Accepted Papers

A Kriging-hdmr Combining With Adaptive Proportional Sampling for Multi-parameter Approximate Modeling

Yili Zhang1, Hanyan Huang1*,1School of Systems Science and Engineering, Sun Yat-Sen University, Guangzhou, Guangdong 510275, PR China

ABSTRACT

High-dimensional complex multi-parameter problems are commonly in engineering, while the traditional approximate modeling is limited to low or medium dimensional problems, which cannot overcome the dimensional disaster and greatly reduce the modelling accuracy with the increase of design parameter space. Therefore, this paper combined Kriging with Cut-HDMR, proposed a developed Kriging-HDMR method based on adaptive proportional sampling strategy, and made full use of Kriging s own interpolation prediction advantages and corresponding errors to improve modeling efficiency. Three numerical tests including coupling test, high-dimensional nonlinear test and calculation cost test were used to verify Kriging-HDMR, and compared with the traditional Kriging and RBF-HDMR in R2, REEA and RMEA measuring the approximate accuracy, results show that the improved Kriging-HDMR greatly reduces the sampling cost and avoids falling into local optima. In addition, at the same calculation cost, when the scale coefficient is 1/2, Kriging-HDMR has higher global approximate accuracy and stronger algorithm robustness, while preserving the hierarchical characteristics of coupling between input variables.

KEYWORDS

Multiparameter decoupling, Kriging-HDMR, Surrogate model, Global approximation.

A Community Based Mobile Application to Reduce Waste From Un-used Bikes Using Social Media

Alexander Junwen Tan1, Jonathan Sahagun2, 1Northwood HighSchool, 4515 Portola Parkway. Irvine, CA 92620, 2Computer Science Department, California State Polytechnic University, Pomona, CA 91768

ABSTRACT

Around 15 million bikes are discarded annually, which poses an environmental risk [1]. The rubber from bike tires takes a long time to decompose, and toxic chemicals are released into the soil during this process [2]. Additionally, the popularity of e-bikes is increasing, and the lithium batteries they use harm the environment during extraction. To address this problem, a bike donation app is proposed, which reduces the number of bikes produced, minimizes waste, and benefits those in need [3]. By operating online, the cost of running the operation is minimal, and the project can reach and help anyone with internet access. However, the app s success relies on a user base, which may be a significant challenge. Furthermore, the app s design may need improvement to attract users. Blind spots in the program may include inaccurate bike donation recommendations and a lack of proper verification for donated bikes safety and condition. An A/B test shows that personalized recommendations through the app increased the conversion rate for successful bike donations. The verification process for donated bikes was effective in ensuring the bikes safety and quality. By developing a mobile app that provides personalized recommendations and addresses bike waste, the project contributes to sustainable transportation and reduces environmental harm [4].

KEYWORDS

Environment, Application, Donation, Bikes .

Design of a Gdi Technique Based Reversible Logic Sram Array for Object Tracking Applications

Dr.G.Sowmya, Ms.B.Alekya Himabindu, Ms.V.Nagamani, Ms.S.Jayamangala, Department of Electronics and communication Engineering, SREC, JNTUA, Nandyal, Andhrapradesh

ABSTRACT

With the advancement of technologies and the form of use of electronic equipment in various applications, data collection or retrieval involves tremendous memories. Because of its high-speed access capabilities, static SRAM cells are usually used. When the size of the memory increases, the power used by the memory is often rising exponentially. Due to their low power characteristics, reversible circuits have gained popularity in recent years. This article discusses a novel SRAM cell architecture using reversible logic. In addition, GDI technique has been successfully introduced in the Reversible logic gates to meet the present day system power consumption requirements. The design specifications of the 4 X 2 SRAM array to store information related to object detection and tracking are also discussed in this article.

KEYWORDS

SRAM, Reversible Logic, GDI technique, Power Consumption.

Cyclic Quantum Teleportation of Two-qubit Entangled States by Using Six-qubit Cluster State and Six-qubit Entangled State

Abdallah Slaoui1, 2, 1LPHE-Modeling and Simulation, Faculty of Sciences, Mohammed V University in Rabat, Rabat, Morocco, 2Centre of Physics and Mathematics, CPM, Faculty of Sciences, Mohammed V University in Rabat, Rabat, Morocco.

ABSTRACT

Cyclic quantum teleportation schemes requires at least the existence of three collaborators acting all as sen- ders and receivers of quantum information, each one of them has an information to be transmitted to the next neighbour in a circular manner. Here, new cyclic quantum teleportation scheme is proposed for perfectly trans- mitting cyclically three arbitrary unknown two-qubit entangled states (a, ß and ?) among the three collaborators. In this scheme, Alice can send to Bob the quantum information contained in her two-qubit entangled state a and receive from Charlie the quantum information contained in the two-qubit entangled state in his possession ? and similarly, Bob can transmit to Charlie the quantum information contained in his two-qubit entangled state ß through a quantum channel of twelve-qubit state consisting of a six-qubit cluster state and a six-qubit entangled state by sequentially and cyclically performing Bell state measurements. Subsequently, each one of the three participants can afterwards retrieve his own desired two-qubit entangled state using classical channel and by performing appropriate unitary Pauli operators and we have shown that our proposed scheme performs efficiently.

