VOLUME 12, ISSUE 8, AUGUST 2023

Keywords: facial liveness detection system, facial recognition system, liveness detection, convolutional neural network. Works Cited: Fatima Ahmed Abubakar, Atika Ahmad Jibrin, Ishaq Muhammad, Abdulrahman Abdulkarim, Zainab Aliyu Musa, Amatullah Yahaya Aliyu "LIVENESS FACIAL RECOGNITION SYSTEM FOR EXAMINATION HALL", IJARCCE International Journal of Advanced Research in Computer and Communication Engineering, vol. 12, no. 8, pp. 1-6, 2023. Crossref https://doi.org/10.17148/IJARCCE.2023.12801

Abstract: In recent years, the global landscape has witnessed a significant surge in the utilization of artificial intelligence (AI) for devising intelligent solutions across corporate entities and institutions of higher education. This paper underscores the profound impact of AI in shaping the landscape of smart solutions and particularly emphasizes the deployment of AI-powered virtual assistants. These virtual assistants hold the potential to revolutionize ICT technical support provision in public universities by expediting the resolution of routine technical issues faced by students and potentially reducing dependency on dedicated ICT support personnel. However, the successful implementation of such transformative technology necessitates a comprehensive understanding of the requisite ICT infrastructure. This study aims to meticulously analyze the essential ICT infrastructure components crucial for the effective deployment of AI-powered virtual assistants. Employing a content analysis approach, the study gathered pertinent data from a corpus of 20 peer-reviewed journal articles centered on the themes of "ICT infrastructure" and "AI-powered Virtual Assistant." The investigation disclosed prominent AI platforms such as IBM Watson, Google Dialogflow, Microsoft Azure, and Amazon Lex as prevalent choices for Natural Language Processing in constructing AI-powered virtual assistants. These platforms exhibited inherent capabilities encompassing natural language comprehension, processing, and generation. Programming languages such as JavaScript, Java, C#, SQL, Python, and Php emerged as popular choices for coding the AI-driven chatbot. Core hardware requirements included servers, smartphones, routers, and PCs. Pertinent software components encompassed Application Programming Interfaces (APIs), Operating Systems (OS), and Integrated Development Environments (IDEs). Furthermore, Facebook and WhatsApp emerged as prevalent messaging platforms for the testing, deployment, and user interaction aspects of the AI-powered chatbot. The outcomes of this study are anticipated to provide valuable insights for policy makers, IT practitioners, and university administrators, facilitating informed decision-making on the strategic integration of AI-powered virtual assistants to bolster ICT technical support for students.

Keywords: ICT Technical Support Provision, AI-Powered Virtual Assistant, ICT Infrastructure Requirements Works Cited: Paul Oduor Oyile, Roselida Maroko Ongare, Anselemo Peters Ikoha "Assessing ICT Infrastructure Requirements for AI-Powered Virtual Assistants in Delivering ICT Technical Support to Kenyan Public Universities", IJARCCE International Journal of Advanced Research in Computer and Communication Engineering, vol. 12, no. 8, pp. 27-36, 2023. Crossref https://doi.org/10.17148/IJARCCE.2023.12804

Abstract: This paper presents a comprehensive study and analysis on customer segmentation using banking data, aiming to enhance the accuracy and effectiveness of segmentation techniques. The rationale for conducting this research lies in the growing need for personalized services in the banking sector, where understanding customer behavior and preferences is crucial for strategic decision-making. The problem addressed in this study is the challenge of accurate and meaningful customer segmentation, considering the intricate patterns and complexities inherent in banking data. Conventional segmentation methods like k-mean, improve k-mean, and fuzzy c have been widely applied; however, their limitations in handling non-linear and complex data structures necessitate the exploration of more advanced techniques. The methodology employed involves a multi-faceted approach to address the segmentation challenge. Initially, conventional methods such as k-means, improved k-means, and fuzzy c-means are applied to the banking data to establish a benchmark for comparison. These methods are effective for relatively simple data distributions but may fall short in capturing intricate patterns. To address this, a novel approach utilizing spectral clustering is proposed. The proposed method, spectral clustering, leverages the spectral properties of the data to capture underlying structures and relationships. Unlike traditional methods, spectral clustering can effectively identify non-linear and complex patterns in the data, making it suitable for the nuances of banking customer behavior. Through experimentation and analysis, the proposed method's performance is evaluated against the established benchmarks, showcasing its potential to yield more accurate and meaningful customer segments. This research contributes to the field of customer segmentation in the banking sector by highlighting the limitations of traditional methods and introducing a novel spectral clustering approach. The customer segmentation using Neural Network and Spectral Clustering performs well compared to the previous research our proposed system gives an accuracy of 99.54 and also gives the best Gini obtained.

