June 17 ~ 18, 2023, Sydney, Australia
Frederick Abban, Dr. Koudjo M. KoumadiDepartment of Computer Engineering, University of Ghana, Legon, Ghana
The universal exponential increase in the demand for high data rate for mobile devices has propelled lots of research in wireless communication. Deploying and implementing a wireless network in a particular geographical area requires proper planning since all existing propagation models are not “one size fit all” models. In this study, path losses of seven empirical propagation model were simulated and compared with results of measurements of received signal strength in Non-Line of Sight (NLOS) scenario for Accra, Ghana on 2300 MHz. The study terrain is similar to most cities on the coast of West Africa. Correction factors were computed and applied to original propagation model equations and the Ericsson 9999 model showed the best fit to the measurement data, thus it can be used to predict received signal strength for Accra and other environments with similar terrains.
4G, Propagation Models, Received Signal Strength, LTE, Path Loss, NLOS.
Shuaiqian Yuan1, Xiangdong Jia1,2, Tongjian Shang1, Yangyang Sun1, 1College of Computer Science and Engineering, Northwest Normal University, Lanzhou, China, 2Wireless Communication Key Lab of Jiangsu Province, Nanjing University of Posts and Telecommunications, Nanjing, China
In order to solve the problem of limited channel resources of the Internet of Things (IoT) and improve the information timeliness of IoT system, a multi-access Cognitive Radio (CR) IoT system model was considered. The model included one Primary User (PU) and two Secondary User (SU) nodes. PU had spectrum access rights, and two SU could share the PU spectrum. Both SUs had Age of Information (AoI) oriented data streams. The difference was that the first user cannot control the generation of status updates and can only generate status updates according to a certain probability, while the second user generated status updates according to any generation strategy. Receiver adopted multiple antennas and transmitter adopted single antenna. Under the constraint of the main user, the average age of information of the First Secondary User node in the First Come First Served (FCFS), Last Come Last Served (LCLS) and the packet dropping queue was analyzed respectively. Then the average age of information of the second user node in the threshold strategy was derived. Finally, an optimization problem was proposed to minimize the average age of information of the first user and lower the average age of information of the second user than the given threshold. The constraint condition of the problem was convex, but the objective function was non-convex, so a suboptimal technique was introduced and the optimal solution was obtained by double convex optimization algorithm. Simulation results show the performance of the proposed algorithm under different system parameters.
Age of Information, Cognitive radio, Double convex optimization, Multiple access channel, Threshold strategy.
Seyyed Rohollah Mirhoseini1, ehrouz Minaei-Bidgoli 2, Rahil Hosseini3 and Bahareh Shaker-Ardakani1, 1Department of Computer Engineering Islamic Azad University North Tehran branch Tehran, Iran, 2Department of Computer Engineering Iran University of Science and Technology Tehran, Iran, 3Department of Computer Engineering, Shahr-e-Qods Branch, Islamic Azad University, Tehran, Iran
Today there are more than 4.39 billion internet users that almost 70 percent of them use social media on mobile devices. Network security is one of the most important aspects to consider while working over the internet, LAN or other networks, no matter how small or big your business is. We know many zero day attacks are continuously emerging because of the addition of various protocols mainly from Internet of Things (IoT). previously known attacks in cyberattacks can detect by using Artificial Intelligence (AI) solutions such as Neural Networks, Machine Learning (ML), Support Vector Machine (SVM), decision tree, Hidden Markov Model (HMM), Hierarchical Clustering, Game Theory (GT), and Natural Language Processing (NLP). But advanced mechanisms of AI are not able to detect all of attacks. On the other hand Deep Learning (DL) techniques which are capable of providing embedded intelligence in the IoT devices and networks, are emerged to cope with different security problems. Then we used Convolutional Neural Network (CNN) and Genetic Algorithm.
Network security, Convolutional Neural Network (CNN), Deep Learning and Genetic Algorithm.