KEYWORDS

Cyclic quantum teleportation, Two-qubit entangled state, Six-qubit cluster state, Six-qubit entan- gled state, Bell states measurement.

Psychological Lights: an Intelligent Led System to Relief Youth Stress Level Using Ai and Internet of the Things

Zixuan Cheng1, Zihang Cheng2, Ang Li3, 1, 2Basis Oro Valley, 11155 N Oracle Rd, Oro Valley, AZ 85737, 3California State University Long Beach, Long Beach, CA 90840

ABSTRACT

The paper discusses the issue of stress among students and proposes using lighting to alleviate stress levels [3]. The authors discuss various techniques for managing stress, including exercise, sleep, and socialization, and suggest that lighting can be used to address seasonal affective disorder (SAD) [4][5]. The paper outlines the challenges faced during the experiment and design, including creating a reliable survey, user interface design, and data privacy. The authors propose using a weighted score for survey responses and adopting simple designs for the app interface. The paper concludes by discussing the potential benefits of using lighting to alleviate stress levels and identifying areas for future research.

KEYWORDS

Mental health, Organization, Smart home, Mobile app.

Multi-document Summarization for Kannada Text Using Deep Learning Techniques

KanishJain, KarthikV, NamanJain, RajeshKumarN, VRBadriprasad, Department of Computer Science, PES University, Bangalore

ABSTRACT

The advent of technological breakthroughs in communication has led people to express themselves in innumerable ways, both verbally and in writing. Language is an instrumental part of this communication. This leads to a hugeamountof data generated and requires automation to analyze,summarize,and gather insights from the data.This paper proposes a method to combine Named Entity Recognition (NER) with deep learning techniques for Kannada textsummarization. Documents are pre-processed to extract name dentities,then with the seas features are trained on a deep learningm odel for summarization. The technique is evaluated against a number of state of the art benchmark datasets and demonstrates notable e?ectiveness in generating high-quality summaries of documents. The results show a considerable improvement over traditional summarization methods, highlighting the potential of this approach for improving the e?ciency and accuracy of multidocument Summarization.

KEYWORDS

NamedEntityRecognition,Encoder,Decoder,OracleSummary.

Role of Image Processing in Dentistry

Ramyaalakshmi1 A and Poonguzhali S2, 1Department of Computer Science, Research scholar VISTAS, Chennai, 2Associate Professor, Department of Computer Applications, VISTAS Chennai

ABSTRACT

Image Processing plays an important role in many industries. One of those is ‘Dentistry’. Image processing is always of great help to all dentists and clinicians for detecting and diagnosing the disease. To identify appropriate treatment the digital dental image must have better contrast of features. Usually, a dental image process is a tedious process and also a time-dragging process because normally human teeth are uneven and non-structural. Moreover, the X-ray images vary due to intensity, noise and contrast leading to more challenges in employing image processing. A dental X-ray is always pre-processed to give a good contrast image. To evaluate the dental disease, segmentation of image features plays a vital role. This paper reviews the image processing techniques, its features along with their applications and gives the comparative study about how the techniques are used.

KEYWORDS

Image Processing, X-rays, dentistry, Applications of Image Processing, Dental radiography.

Exploring ICT Adoption in Teaching and Learning of Science: a Case of Senior Teachers in Kenya

David Ochieng Odhiambo1 and Dr. Winston Edward Massam (PhD)2, 1Graduate student (Science education), Aga Khan University, IED-East Africa, Tanzania and 2Faculty, Aga Khan University, IED-East Africa, Tanzania

ABSTRACT

This study was conducted in Homa Bay Sub County, Kenya to explore the adoption of ICT among senior teachers, aged between 45-60 years, in teaching and learning of science. A qualitative research approach using a case study design was employed. Eight secondary school senior teachers of science, 2 females and 6 males, within Homa Bay Sub County were purposefully selected. Semi structured interviews were conducted one-on-one basis followed by classroom observation and document analysis. Analysis of the data revealed that senior teachers of science generally integrate ICTs in planning, assessment and in classroom teaching; however pedagogical preferences and teacher’s workload had an influence on their choice of pedagogy. Subsequently, traditional pedagogy dominated their classroom lessons while ICT partially integrated as complementary to teach specific topics in science. The major factors revealed, that stalled ICT adoption in science teaching among the senior teachers, were insufficient technical support that reduced confidence of ICT integration, scarce practical training on ICT usage, basic expertise level in using ICT and inadequate ICT infrastructures in schools. Nonetheless, the senior teachers of science acknowledged that ICT is convenient and beneficial in teaching and learning. Thus, for increasing ICT integration by senior teachers of science in teaching and learning, it was recommended that the senior teachers of science be exposed to consistent ICT training to be abreast with current ICT knowledge, policies developed for teachers’ preparation and professional development be supportive and emphasize ICT integration in pedagogy.