Keywords: Customer Segment, K–Means, Machine Learning, Banking Profiling, Spectral Clustering.

Abstract: This paper aims to analyse exam questions over four academic years in three colleges, to know levels of learning in final examination questions according to Bloom's levels of knowledge. The paper method used is quantitative and qualitative approach and content analysis. To collect the data, this study used documentation , 2318 questions were examined, taken from 110 final exams papers and from different levels, and the exam of the same course must be in all four years or three years at least , from 2015-2018. The analysis was carried out according to Bloom's classification of cognitive levels (knowledge - comprehension - application - analysis - synthesis - evaluation). There are two evaluation performed on this data. The first evaluation done by the experts and the second is electronic evaluation. The questions were distributed on all levels, but in different proportions, and the results were the following: The level of knowledge ranked first with 62.5%, the level of application in the second rank with 18.1%, and the level of comprehension in the third rank by 12.6%, while the percentage of the level of analysis was 3.9%, which is the fourth rank, and the level of evaluation was in the fifth rank by 1.6%, and the percentage of the level of synthesis was in the last rank. By 1.3%. There are no significant differences in the sample of the study in the level of knowledge between expert assessment and electronic assessment which is 0.49% , as well as there is  no statistically significant differences in the level of comprehension which  is 0.40%, and there is no statistical significance at the application level which is 0.84%, and there is no statistical significance at the level of analysis which 4.17%, and there are no statistical differences at the level of synthesis which is  2.22%, and there are no statistically significant differences in the evaluation level which is 3.85%.

Keywords: Bloom’s Taxonomy, Exam, Natural Language Processing. Works Cited: Adlan Balola Ali, Al shefa Abd algadir Hassan, Omar Balola Ali "Electronic Evaluation of Quality of Exams' Questions Written in Arabic Language Based on Bloom’s Taxonomy", IJARCCE International Journal of Advanced Research in Computer and Communication Engineering, vol. 12, no. 8, pp. 50-61, 2023. Crossref https://doi.org/10.17148/IJARCCE.2023.12806

Abstract: Cloud Web APIs have become an essential component of modern software development, allowing developers to access and utilize various services and functionalities provided by cloud providers. Java, a widely used programming language, is frequently employed for developing robust and scalable systems. This research paper aims to provide comprehensive recommendations for effectively integrating and utilizing cloud web APIs within Java-backed systems. The paper discusses the benefits of using cloud web APIs, highlights common challenges, and presents best practices and strategies for successful integration. It also covers security considerations, performance optimization, and potential future developments in the field. Works Cited: Ravikiran Kandepu "Recommendations for the Use of Cloud Web APIs in the Insurance Industry and Its impact on Artificial Intelligence (AI) and Machine Learning", IJARCCE International Journal of Advanced Research in Computer and Communication Engineering, vol. 12, no. 8, pp. 62-70, 2023. Crossref https://doi.org/10.17148/IJARCCE.2023.12807

Abstract: Most people prefer search engines as their initial method of browsing information on the internet. Most of the time, the returned information is not the anticipated search material or is not requires. Search engines progress continuously for fulling human search behaviour for returning the best information available and to identify the search query. Search engines have a time to decide which is the most proper information to give to the searcher since, in these days, the information is extensive and so numerous. The Web crawlers predicts the searching quality on behalf of search engines.