Chaitanyateja Thotadi1, Monith Debbalav1, Subba Rao1, Alavalapati Goutham Reddy2*, Basker Palaniswamy3, Vanga Odelu1, 1National Institute of Technology, Andhra Pradesh, India, 2Fontbonne University, St. Louis, USA, 3Queensland University of Technology, Brisbane, Australia, 4Indian Institute of Infromation Technology Sri City, Chittoor, India
Social network providers offer a variety of entertainment services in exchange for end users’ personal information, such as their identity. The majority of users access social networking sites via their smartphones, which they utilize in conjunction with a traditional authenticator like a password. On the other hand, aggregators, which pull content from multiple social networks, are often used to get into smartphone apps that may involve mobile ticketing, identification, and access control. They are a potential target for malware and spyware injections due to their powerful position. Malware is capable of circumventing authentication mechanisms in order to get access to social networking services, which may result in stealing the personal information of users. To deflect any type of attack from malicious software, BrightPass , a malware-resistant method based on screen brightness, was introduced. Conversely, we have demonstrated that the BrightPass user’s personally identifiable information, such as PIN numbers, may be recovered by evaluating the variations between the recorded input from many authentication sessions. We have then offered various enhanced BrightPass versions to address the observed vulnerability. Our enhanced BrightPass versions are both simple and secure to use when it comes to accessing social networks via mobiles.
Smartphones, Social Networks, Malware, Authentication, Security.
Fatai Jimoh1, Mo Saraee2, Azar Shahgholian3, 1,2School of Engineering Science and Humanity, University of Salford, United Kingdom, 3Liverpool Business School, Liverpool John Moore University, Liverpool, United Kingdom
The conventional data mining approach to crime prediction models often depends on historical data. However, considering the current global crime trend where offenders frequently register their criminal intent on social media and also invite others to witness and/or participate in various crimes, there is a need for an alternative and more dynamic approach.This paper, therefore, applied an ensemble of machine learning algorithms to reducethe crime rate in Manchester cities by combining historical crime data with tweet data. To overcome the problem with data qualities and bias in our prediction we ensured that all the features used in this work were from the same year and same month. The ensembled method showed the highest performance in predicting different categories of crimes with the highest accuracy when compared with base models and our study also revealed the contribution of sentiment score to the overall performance of the model. Finally, we conclude that social media data if properly mined would contribute to an improved prediction on the likelihood of crimes occurrence as well as their prediction.
Data Mining, Crime Prediction, Machine learning, Sentiment Analysis, Stacked Ensemble.
Seyed Mohammad Hossein Abedi Nejad, Mohammad Amin Behzadi and Abdolrahim Taheri, Iran
Overloading in DC servo motors is a major concern in industries, as many companies face the problem of finding expert operators, and also human monitoring may not be an effective solution. Therefore, this paper proposed an embedded Artificial intelligence (AI) approach using a Convolutional Neural Network (CNN) using a new transformation to extract faults from real-time input signals without human interference. Our main purpose is to extract as many as possible features from the input signal to achieve a relaxed dataset that results in an effective but compact network to provide real-time fault detection even in a low-memory microcontroller. Besides, fault detection method a synchronous dualmotor system is also proposed to take action in faulty events. To fulfill this intention, a one-dimensional input signal from the output current of each DC servo motor is monitored and transformed into a 3d stack of data and then the CNN is implemented into the processor to detect any fault corresponding to overloading, finally experimental setup results in 99.9997% accuracy during testing for a model with nearly 8000 parameters. In addition, the proposed dual-motor system could achieve overload reduction and provide a fault-tolerant system and it is shown that this system also takes advantage of less energy consumption.
Embedded AI, CNN, signal transformation, real-time fault-detection, dual-motor fault-tolerance.
Saachin Bhatt, Mustansar Ghazanfar and Mohammad Hossein Amirhosseini, University of East London, London, E16 2RD, United Kingdom
The purpose of this research is to investigate the impact of social media sentiments on predicting the Btcoin price using machine learning models, with a focus on integrating on-chain data and employing a Multi Modal Fusion Model. For conducting the experiments, the crypto market data, on-chain data, and corresponding social media data (Twitter) has been collected from 2014 to 2022 containing over 2000 samples. We trained various models over historical data including K-Nearest Neighbors, Logistic Regression, Gaussian Naive Bayes, Support Vector Machine, Extreme Gradient Boosting and a Multi Modal Fusion. Next, we added Twitter sentiment data to the models, using the Twitter-roBERTa and VADAR models to analyse the sentiments expressed in social media about Bitcoin. We then compared the performance of these models with and without the Twitter sentiment data and found that the inclusion of sentiment feature resulted in consistently better performance, with Twitter-RoBERTa-based sentiment giving an average F1 scores of 0.79. The best performing model was an optimised Multi Modal Fusion classifier using TwitterRoBERTa based sentiment, producing an F1 score of 0.85. This study represents a significant contribution to the field of financial forecasting by demonstrating the potential of social media sentiment analysis, onchain data integration, and the application of a Multi Modal Fusion model to improve the accuracy and robustness of machine learning models for predicting market trends, providing a valuable tool for investors, brokers, and traders seeking to make informed decisions.