KEYWORDS

ICT adoption, senior teachers of science, pedagogy, ICT integration in teaching and learning, diffusion innovation.

Data Privacy Between Perception and Learning: an Educational Chatbot to Get the User Sensitized and Trained to Rights Awareness Through Human-centered Design and Personalized Learning

Sergio Guida and Independent Researcher, Data Governance, AI / Sr Man.Cons., Italy

ABSTRACT

This study was conducted in Homa Bay Sub County, Kenya to explore the adoption of ICT among senior teachers, aged between 45-60 years, in teaching and learning of science. A qualitative research approach using a case study design was employed. Eight secondary school senior teachers of science, 2 females and 6 males, within Homa Bay Sub County were purposefully selected. Semi structured interviews were conducted one-on-one basis followed by classroom observation and document analysis. Analysis of the data revealed that senior teachers of science generally integrate ICTs in planning, assessment and in classroom teaching; however pedagogical preferences and teacher’s workload had an influence on their choice of pedagogy. Subsequently, traditional pedagogy dominated their classroom lessons while ICT partially integrated as complementary to teach specific topics in science. The major factors revealed, that stalled ICT adoption in science teaching among the senior teachers, were insufficient technical support that reduced confidence of ICT integration, scarce practical training on ICT usage, basic expertise level in using ICT and inadequate ICT infrastructures in schools. Nonetheless, the senior teachers of science acknowledged that ICT is convenient and beneficial in teaching and learning. Thus, for increasing ICT integration by senior teachers of science in teaching and learning, it was recommended that the senior teachers of science be exposed to consistent ICT training to be abreast with current ICT knowledge, policies developed for teachers’ preparation and professional development be supportive and emphasize ICT integration in pedagogy

KEYWORDS

Human-Centered Design, User experience, Morenian psychodrama, Situated Pychological Agents Framework, Technology-Enhanced Learning, data privacy rights, educational chatbot. .

IoT Network Proposal for the Identification, Monitoring and Location of Crocodiles in the Estuary of Puerto Vallarta, jalisco, Mexico

Miguel Angel Gallardo Lemus1, Juan Carlos RodriguezRamos2 and Rodrigo Oliver Delgado Arcega3, 1Academia de Sistemas E ITICs, Campus Vallarta, ITMMPyH, Mexico 1, 2Academia de Sistemas E ITICs, Campus La Huerta,ITMMPyH, Mexico 2 and 3Academia de Electromecánica, Campus Vallarta, ITMMPyH, Mexico 3

ABSTRACT

In the present work, the analysis and designproposalofanetworkofwirelesssensorsforthemonitoring and surveillance of rocodiles founding the area of the estuary and the Marina of Puerto Vallarta, Jalisco, which is an area surrounded by commercial and residential areas. A LoRa network and a NoSQL database service such as Firebase and a data visualization using React Native are proposed.

KEYWORDS

IoT, Wireless Sensor Networks (WSN), react native, LoRa, animal monitoring.t. .

Sustainable Electrical and Electronics Technology Education in Nigeria

Tombari James, Department of Electrical and Electronics Technology, Federal College of Education (Technical), Omoku, Rivers State, Nigeria

ABSTRACT

Over the past two years, educational institutions and the society at large has witnessed certain daring challenges that altered the nature of work as well as formal education. There is an increasing awareness of global action plans. Hence this paper reviews scholarly literature on the nature (curriculum content, pedagogies, and teacher education) of education needed to sustain students positive motivation, ensure transformative learning, and ensure achievement of the sustainable development goals specifically with respect to electrical and electronic technology education in the Nigerian situation. Nigeria’s contemporary needs were identified, likewise the benefit of a sustainable electrical and electronic technology education in mitigating those needs.

KEYWORDS

Electrical and Electronic, Technology Education, Sustainability, Education for Sustainable Development, Contemporary challenges, Nigeria.

Educating the Expat Educator in Japan: Why Awareness of the Politicization of English Education in Japan Matters

Peter Joun, Department of Economics, Hosei University, Tokyo, Japan

ABSTRACT

Despite huge investments of time and money as well as decades of research into classroom methodology, the English proficiency of Japanese students arguably remains less than satisfactory, with recent surveys finding that Japan has fallen significantly behind South Korea, Vietnam, and China. Such results seem to reflect comments heard amongst expatriate instructors of English who work in Japan that express frustration at getting Japanese students to participate. Research has found that it is often the western instructors’ unfamiliarity with Japanese styles of learning and classroom interaction patterns that results in less-than-ideal engagement. The pronounced tendency to instruct students from a view which prioritizes the native speakers cultural assumptions, may in turn be linked to how heavily politicized English education in Japan has been in the post-war era.

KEYWORDS

standardized testing, TOEIC, political environment, hegemony, critical discourse analysis.