Abstract: The aim of the research is to determine the impact of pradhan mantri ujjwala yojana on rural family status . The objective of 'Pradhan Mantri Ujjwala Yojana' is to provide cooking gas connections to women living below the poverty line. When women are living safe, and productive lives, they can reach their full potential, contribute their skills to the workforce, and raise happy and healthy children. The research population consists of beneficiaries of the Pradhan Mantri Ujjwala Yojana (PMUY) who reside in Aurangabad district. A total of 300 samples were the target population of the study. The dimension of Pradhan Mantri Ujjwala Yojana (PMUY) Scheme selected as Sustainability, Energy Affordability, Accessibility, Safety, environment-friendly and Health & hygiene. The parameter for famility status is includes as Frequency of nutritional diets, Political activism, Increase in employment opportunity, Access to schools, colleges without any hindrance, Moveable and immoveable properties. The household survey was undertaken from the beneficiaries of PMUY . The data was collected through demographic information and interview schedule from   women residing in the rural sector in Aurangabad district of Maharashtra. Descriptive statistics, Chi-square and Regression analysis (By using SPSS) will be used for data processing. The result shows the Pradhan Mantri Ujjwala Yojana (PMUY) was regressed on the predictor women empowerment with respect to family concern

Keywords: Nutritional diets , Political activism, employment opportunity, Access to schools , properties, Health & hygiene

Abstract: Breast cancer is one of the common types of cancer which is affecting health of women population in the world from last few decades. Breast cancer treatments always depend upon early detection, personalized approach and knowledge of disease. From last decade there are many deep learning and machine learning algorithm are implemented by many researchers but accuracy and precision not up to the mark hence mammographic breast density classification is done subjectively by radiologist. In this research article implementation of machine learning algorithm is proposed for mammographic breast density classification. In this approach input images are preprocessed with help of morphological operations; pectoral muscle is removed by Hough transform and Canny Edge detection techniques are used. The images are segmented with the help of Gaussian mixture model and features are extracted using GLCM feature extraction method and then SVM classification is performed on the images. With certain modification this algorithm is suitable for clinical practice.

Keywords: Breast Cancer, BI-RADS Classification, Preprocessing, Segmentation, Feature Extraction, Mammographic Breast Density.

Abstract: The aims of this study was to comprised selected personal characteristics and Analytical Skills between male and female students who newly admitted in the bachelor degree in the College . A total of 55 male English and 55 female English students were selected from various affiliated colleges of Dr. Babasaheb Ambedkar Marathwada University, Aurangabad. The data was collected through respondents in the form self-design questionnaire from different students The data was collected through respondents in the form of different descriptive tests. The Personal Characteristics about, use of smart phone Use of Internet and Participation in sports was obtained before seeking responses. The self-design questionnaire was used to measure the Analytical Skills of the students. Result reveals that 18.18% male students reported mild stress, 27.27% male students reported moderate stress and 54.54%male students reported severe level of stress. Result reveals that 20.00% female students reported mild stress, 23.63% female students reported moderate stress and 56.36% female students reported severe level of stress. The findings of the study revealed that significant differences were found in Analytical Skills between English and other students. Male English students was to found better analytical Skills. Key words: Gender, Analytical Skills, literature, students.

Abstract: This paper offers an extensive examination of energy harvesting approaches, with a special focus on the utilization of piezoelectric energy harvesting. This method capitalizes on the unique capability of specific materials to produce an electric field when subjected to mechanical forces, a phenomenon termed the direct piezoelectric effect. Piezoelectric transducers are available in various forms and materials, rendering them adaptable for a wide array of applications. In addition to this, the paper delves into the significance of modeling the behavior of piezoelectric materials in both the time and frequency domains to enhance their utility. Additionally, it investigates several circuit configurations designed to optimize the efficiency of energy harvesting from piezoelectric devices.

Keywords: Piezoelectric harvesting, Direct piezoelectric effect, Transducer, Efficiency enhancement, Time and Frequency Modelling.

Abstract: This paper proposes an unauthorized entry detector using face recognition technology and a feature to send an SOS message alert. The system consists of a camera that captures images of people entering a restricted area and a face recognition algorithm that compares the captured images with a database of authorized individuals. If an unauthorized individual is detected, the system can send an SOS message alert to security personnel or other designated recipients. The proposed system was evaluated through experiments conducted in a controlled environment. The results showed that the system can accurately detect unauthorized individuals and send an SOS message alert with a high degree of accuracy. The proposed system has the potential to enhance security in various settings, such as airports, banks, and government buildings, by providing an immediate response to unauthorized entries. However, it is crucial for organizations to balance the benefits of this technology with the potential privacy and security concerns and take necessary measures to ensure ethical and responsible use.

