study ahead of their teachers; improves students reading competence; promotes their computer skills towards academic activities among others. Nevertheless, few of the respondents reported that Internet usage distracts their attention and prevents them from attending lectures regularly. Based on that, it was recommended that school counselors with the support of the administrators organise enlightenment programmes for students on how to use the internet to improve academic performance. Students in the understudy institution should be encouraged to use the Internet in searching for information that will enhance and improve their academic performance. It is important also to expose the school counselors on training to computer appreciation so that they can give right counselling direction on Internet usage by students regarding their academic activities.

Oral Reading Fluency Can Be Estimated Across Languages with Text-To-Speech Software

Alberto Luis Fernández, Gabriel Enrique Jáuregui — Sun, 22 Oct 2023 00:00:00 +0000

Oral reading fluency (ORF) is a good index of the reading skill level, measured as the number of words read correctly per minute (WRPM). However, ORF tests are not available in many languages. This study tested if the mean of WRPM could be estimated in languages for which ORF tests have not been developed by using free text-to-speech software. Mean time taken by Google Translate (GT) to read out loud 10 texts in 16 languages from the International Reading Speed Texts was compared with the mean time taken by human participants. An English/Other languages ratio was obtained for both reading systems. Both ratios were highly similar, showing that GT is a valid tool to estimate mean WRPM in multiple languages.

Enhanced Covid-19 Contact-Tracing System

Gabriel Ihuoma Lilian, Barifaa Naakorobee — Fri, 10 Nov 2023 00:00:00 +0000

Covid-19 is a global pandemic that has brought the world to a standstill. The virus originated from Wuhan China and has claimed the lives of over 5 million people according to World Health Organization. The Nigeria centre for Disease control is an agency that manages pandemics in Nigeria. They have created awareness son the management of Covic-19. Contact tracing of people that have come in contact with infected people poses a lot of problem. In this study, optimized system for con tact tracing of Covic-19 was carried out. Object Oriented Analysis Design Methodology (OOADM) was adopted and implementation was achieved with python programming language. The result obtained showed better and optimized performance in contact-tracing based on symptomatic (1) and asymptomatic (1+1) infection generation using fuzzy logic as an accurate decision making tool.

Examining the Advantages of Artificial Intelligence Alongside Its Potential Risks on Human Wellbeing, Data Privacy, and National Security

Olushola Agbaje — Mon, 30 Oct 2023 00:00:00 +0000

This study seeks to comprehensively analyze the benefits and risks of artificial intelligence and discuss strategies and policies to balance them. The paper assesses AI's positive impact on four industries - healthcare, finance, transportation, and education – juxtaposed with its negative welfare, privacy, and security effects. The study utilizes a semi-systematic review methodology to explore diverse narratives surrounding AI's societal implications. Key findings suggest AI can improve decision-making, productivity, and quality of life but risks exacerbating bias, unemployment, and insecurity if not developed responsibly. The paper discusses practical strategies, policies, and regulatory interventions to help balance AI's pros and cons, including human-centered design, explainable AI, and governance frameworks. It also suggests actionable recommendations for individual, organizational, and national stakeholders. Suggestions for future research include developing robust AI resilient to attacks, increasing AI transparency and accountability, assessing long-term societal impacts, and addressing legal and ethical dilemmas. This timely study contributes a measured perspective to current debates on AI and provides a framework to help appropriate its advantages while mitigating its perils.

Implementation of Artificial Intelligence in Traffic Management in the United States

Olushola Adegoke — Sun, 10 Dec 2023 00:00:00 +0000

This paper investigates the application and deployment of artificial intelligence (AI) in enhancing traffic management within the U.S., focusing mainly on predicting future traffic demand using machine learning and deep learning models. Utilizing datasets from the Tom-Tom Traffic Index and the Python programming language for data processing, the study aims to mitigate traffic congestion through accurate traffic prediction. The study specifically examines Baltimore, Maryland (used as a proxy for major U.S. cities) to assess the efficiency of AI technologies on traffic levels and provides a comparative analysis of machine learning and deep learning algorithms (decision tree, random forest, logistic regression, and deep learning neural network). The results revealed that decision tree models surpass other algorithms with an 85% accuracy rate in congestion prediction. The study contemplates the technical aspects of traffic management systems and addresses the practical implications for city planning and the overarching goals of reducing congestion and facilitating transportation logistics. The paper offers valuable insights to transportation planners, logistics managers, and academic researchers.