A Mathematical Related Program Used for Amc Competition Practicing Using Unity Engine

Andrew Zhou1, Moddwyn Andaya2, 1Northwood High School, 4515 Portola Parkway Irvine, CA 92620, 2Computer Science Department, California State Polytechnic University, Pomona, CA 91768

ABSTRACT

The American Mathematics Competitions (AMC) are a set of challenging exams and curriculum materials designed to improve the problem-solving abilities and mathematical knowledge of middle and high school students. The competition is divided into three levels based on student age, with AMC8 for 8th grade or lower, and AMC10 and AMC12 for 10th grade or lower and 12th grade or lower, respectively. However, not all students have the necessary skills to excel in these competitions, and the average scores obtained by students range from 50 to 70. To address this issue, this article proposes a novel practice method that combines gaming and math problem-solving. The proposed game is based entirely on mathematical problem-solving, with questions drawn from past AMC exams. By incorporating the visual appeal and interactive nature of gaming, students can stay engaged and focused while practicing their math skills. Additionally, the game can simulate the rules of AMC exams, such as time limits and pass lines, making it an effective tool for practicing for the actual competition [1]. This approach promises to significantly enhance the efficiency of math problem-solving practice while making it more enjoyable for students.

KEYWORDS

AMC, Unity, Learning, Mathematical.

Self-perception of Physical Beauty and Its Influence on Personality

Rivas-Huaman, Rolly Guillermo, National University of Education Enrique Guzmán y Valle

ABSTRACT

Introduction: Beauty is inherent to the human being, whether in what it observes, in what it can produce or in what it can reflect in the sight of other people. Beauty has always had an influence on mans behavior, on the formation of his personality and on his psychological health. Objectives: to rehearse an analysis of self-perception of beauty from philosophy and psychology and its repercussions on psychological health. Development: Philosophy could not be alien to it and through various exponents; it tries to answer what beauty is and what beauty can generate in man. Psychology, on the other hand, tries to explain what personality and behaviors are how it is formed, what factors influence it. The self-perception of physical beauty can acquire a determining role in the personality, but ideally, it should take a back seat. This essay aims to explain the self-perception of physical beauty and its relationship with the onset of psychological disorders. To do this, a brief journey through history is made, considering the philosophical and psychological foundations with respect to beauty and personality. Conclusions: if the self-perception of physical beauty is empowered exaggeratedly or poorly, they can awaken psychological disorders. Parents play an important role for better or for worse, consciously or unconsciously. The importance of a balanced and comprehensive education, stimulating the self-perception of inner beauty, seeking its prevalence over the self-perception of physical or external beauty, in the training of children, is essential to avoid the appearance of psychological disorders related to the self-perception of physical beauty.

KEYWORDS

Physical beauty, personality, psychological disorders.

A Data-driven Application for Matching Student Traits With Learning Opportunities Using Artificial Intelligence

Chongda You1, Andrew Park2, 1Portola High School, 1001 Cadence, Irvine, CA 92618, 2Computer Science Department, California State Polytechnic University, Pomona, CA 91768

ABSTRACT

It is often limited to students the opportunities they can get to apply their knowledge and learn new things, as different opportunities have vastly different ways of advertising. Just like the students, many organizations are looking for passionate students to apply their knowledge and energy to benefit society [2]. To solve this problem and ease the difficulties in finding the right opportunities, Maclever aims to be a place where all organizations can post their opportunities for students, and students can use the artificial intelligence-based feature in this application to find opportunities that best fit their skills. Maclever aims to be a simple and effective connection between organizations and students [3]. Leveraging tools such as sentiment analysis and utilizing models based on user behaviors and preferences to better match valuable connections allows us to create a system that gives us a much stronger ability to address the goals Maclevers sets out to solve.

KEYWORDS

Data-Driven, Sentiment Analysis and Models, User Interaction, Sentiment Analysis Model.

AI-Related Offensive Technologies in(Cyber) Warfare and Nuclear Security

Tatiana (Anastasia) Kyttaroudi and Department of Applied Informatics, University of Macedonia, Thessaloniki, Greece

ABSTRACT

Despite the world’s generalized stability, new technologies have affected every aspect of everyday life, from transportation and communication to military capabilities. The new possibilities technologies like Artificial Intelligence have introduced, along with the unexplored aspects they entail, present an unprecedented multi-faceted reality in the military domain. Weaponized technologies are being developed in parallel with defensive capabilities, enabling a race between the two and between the states that engage in relevant research. These novelties raise a new modus operandi on the conventional battlefield and in cyberspace, reshaping the very nature of war. Incorporating AI-enabled offensive technologies in war is followed by unfamiliar benefits, challenges, legal and ethical questions. In this paper, some of the most common AI-enabled technologies used in warfare will be discussed, with relevant case studies and some of the major states’ national developments, all while presentingthe benefits and challenges of integrating cutting-edge technologies in the military sphere.

KEYWORDS

AI warfare, AI nuclear security, AI cyber warfare, combat AI.