Keywords: Eigen Face Recognition, Principal Component Analysis.

Abstract: In today’s world, the rising popularity of the web has drastically expanded the probability of sharing significant information and knowledge on a large scale certainly not seen before. Owing to the availability of the bulk amount of data, the search services on the World Wide Web (WWW) are fetching more demand among users. Regardless of its beneficial part by conventional term-based search engines, precise filtering and retrieving relevant data from the web is considered as a challenging task. In fact, page relevance is the fundamental aspect for the web search, as it supports the current and novel search engines, indexing, crawling, and ranking. Relevancy and popularity of a website are two different things in the world of search engines.

Abstract: Data is the driving force behind the power of modern-day Machine Learning or Deep Learning algorithms. Accuracy and efficiency of these algorithms are largely dependent on the quality of the data that they are trained on; consequently, data poisoning poses a significant threat to these models. Data poisoning attacks present a significant challenge in maintaining the integrity of machine learning models. Currently automated methods and human inspection techniques often fail to identify clean subsets with high precision. In this paper target class of samples for this study’s layer-wise feature analysis includes both poisoned and benign samples. It discovers that the key layer-which is frequently overlooked by existing defences is what distinguishes dangerous from innocuous substances. Key layer analysis of characteristic differences between suspicious and benign samples suggests a simple yet effective approach to filter poisoned samples. Effectiveness of the defences has been verified by in-depth experiments on two benchmark datasets.

Keywords: Data poison, machine learning, Deep Neural Networks, Feature Analysis.

Abstract: Image filtering algorithms are employed to eliminate various types of noise from established images, such as the Lena image. This noise can either be present in the image during the image capture process or as a result of transmission. The present study aims to compare the performance of mean, median, and Gaussian filters in the de-noising of Lena images affected by Gaussian noise, salt and pepper noise, and speckle noise. The evaluation of a performance is carried out using the Peak Signal to Noise Ratio (PSNR) metric.

Keywords: Gaussian noise, Salt & Pepper noise, Speckle noise, Mean filter, Median filter, Gaussian Filter, and PSNR.

Abstract: The integration of powerful technologies such as Internet of Things (IoT) and Multi-Agent Systems (MAS) in healthcare addresses the complex nature of the industry, facilitating communication, coordination, and decision-making among various departments. This becomes particularly crucial in the context such as the COVID-19 pandemic, where developing countries faced increased demand for healthcare services, limited resources, and a lack of robust health systems. Through the utilization of IoT and agent-based systems, remote consultations and virtual doctors can provide essential healthcare services by analyzing patient data and medical history. This paper examined the existing Internet of Things (IoT) approaches in use in health consultancy in Kenya. The study included referral hospitals as sample units, focusing on medical consultants and utilizing scholarly literature recommendations. Three (3) Hospital facilities were selected based on their capacity for training, research, and referrals. The study respondents comprised the general superintendent, medical consultants, health system managers, medical students, and patients. Interviews and survey questions were used for data collection. The instruments average validity test score was (.84) and reliability score of (.799) based on Chronbach’s Alpha. The findings of the study reveals that some hospitals have integrated IoT technology in health consultancy services in Kenya with a significant improvements in data management, diagnosis accuracy, and patient outcomes, but there is need to address concerns regarding data security, privacy, standardization and infrastructure which was pointed out to be crucial for fully harnessing its potential. It also indicates that in Kenya, a majority of healthcare facilities are using Electronic Health Record (EHR) systems and have reliable internet connectivity, although there are variations in the availability of hardware and software technology, suggesting the need for targeted improvements and investments in the healthcare technological infrastructure.