A Systematic Review of NOMA Variants for 5G and Beyond

Duaa Saleem Hassan, Ahmed G. Wadday — Sun, 26 Nov 2023 00:00:00 +0000

With the fast expansion of the Internet of Things (IoT), there is an exponential need for mobile intelligent terminals .However, the connectivity of large-scale intelligent terminals is constrained by increasingly restricted spectrum resources. To address this issue, non-orthogonal multiple access (NOMA) technology, which can handle more users with less resources, is predicted to enable future wireless networks beyond 5G,., 6G, to give huge terminal access. The fundamental idea behind NOMA is to superimpose signals from numerous users on the same time-frequency resource prior to transmission. At the receiver, serial interference cancellation (SIC) technology is used to reduce interference among users. In this review paper we discusses the principles of the strong candidate Non-Orthogonal Multiple Access (NOMA) approach, as well as how it can best match the requirements of the Fifth Generation (5G) requirements in practical applications.

Assessing Machine Learning's Accuracy in Stock Price Prediction

Aryan Bhatta, Pranshu Poudyal, Drishant Kumar Maharjan, Aryaa Thapa — Mon, 25 Sep 2023 00:00:00 +0000

This research examines how well machine learning models can predict the closing price of traded stocks. The financial industry has seen an increase, in the use of these models due to the availability of datasets and technological advancements. The study compares machine learning models such as Linear Regression, Random Forest and K Nearest Neighbor (KNN) to determine which ones are the accurate predictors and what factors contribute to their effectiveness. To gain insights into model performance a diverse dataset consisting of five stocks from sectors is used. Data analysis and modeling are conducted using Python programming language with libraries, like Pandas, NumPy, Matplotlib and Scikit learn. The performance evaluation metric utilized is Mean Squared Error (MSE). The research findings have the potential to assist investors and traders in making decisions while also contributing to the growth of the financial industry.

A Lightweight Way to Secure Automotive Networks Using CAN/CAN-FD

Sukhyun Seo — Wed, 27 Sep 2023 00:00:00 +0000

In-vehicle communication uses the CAN/CAN-FD bus, and communication speed and security are important. As current CAN/CAN-FD communication is used without encryption, many cases of vehicle hacking have been reported over time. With the advent of autonomous driving and connected cars, vehicles are no longer independent; they can be infiltrated from the outside and personal information such as vehicle location and driving habits can be accessed through the vehicle, posing a serious threat to personal privacy and life. Therefore, communication data needs to be encrypted to increase the security of communication. In this paper, data frames are encrypted using a shuffling algorithm in the CAN/CAN-FD communication system environment. We also compare and analyse standardised encryption methods, namely AES and ARIA, and shuffling algorithms, and suggest ways to increase the security and communication speed in the vehicle.

Formulation of a Computational Model for Predicting Drug Reactions Using Machine Learning

Christopher Agbonkhese, Hettie Abimbola Soriyan, Kolawole Mosa — Sun, 26 Nov 2023 00:00:00 +0000

In the rapidly evolving landscape of healthcare, the efficient detection of drug reactions is of paramount importance to ensure patient safety and optimize treatment outcomes. This article presents the formulation of a computational model for the prediction of drug reactions in clinical settings using machine learning techniques. Our research leverages state-of-the-art machine learning algorithms to extract valuable insights from health records and prescription data. By systematically analyzing the relationships between prescribed medications and observed patient reactions, our computational model will be able to identify potential drug reactions emanating from drug prescription in clinical a clinical setting.

Delineating International Cooperation in the Fight against Cybercrime in Cameroon

Kwei Haliday Nyingchia, Clinton Atabongakeng Fobellah, Elvin Fuwain Ndiwum — Mon, 25 Sep 2023 00:00:00 +0000

The advent of new technologies and the increase in their use have ushered in a new chapter in how things are being done in contemporary society. Though plausible, it has also paved the way for crimes (cybercrimes) to be committed through electronic means on a global scale. This has greatly undermined the territorial integrity of nations, and it poses a significant problem to the global community in general. Currently, the effect of cybercrime is something the global economy cannot afford to ignore. It has increased security risks of critical infrastructures, brought about massive privacy invasion and attacks on businesses, and state security. It is difficult to stop crimes of this nature since technology is always evolving and the world is becoming more connected. It therefore requires a well-coordinated and concerted effort from governments around the world to contain crimes of this nature. It is in this line of reasoning that the Cameroon government has made significant strides through the 2010 law on cyber security and cyber criminality (Hereafter referred to as the Cyber Law) to foster cooperation with other nations in a bit to curb the spread of cybercrime in Cameroon. Despite so, the efforts are not sufficient and the prevalent nature of these offenses today still largely smashed government efforts to the ground. This paper sets out to examine the efficiency of the measures taken by the Cameroon Government to forge international cooperation with the aim to combat cybercrime.