Efficient Implementation of Tanh: A Comparative Study of New Results

Samira Sorayaasa1 and Majid Ahmadi2,FIEEE, FIET, Department of Electrical and Computer Engineering, University of Windsor, Windsor, Ontario, Canada

ABSTRACT

Hyperbolic tangent (Tanh) activation function is used in multilayered artificial neural networks (ANN). This activation function contains exponential and division terms in its expressions which makes it accurate digital implementation difficult. In this paper we present two different approximation techniques for digital implementation of Tanh function using power of two and coordinate rotation digital computer (CORDIC) methods. A comparative study of both techniques in terms of accuracy of their approximation in hardware costs as well as speed of their implementations on FPGA is also explained.

KEYWORDS

ANNs, Tanh, activation function, approximation with power of two, CORDIC algorithm, FPGA, optimization, hardware resource, latency, error.

Predicting the Acceptance of Academic Papers According to Writing Style

Xi Deng1, Shasha Li1, Jie Yu1, Jun Ma1, Bin Ji1, Wuhang Lin1,Shezheng Song1 and Zibo Yi2, 11College of Computer, National University of Defense Technology, Changsha, China, 12Information Research Center of Military Science PLA Academy of Military Science, Beijing, China

ABSTRACT

The task of predicting the acceptance of academic papers is of great significance for novices in the field of dissertation writing. Besides paper texts, most existing models rely on additional information without focusing on the writing style of the paper texts. More importantly, these models make fewer attempts to exploit the interactions and aggregations in the hierarchical structure of the paper text. To address these issues, we propose a novel modular hierarchical model(MHM) to make predictions of paper acceptance. The input to our model is only the original paper, and no other extra information is required. We capture the hierarchical structure of paper texts with three encoders: a WtoS encoder, a StoP encoder, and a paper encoder. To this end, the WtoS encoder uses the pre-trained language model SciBERT to obtain the sentence representation from the word representation. The StoP encoder lets sentences in the same paragraph interact and aggregates them to get paragraph embeddings based on importance scores. The paper encoder does interaction among different hierarchical structures of three modules of a paper text: the paper title, abstract sentences, and body paragraphs. Then this encoder aggregates new representations generated into a compact vector. In addition, the paper encoder models the guiding role of the title and abstract, respectively, generating another two compact vectors. We concatenate the above three compact vectors and additional four manual features to obtain the paper representation. This representation is then fed into a classifier to obtain the acceptance decision. Experimental results on a large-scale dataset show that our model consistently outperforms the previous strong baselines in four evaluation metrics. Quantitative and qualitative analyses further validate the superiority of our model.

KEYWORDS

Acceptance, Academic Papers, Modular, Hierarchical, Interact.

An Automated Generation Fromvideo To3dcharacteranimation Using Artificial Intelligence and Poseestimate

Daniel Haocheng Xian1, Jonathan Sahagun2, 1Catlin Gabel School, 8825 SW Barnes Rd, Portland, OR 97225, 2Computer Science Department, California State Polytechnic University, Pomona, CA91768

ABSTRACT

This paper presents a novel approach to automatically generate 3D character animation from video using artificial intelligence and pose estimation [3]. The proposed system first extracts the pose information fromthe input videousing a pose estimation model [2]. Then, an artificial neural network is trained to generate the corresponding3Dcharacter animation based on the extracted pose information [1]. The generated animation is then refined usingaset of animation filters to enhance the quality of the final output. Our experimental results demonstrate theef ectiveness of the proposed approach in generating realistic and natural-looking 3D character animations fromvideo input [4]. This automated process has the potential to greatly reduce the time and ef ort required for creating3D character animations, making it a valuable tool for the entertainment and gaming industries.

KEYWORDS

3D modeling, Artificial Intelligence, Animation.

Predicting the Effectiveness of New Product Cigarette Launch Strategy: a Study Based on Synthetic Control Method

Yu-Hua Mo1, Chao Deng2, Fei-Jie Huang3, Qian Tan4* and Yuan-Kun Li5, 1,2,3,4*China Tobacco Guangxi Industrial Co., Ltd. Nanning, China, 5PBC School of Tsinghua University, Beijing, China

ABSTRACT

Under the constraints of the short time span of past sales data and the high volatility of new product sales, how to accurately predict the ef ect of a new product cigarette launch strategy has become an urgentproblem that needs to be solved for the development of the current tobacco industry. We take 18 months of cigarette sales data in city B of province A as the research sample, take new cigarette Cas the research object, and use the random forest method to fix the errors and missing data. Then, we first use the mature cigarette brands short-term historical sales and multiple labeling systems including the mature cigarette brands historical sales data, retailer sales data, merchant circle crowd portrait data, based on various machine learning method, to predict the mature cigarette sales. We calculate the fittingweights of mature cigarettes to new cigarettes and then simulate and predict the sales trend of newcigarettes. The application ef ect test found the accuracy of new cigarette sales prediction based on thetraditional LSTM model was only 33.31%. In comparison, the prediction accuracy after fitting weights obtained based on the synthetic control method and machine learning could reach 94.17%. We address the limitations encountered in new cigarette sales prediction, fill the research gap in new cigarette launchmodels, and help the tobacco industry develop new cigarette launch strategies.