Keywords: agent-based systems, health care system, health consultancy services, internet of things, healthcare

Abstract: Pervious concrete is a unique and eco-friendly alternative to traditional concrete, known for its high porosity and permeability. Different mix designs, with and without fines, were tested to determine their mechanical properties. The study found an inverse relationship between compressive strength and permeability. Various mixtures were evaluated using aggregate sizes ranging from 12 mm to 4.75 mm, with consistent water content and varying fines content. Pervious concrete typically exhibits compressive strength between 9.18 MPa and 14.06 MPa, and permeability ranging from 5.9% to 12.7%. The desired void ratio falls between 15% and 20%. The study emphasizes the influence of factors like shape, angularity, paste content, size, and water-to-cement ratio on the strength and permeability of pervious concrete. The goal is to implement pervious concrete in different applications to replenish groundwater resources

Keywords: Pervious concrete, Permeable concrete Compressive strength, Void ratio, Ground water Recharge

Abstract: Artificial intelligence and deep learning methods are used in the suggested fall detection system for senior persons in order to reliably recognize falls in real-time. In contrast to conventional systems, which rely on Internet of Things (IoT) gadgets, this system uses wearable gadgets and sensors to collect data, which is then analyzed using AI algorithms. The device can tell the difference between falls and other movements with great accuracy, alerting caregivers or emergency personnel as needed. The system can continuously learn and increase its accuracy over time thanks to the application of AI and deep learning, which ensures accurate fall detection for senior people. As falls are a primary cause of injury and death in the aged population, fall detection systems are becoming more and more crucial. In order to detect falls and notify caretakers or emergency services, traditional fall detection systems rely on Internet of Things (IoT) devices, such as wearable sensors or smart home technologies. These systems, however, can be expensive and might not be available to everyone. In this study, we suggest an IoT-free fall detection system for older people that makes use of artificial intelligence (AI) and deep learning techniques. Our technology uses information from furniture found in most homes, including chairs, tables, and bed frames, to identify falls and notify caretakers. Our system accurately distinguishes falls from other movements and ascertains the fall severity using machine learning techniques.

Keywords: Fall detection system in aged population, IoT, Convolutional Neural Network model, Convolution Neural Network (CNN) architecture, You only look once version (YOLO), MobileNet, ResUNet and DeepUNet.

Abstract: Predicting student dropout using data mining is an important application in the field of education. Dropout prediction models can help educational institutions identify at-risk students early and provide appropriate interventions to improve retention rates.

Keywords: Data Mining, Decision Tree, Deep Learning, Survival Analysis

Keywords: Monozygotic twins (MZ), Intelligence quotient (IQ), Brain size analysis, Algorithmic analysis, Neuroimaging, MRI scans, Computational neuroscience, Brain structure, Automated analysis, Medical research, Cognitive conditions, Neurological disorders, Brain health, Data processing, Computational methods, Research applications, Brain development, Neuroinformatics, Image processing, Neurological insights, Precision measurements, Advanced technology.

Abstract: Natural language processing (NLP), a field of artificial intelligence, has grown and innovated remarkably over the last several years. It is an area of research and application that explores how computers can be used to understand and manipulates natural language text or speech to do useful things. This review paper discuss about the most recent advancements in NLP, taking into account its historical context, its important approaches, cutting-edge models, and applications. It also covers challenges under NLP and future prospects of NLP. This paper could be beneficial to those who wish to study and learn about NLP. Keywords: Natural Language Processing, NLP Advancements, Language Understanding, NLP Methodologies, Ethical Considerations.

Abstract: The evolution of education in the digital era has prompted a fundamental shift towards online learning platforms. This project introduces an advanced E-Learning Management System that harnesses the power of Amazon Web Services to offer a robust, scalable, and secure online educational platform. It is a comprehensive e-learning solution designed to facilitate seamless management and delivery of educational content. It empowers educators to create, organize, and administer courses efficiently, while providing students with a dynamic and interactive learning experience. Leveraging AWS's cloud infrastructure, it ensures high availability, enabling uninterrupted access to educational resources from anywhere, at any time. Key features of E-LMS include user-friendly content creation tools, real-time collaboration, automated assessments, and comprehensive analytics. Educators can create engaging courses, monitor student progress, and offer timely feedback, all while benefiting from AWS's automatic scaling to accommodate varying workloads. Security and data privacy are paramount in E-LMS. AWS's rigorous security protocols are integrated into the system, ensuring the protection of sensitive data and uninterrupted service delivery. Students and educators can engage with confidence, knowing their information is safeguarded.