Natural Language Processing for Cyberbullying Detection

Jerry He, Lisa Chalaguine — Sun, 22 Oct 2023 00:00:00 +0000

With the development of digital technologies and the popularity of social media, cyberbullying has become a serious public health concern that can lead to increased risk of mental and behavioral health issues or even suicide. Artificial intelligence like machine learning opens a lot of possibilities to combat cyberbullying, e.g. automatic cyberbullying detection. Most recent research focuses on improving performance by developing complex models that demand more resources and time to run. The research uses publicly available datasets without carefully evaluating their feasibility and limitations. This study uses natural language processing (NLP) to evaluate the model performance and examine the difference between fine-grained classification and binary classification as well as assess the feasibility and quality of the publicly available dataset. The results show that simple classifier can also achieve similar performance as that of more complex models if appropriate preprocessing is used, and the publicly available dataset may have limitations and quality issues that researchers should consider when using the data.

Student Attrition Prediction Using Machine Learning Techniques

Sat, 02 Sep 2023 00:00:00 +0000

In educational systems, students’ course enrollment is fundamental performance metrics to academic and financial sustainability. In many higher institutions today, students’ attrition rates are caused by a variety of circumstances, including demographic and personal factors such as age, gender, academic background, financial abilities, and academic degree of choice. In this study, machine learning approaches was used to develop prediction models that predicted students’ attrition rate in pursuing computer science degree, as well as students who have a high risk of dropping out before graduation. This can help higher education institutes to develop proper intervention plans to reduce attrition rates and increase the probability of student academic success. Student’s data were collected from the Federal University Lokoja (FUL), Nigeria. The data were preprocessed using existing weka machine learning libraries where the data was converted into attribute related file form (arff) and resampling techniques was used to partition the data into training set and testing set. The correlation-based feature selection was extracted and used to develop the students’ attrition model and to identify the students’ risk of dropping out. Random forest and random tree machine learning algorithms were used to predict students' attrition. The results showed that the random forest had an accuracy of 79.45%, while the random tree's accuracy was 78.09%. This is an improvement over previous results where 66.14% and 57.48% accuracy was recorded for random forest and random tree respectively. This improvement was as a result of the techniques used. It is therefore recommended that applying techniques to the classification model can improve the performance of the model.

A Validity of in-Vehicle Networks Using CAN-FD

Sukhyun Seo — Wed, 27 Sep 2023 00:00:00 +0000

The most common communication interface for automotive electronic control units is CAN (Controller Area Network). Since CAN was first introduced in Daimler vehicles in 1991, all automotive manufacturers have adopted CAN communication for in-vehicle networks. However, as the number of electronic control units connected to the CAN network grows rapidly, the CAN protocol is reaching its technological limits. To overcome this limitation, Bosch has introduced a new communication protocol, CAN-FD (Flexible Data-rate). In this paper, we analyse the characteristics and limitations of CAN-FD communication according to the topology under the in-vehicle wiring harness environment designed based on the existing classic CAN communication.

Smart Drone with Renewable Smart System

Mays A. Al-bahrany, Ahmad T. AbdulSadda — Wed, 29 Nov 2023 00:00:00 +0000

In order to lessen its negative effects on the environment and to maintain its future operations in a clear, renewable, and sustainable manner, the aviation industry has begun developing designs that are dependent on alternative energy sources but also friendly to the environment and conventional energy. Solar energy has been suggested as a potential remedy. Aerial vehicles driven by solar energy are viewed as essential to limiting the consequences of global warming. In this study, a MATLAB/Simulink environment is used to simulate a mathematical model of a solar-powered BLDC motor of a UAV. under photovoltaic (PV) array systems, the phrase "maximum power point tracking" (MPPT) is crucial to ensuring that, under specific circumstances, the connected systems receive the greatest power output. This study simulates "fuzzy logic control," one of the preferred MPPT methods, using a solar-powered BLDC motor for an unmanned aerial vehicle (UAV) design. The PV cell, MPPT, buck-boost converter, and BLDC motor models in the cascade structure are simulated, tested, and the results are compared to the DC motor technical data. As a result, despite changes in irradiance, the results of mathematical model simulation overlap with motor technical reference values. A mathematical model of a solar-powered BLDC motor for a UAV is created and simulated using the MATLAB/Simulink environment, in contrast to prior solar-powered BLDC motor literature efforts. The fuzzy logic control MPPT technique is preferred for adjusting the maximum power output at the solar cell, and a buck-boost converter structure is connected between the MPPT and the BLDC motor mathematical model. It is recommended for usage in solar-powered UAV designs in the future.