KEYWORDS

Cigarette sales forecast, Synthetic control method, Machine Learning , Multiple labeling systems

Research on Helmet Recognition Based on Neural Network Model Learning Algorithmt

Shuxuan Feng and Jun Lu, Heilongjiang University, Harbin, China

ABSTRACT

The model based on object detection algorithm shows great advantages in accuracy and precision rate. However, due to the different sizes of the targets to be detected in the images, coupled with the interference of factors such as occlusion and scene complexity, at the same time, the object detection has the problem of too small a percentage, which is prone to miss detection and false detection. Therefore, the performance of the model needs to be further improved. In this paper, based on the shortcomings of existing models, we propose the Invo-YOLOv5s model, which can well improve the accuracy of detection and anti-interference ability. We conducted experiments on the model on the helmet dataset, and the experiments showed that the Invo-YOLOv5s model was selected for training, and the final obtained model detection accuracy reached 94.9%, which is 2.3% higher than the accuracy of the original YOLO model.

KEYWORDS

Object Detection, YOLOv5s, Computer Vision, Attention Mechanism

Growth and Performance of Regional Rural Banks in India

Sakshi Gupta, Assistant Professor, Department of Commerce, Sanatan Dharam Mahila Mahavidyalaya, Hansi (Haryana) India

ABSTRACT

on the Advice of the Narasimhan Working Committee, Regional Rural Banks Were Created in 1975. Rrbs Primary Goal is to Give Rural Credit to the Rural Sector of Society. After Amalgamation, Rrbs Reduced the Total Number to 82 in March 2011. During 2012-2013, 31 Rrbs Merged Into 13 Rrbs, Bringing the Overall Number of Rrbs to 64 in March 2013. However, There Are Currently 43 Rrbs in India. The Central Government Contributes 50% of the Entire Capital of Rrbs, the State Government Contributes 15%, and Sponsor Banks Contribute 35%. This Study Helps in Measuring the Growth and Performance of Rrbs in India. The Study Covers the Years From 2016-17 to 2020-21, and the Statistical Tools Involved in This Work Are Growth Rate, Aagr, and Cagr. The Data for This Research Was Gathered From the Annual Reports of Nabard.

KEYWORDS

Performance, Growth, Average annual growth, Compound annual growth.

E-learning Pedagogical Integration for Multidimensional Educational Transformation

Hagar Gamal Khouder, Department of languages, The American University in Cairo, School of Continuing Education, Cairo, Egypt

ABSTRACT

This paper aims to show how pedagogy influences the learning process, and how the use of ICT tools has a significant positive impact on enhancing the educational environment. ICT tools are required to be used in the modern era because the current generation is tech-savvy and the use of technological devices is indispensable. The use of e-learning became obvious, especially after the pandemic, which prompted extensive research on the proper use of multiple platforms to achieve the interactive learning that educators desired while being apart. Additionally, incorporating AI into the learning process could have a significant impact on improving learning, as well as ask for a multidimensional transformation of both students and educators. However, the use of AI is not entirely independent because students mindsets need to be changed to help them feel like they are in control of their own learning to make them active participants and active citizens.

KEYWORDS

Third-Space Learning, Multidimensional Learning, Transformative Change, Learning Pedagogy, Toolkits.

An Intelligent Program to Monitor 3d Printing and Detect Failures Using Computer Vision and Machine Learning

Christine Li1, Y jia Zhang2, 1Sage Hill School, 20402 Newport Coast Dr, Newport Beach, CA 92657, 2Computer Science Department, California State Polytechnic University, Pomona, CA 91768

ABSTRACT

This paper proposes a novel solution for tracking the 3D printing process using an application that provides users with real-time updates on its progress [1]. The approach involves taking pictures of the 3D printer during the printing process, which are then analyzed by an AI model trained on thousands of labeled images to detect print failures [2]. The system is implemented using a Raspberry Pi and a camera, which capture images of the 3D printer and upload them to an online database [3]. The proposed application accesses this database to keep the user informed of the printers current state, ensuring a seamless printing experience.

KEYWORDS

3D Printing, Machine Learning, Computer Vision, AI.