Keywords: Cloud computing technique, Prescriptive Analytics, Sentiment Analysis, Content-Based Filtering

Abstract: Data from the physical world is connected to IoT-powered computational models through wireless sensor networks. Specialized transducers used in wireless sensor networks enable limited-energy Iot devices to sense their surroundings. One of the most important design considerations in WSNs is energy consumption due to the practical difficulty of replacing batteries in sensor networks. The clustering method is essential for raising the energy effectiveness of a sensor network. The right cluster head choice can help with network load balancing, energy conservation, and increased durability. The cluster head selection process using the Cuckoo search algorithm and the data transmission process using multi hop AODV routing are the key themes of the study. By calculating network efficiency using an average of throughput and remaining energy, the final conclusions are obtained. Simulated findings show that the suggested strategy performs better than the current I-SEP method.  

Keywords: AODV, Wireless Sensor Networks, Internet of Things, Clustering, Energy Efficiency, CSA

Abstract: For contemporary enterprises, the significance of wireless enterprise networks has expanded due to their heightened adaptability and mobility in terms of connectivity and access to information. However, these networks are also vulnerable to various forms of cyber-attacks, including intrusions from external parties, breaches of data, and outbreaks of malware. To counter these threats, it is imperative to possess efficient intrusion detection systems (IDSs). One potential strategy to enhance the performance of IDSs for wireless enterprise networks is the utilization of ensembled machine learning models. To construct an IDS model in a wireless environment employing the AWID dataset, this investigation integrates the prognostications of three distinct classification techniques: specifically, the hybrid model of CNN-SVM, the ensemble model of SVM-MLP, and the ensemble model of DT-KNN. The efficacy of the model is assessed based on the statistical information derived from the confusion matrix, such as accuracy, recall, precision, and F1-scores.

Keywords: Intrusion Detection System (IDS); Machine Learning; Ensemble models; AWID.

Abstract: With the rapidly increasing number of sensors in today’s automotive designs, modern Advanced Driver Assistance Systems (ADAS) are highly dependent on high-performance multi-core computing platforms and high-speed Ethernet networking. In this paper, we first present an industrial-like interconnection architecture between ADAS clusters and computer modules running AI application software and highlight the unique challenges induced with the simultaneous operation of the Ethernet AVB time-triggered and standard traffic classes. We then demonstrate the advantages of creating a flexible AI development platform with the ability to run simulation programs that abstract the realistic interaction of the AI with the ADAS domain. We provide background information on contemporary automotive cluster controllers as well as the real-time data flow components of typical ADAS ECUs, in order to elucidate the associated complexities. Finally, we propose the adaptable Partitioned Gateway module architecture that integrates an automotive microcontroller with a high-performance Gb Ethernet switch to support various ECU clusters, while minimizing latency and network loading for ADAS sensor data preprocessing.

Keywords: ADAS (Advanced Driver Assistance Systems), Ethernet AVB (Audio Video Bridging), Multi-core computing platforms, AI application software, Interconnection architecture, Simulation programs, Automotive cluster controllers, Real-time data flow, Partitioned Gateway module, Gb Ethernet switch

Abstract: The intersection of sustainable finance and artificial intelligence (AI) represents a transformative shift in how financial institutions assess, manage, and promote environmentally responsible investments. With growing concerns over climate change and the need for a sustainable global economy, the role of AI in driving green investments has become more critical than ever. This paper explores the utilization of AI technologies in the realm of sustainable finance, emphasizing their ability to provide data-driven insights that enable smarter, more impactful green investments. By leveraging AI tools such as machine learning, big data analytics, and predictive modeling, investors and financial institutions are better equipped to evaluate environmental, social, and governance (ESG) factors, identify profitable green investment opportunities, and mitigate risks. The integration of AI facilitates the development of green bonds, sustainable funds, and eco-friendly portfolios, contributing to the global shift toward sustainability. This paper also discusses the challenges and limitations of incorporating AI into sustainable finance, including data privacy concerns, lack of standardization in ESG metrics, and ethical considerations. Finally, it concludes by exploring future trends and the potential of AI to further revolutionize sustainable finance, ultimately aligning financial strategies with global sustainability goals.

Keywords: Artificial Intelligence, Sustainable finance, green technologies, carbon credits, Bigdata, ESG.