Cryptocurrency, Bitcoin Price, Social Media, Sentiment Analysis, Machine Learning, K-Nearest Neighbors, Logistic regression, Gaussian Naive Bayes, Support Vector Machine, Extreme Gradient Boosting, Multi Modal Fusion.
Neel Chitre, Manjusha Tatiya, Anmol Dhage and Neha Sharma, Department of Computer Engineering, Indira College of Engineering and Management, SPPU, India
Prescription reading is an important task in the healthcare industry, as it helps to ensure that patients receive the correct medications and dosages. However, manual prescription reading can be time-consuming and error-prone, leading to potential harm for patients. Machine learning has the potential to automate this task, improving efficiency and accuracy. In this paper, we review the state-of-the-art in prescription reading using machine learning techniques, including support vector machines, deep learning, and recurrent neural networks.We also propose a novel approach using convolutional neural networks for handwritten prescription recognition.
Prescription, Deep Learning, CNN, SVM, RNN.
Huakun Hou, Xiaoke Deng, Zekun Lu & Linbo Zhai, Department of Information Engineering, Shandong Normal University, China
Today, educational prediction analysis has become an important tool for educational institutions to analyze students, and the performance of students in educational prediction analysis is a critical feedback. However, there are serious challenges in dealing with multi-factor datasets to improve the convergence and accuracy of predicting student performance. Therefore, this article comprehensively analyzes machine learning technologies, analyzes the community activities of middle-term students, and explores the impact of community activities on middle-aged students practical skills. This paper uses Random Forest and Pearson Correlation Coefficient to analyze the relevance of the impact of community activities on student hands-on performance and use Teaching and Learning Algorithms (TLBO) to optimize Back propagation (BP) neural networks applied to predict students future trends. The results show that the TLBO-BP model can more accurately predict dynamic changes in student performance and predict high accuracy in simple patterns.
Grade prediction, Feature analysis, Teaching and learning optimization algorithms, BP neural network.
James Wang1, Ang Li2, 1Dougherty Valley High School, 10550 Albion Rd, San Ramon, CA 94582, 2Computer Science Department, California State Polytechnic University, Pomona, CA 91768
This paper addresses the problem of predicting future medication models in a database using a powerful AI system . The background highlights the importance of accurate predictions in healthcare for effective decision-making and improved patient outcomes. The proposed solution involves the development of an AI model trained on a diverse dataset of historical medication models, incorporating advanced machine learning algorithms and techniques. The key technologies and components of the program include data preprocessing, feature selection, algorithm comparison, and performance evaluation . Challenges encountered during the project, such as data quality and model generalizability, were mitigated through careful data cleaning and fine-tuning of the AI model . The application of the system to various scenarios during experimentation demonstrated its robustness and versatility. The most important results include high prediction accuracy, precision, and recall in forecasting future medication models. The system showed promising performance across different patient populations, suggesting its potential for personalized treatment planning and decision-making. The idea presented in this paper offers a valuable solution for healthcare professionals and researchers seeking accurate predictions in medication modeling, facilitating better patient care and optimized treatment strategies.
Medication modeling, AI system, Prediction accuracy.
Olta Llaha and Azir Aliu, Faculty of Contemporary Sciences and Technologies, South East European University, Tetovo, North Macedonia
The objective of this paper is to present a practical case illustrating the practicality of data visualization techniques and machine learning algorithms. Through the assessment of various algorithms, we aim to examine data and forecast the influence of data visualization on decision-making. Our findings offer substantiation in favor of the assertion that data visualization indeed impacts decision-making. Furthermore, we explore the ramifications of employing data visualization technology within the academic community, encompassing both faculty and students. Additionally, we assess the impact of data visualization on decision-making in higher education institutions, underscoring its capacity to augment the decision-making efficiency and promptness of stakeholders.
Higher Education, Machine Learning, Data Visualization.