Detecting Foreign Object Debris in Aircraft Fuel Tanks via Machine Learning and Computer Vision

Taha Shaikh and Patrick Marlow, Bradley Department of Electrical and Computer Engineering, Virginia Tech, Blacksburg, USA

ABSTRACT

Foreign object debris, abbreviated as FOD, is defined as “any object that does not belong in or near airplanes and, as a result, can seriously injure airport or airline personnel and damage airplanes (Bachtel, 2013). FOD costs American airline companies an upwards of $12 billion per year. FOD can be anything from loose screws to gravel and stones, which can cause engine failures, tire damage, and other critical issues. In recent years, there has been a growing interest in using machine learning for FOD detection in aircraft fuel tanks to reduce the risk of FOD-related incidents.The goal of this project is to minimize FOD-related risk by detecting FOD in aircraft fuel tanks during production or maintenance work. Currently, this process is done manually, with workers visually inspecting fuel tank interiors. These fuel tanks are small, dif icult to navigate and are usually dark with several obscured areas. In order to supplement the manual detection of FOD, this project proposes a FOD detection device that captures images of the fuel tank before and after maintenance work takes place. This detection device will consist of a camera that can detect small pieces of FOD undetectable to the human eye so that workers can remove it before the aircraft leaves the factory. This camera will be equipped with digital image processing software that comprises image stitching, alignment, equalization, YOLOv3 image classification, blurring, and edge thresholding. The purpose of this software is to highlight potential FOD in the fuel tank and make it highly visible for workers and management so that it can be safely removed. The algorithm was successful at detecting FOD placed inside a simulated fuel tank; while there were problems with the algorithm misclassifying image perspective dif erences as FOD, it was able to clearly highlight small pieces of FOD such as tissues, washers, bolts, and tiny plastic pieces.

KEYWORDS

Foreign Object Debris, FOD, fuel tank, image classification, object detection, YOLOv3, histograme qualization, dif erence detection, aviation, RANSAC, SIFT, panoramic image stitching.

Gambling Addiction Games: a Mobile Platform for Gambling Psychological Simulation and Research Based on Artificial Intelligence and Machine Learning

Michelle Tay1 and Jack Wagner2, 1Crean Lutheran High School, 12500 Sand Canyon Ave, Irvine, CA 92618 and 2Computer Science Department, California State Polytechnic University, Pomona, CA 91768

ABSTRACT

Gambling addictions are an issue that not only affects someone’s financial stability, but also their relationship with their loved ones [4]. Attempting to quit these addictions cold turkey is hard and not healthy. However, could turning down certain factors of a slot machine reduce the likelihood of a gambling addiction [5]? This paper develops a software to simulate four virtual slot machines, each implementing a specific set of Audio Visual factor’s using the UnityEngine, C# scripts, and Procreate [6]. We applied our application to find out if Visual and Audio cues effect slot machine play time and conducted a qualitative evaluation of the approach [7]. The results show that using this tool we can successfully simulate a live slot machine and easily remove and add things. In this project, we used Unity to create a virtual slot machine to collect data on which slot machine would have more plays by changing their audio and visual cues.

KEYWORDS

Gambling Addiction, Slot Machines, Audiovisual Cues

The Effectiveness of Shielding Methodologies on VLSI Integrations

CO Adeogun, Mountain Top University, Nigeria

ABSTRACT

As the technology advances into deep sub-micron era, crosstalk reduction in VLSI interconnect has become more important for high speed digital circuit design. Shielding is an effective and common technique to deal with signal integrity issues such as crosstalk noise and delay uncertainty. In this survey paper, the basic idea of shielding to reduce capacitive and inductive coupling effect is presented respectively. The effectiveness of shielding is discussed based on simulation results in different shielding cases. Shield insertion algorithms are introduced to minimize the routing area under the given noise specifications. Conclusions are drawn and discussions are made regarding the existing and future work.

KEYWORDS

VLSI Interconnect, Digital Circuit Design, Signal Integrity, Coupling Ef ect, Algorithms.

Developing an Objective Refereeing System for Fencing: Using Pose Estimation Algorithms and Expert Knowledge Systems to Determine Priority and Ensure Fairness

Haokai Zhou1, Aleksandr Smolin2, 1Tarbut V’ Torah Community Day School, 5200 Bonita Canyon Dr, Irvine, CA 92603, 2Computer Science Department, California State Polytechnic University, Pomona, CA 91768

ABSTRACT

Fencers in foil and sabre are often concerned with their referees’ preferences when determining priority, which determines who receives the point in a bout [1]. Oftentimes, humans fail to rationally determine priority and apply the rules fairly, leading to inconsistencies in decisions in the same bout. This often causes heated arguments and much discord in fencing competitions [2]. This paper develops software to identify fencers on a video recording, locate key points in their body’s structure, record their movements and critical metrics about their performance, and match them with an objective expert knowledge system in order to determine who truly has priority at any given time in the match. We tested out several pose estimation algorithms, such as Yolov5, Yolov7, and MediaPipe in order to determine which one has better accuracy and performance in order to be able to deliver precise, unbiased, and fair refereeing decisions in a short period of time, and then allow the referees to reference the logic behind the decision, as well as see all the data that the decision was based upon in order to validate its veracity [3][4]. We also use caching technology to be able to quickly reload and review previous decisions in case any doubt about the bout’s outcome arises post-fact.

KEYWORDS

Python, Yolov7, OpenCV, Fencing.

Data Stream Classification in Dynamic Feature Space Using Feature Mapping

Reza Sajedi and Mohammadreza Razzazi, Department of Computer Engineering, Amirkabir University of Technology, Tehran, Iran

ABSTRACT

Nowadays, stream learning in dynamic feature space is becoming a trending topic. In this problem, it is assumed that each instance of the data stream has different features, and the feature spaces of the classifier and the instances may differ. In this study, we propose a general algorithm for data stream classification in dynamic feature space using feature mapping. In contrast to the other studies, our algorithm is not based on a specific classifier and can cooperate with any classifier best suited for an intended application. It discovers the relationship between the features and estimates the unavailable features previously observed by the classifier. Using this technique leads to exploiting the full potential of the classifier. Furthermore, through empirical experiments and comparison with two recent algorithms, we show that our algorithm has a higher accuracy.

KEYWORDS

Algorithm, Varying Feature Space, Feature Evolution

Heart Disease Prediction Using the Chi-square Test and Linear Regression

Dinesh Kalla and Arvind Chandrasekaran, Department of Computer Science, Colorado Technical University, Colorado, USA

ABSTRACT

Heart disease is the most common disease reported in the United States among both genders. According to official statistics, about fifty percent of Americans suffer cardiovascular disease. This paper performs chi-square tests and linear regression analysis to predict heart disease based on the symptoms like chest pain and dizziness. This paper will help healthcare sectors to provide better assistance for patients suffering from heart disease by predicting it in the beginning stage of the disease. The Chi-square test is conducted to identify whether there is a relation between chest pain and heart disease cases in the United States by analyzing the heart disease dataset from IEEE Data Port. The test results and analysis show that males in the United States are most likely to develop heart disease with the symptoms like chest pain, dizziness, shortness of breath, fatigue, and nausea. This test also shows that a weak correlation of 0.5 is identified, which shows that people of all ages, including teens, can face heart disease, and its prevalence increase with age. Also, the tests indicate that 90 percent of the participant facing severe chest pain suffers from heart disease, where the majority of the successful heart disease identified is in males, and only 10 percent of participants are identified as healthy. The evaluated p-values are much more significant than the statistical threshold of 0.05, which concludes that factors like sex, Exercise angina, Cholesterol, old peak, ST_Slope, obesity, and blood sugar play a significant role in the onset of cardiovascular disease.

KEYWORDS

Chi-Square Test, R; Data Mining; Big Data; Linear Regression Analysis; Heart Disease; Risk Factor; Machine Learning; Cardiovascular Disease.

Lightweight Unsupervisedgraphembedding Method via Dfs Basedn-edgesubgraphs

Mengmeng Jia1 and Hao Feng2, 1Key Laboratory of Cyberspace Situation Awareness of Henan, Zhengzhou University, Henan, China, 2Key Laboratory of Cyberspace Situation Awareness of Henan, Henan, China

ABSTRACT

The graph embedding method can transform the graph structure into a vector so that the similarity between the graphs can be measured quickly and conveniently. The existing graph embedding methods mainly call the neural network model to obtain a vector representation of the graph based on the rootedsubgraphs or frequent subgraphs. These methods have the following problems: Firstly, some algorithms are complex in sub-graph extraction, and their versatility is not high; secondly, whether rootedsubgraphs or frequent subgraphs, they are only a local part of all the subgraphs in a graph, the final embeddings cannot represent the entire original graph, which leads to poor classification ability for graphs. To further improve the learning ability of the graph embedding method, we propose asubstructure2vec (substructure to graph vector) method via the depth-first search based N-edge subgraphs. The method prevents the high time complexity of subgraph traversal by limiting the maximumnumber of edges N of the extracted subgraphs. Furthermore, the subgraphs can be extracted more comprehensively by performing depth-first search based on N-edge subgraphs extraction over the entire graph. Our method has the following advantages compared with the current graph embedding methods: lower time complexity, unsupervised learning, and learning the whole graph embeddings. Experimental comparison of classification tasks on multiple datasets shows that our approach increases the classification accuracy rate by 2%-100% on most of the datasets.

KEYWORDS

Graph embedding, Sub-graph extraction, Depth-first search, Unsupervised learning.

A Novel Exploit Traffic Traceback Method Based on Session Relationship

Yajing Liu, Ruijie Cai, Xiaokang Yin, and Shengli Liu, State Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou 450001, China.

ABSTRACT

Vulnerability exploitation is the key to obtaining the control authority of the system, posing a significant threat to network security. Therefore, it is necessary to discover exploitation from traffic. The current methods usually only target a single stage with an incomplete causal relationship and depend on the payload content, causing attacker easily avoids detection by encrypting traffic and other means. To solve the above problems, we propose a traffic traceback method of vulnerability exploitation based on session relation. First, we construct the session relationship model using the session correlation of different stages during the exploit. Second, we build a session diagram based on historical traffic. Finally, we traverse the session diagram to find the traffic conforming to the session relationship model. Compared with Blatta, a method detecting early exploit traffic with RNN, the detection rate of our method is increased by 50%, independent of traffic encryption methods.

KEYWORDS

Exploit, Malicious Traffic Detection, Session Relationship, Traffic Analysis.