Xiaoke Deng, Huakun Hou, Meiyu Jin and Linbo Zhai, School of Information Science and Engineering, Shandong Normal University, Jinan, China
With the rapid development of information technology and its penetration into the field of education, it has become a worthwhile research topic to explore how to combine OBE education models with data mining techniques to obtain predictions of secondary school students employment rates. Therefore, this paper first collected attribution data affecting the employment rate of secondary school students, and then used three models to predict the highly correlated attribution data. Finally, based on the analysis and prediction results, the OBE education model was implemented for the corresponding classes. The experimental results found that the combination of data mining techniques and educational practices facilitated the development of secondary education and significantly increased the employment rate of secondary school students.
Education data mining, OBEmodel, Student employment prediction.
Abhinav Gullapalli and Neil Patel, College of Computing, Georgia Institute of Technology, Atlanta, USA
The rapid development of Large Language Models (LLMs) over the 21st century has sparked curiosity in diverse opportunities to apply deep learning with natural language. Our work evaluates the application of LLMs to diagnose patients in the clinical setting. Specifically, we apply OpenAIs GPT-3.5-Turbo1 with zeroshot learning and fine-tune Big Bird2 and OpenAIs GPT-3-Turbo1 on an open-source French dataset3 encompassing 49 pathologies with various related questions and answers which we translate to English. We further extend the French dataset with manually labelled ICD-10 codes for a standardized, languageagnostic representation of each pathology. Model performance for diagnosing patients is evaluated on three increasingly difficult tasks to identify partial representations and the complete representation of each pathologys respective ICD-10 code. We find that Few-Shot GPT-3-Davinci performs best on the primary task of interest, namely identifying the complete ICD-10 code for patient diagnosis, which is inherently the hardest of the three tasks evaluated in this work. The development of future DrGPTs, while exciting and potentially useful, should be used responsibly while considering the ethical implications of diagnosing and treating human patients through LLMs.
large-language models, prompt engineering, fine-tuning.
Venkat Amith Woonna and Rahul Sandireddy, Department of Computer Science, Vellore Institute of Technology, Chennai, India
The process of monitoring the price of a product and receiving alerts if the price drops involves entering the URL of the product, setting a threshold price, and receiving an alert email if the price drops below the set threshold value. This process can help users to monitor the price of a product and potentially save money if the product goes on sale or experiences a price drop. Price tracking tools that use data extraction systems such as this can be very useful for users who want to monitor the price of a particular product and get alerted when the price drops to a certain level. The system being described is a price tracking tool that extracts data from e-commerce websites based on the user’s URL input, specifically for Flipkart and Amazon. This system employs the ETL (Extract, Transform, Load) process, where the first step is to extract the data, followed by transforming the data, and then loading the data into the system. Once the system has collected the price data, it sends an email alert to the user with the lowest available price. The model periodically checks the price of products for every 4 hours, which can be set to any duration by the user on e-commerce websites, extract data on price history, and provides insights into the opinions of other users about the product. It is worth noting that data extraction from websites without permission is not legal and can result in legal action against the perpetrator. However, many e-commerce websites offer APIs or other methods for authorized data extraction that can be used to create price-tracking tools or other applications. Additionally, it is important to consider the ethical implications of using such systems, particularly in terms of potential harm to small businesses and consumers. Price tracking tools can be useful for users who want to make informed decisions about purchasing products at the optimal price point or understand how other users feel about the product. The system described can help users monitor the price of a particular product and get alerted when the price drops to a certain level, and can also provide insights into the product’s pricing history and patterns. However, it is important to use such systems ethically and legally and to consider the potential harm to small businesses and consumers.
Price monitoring , Price alerts , Price tracking tool, Data extraction, E-commerce.
Ariestelo A. Asilo1,2, Czarina Anne A. Villareiz1, Arnan B. Araza3, Jose Arnold S. Bagabaldo4, and Francis G. Balazon5, 1Varacco Inc & 2ThinnkFarm, 3Wageningen University and Research, 4Packetworx, 5Batangas State University, The National Engineering University Lipa Campus
This study focuses on the use of Internet of Things (IoT) devices for monitoring microclimate conditions in coffee production, specifically in the highland and lowland areas of the Philippines. The aim is to assess the effectiveness and accuracy of IoT-based data in decision-making for farm interventions, compared to data from the well-known weather prediction app AccuWeather. This comparison is important in developing prediction models useful for upscaling the study to other farms. The results show that there are systematic differences in the temperature, relative humidity, and amount of rainfall data collected by the two methods, indicating the importance of on-site IoT devices for localized microclimate monitoring. However, the study also highlights some challenges in the implementation of IoT, such as malfunctions and hardware issues. This research underscores the need for further testing and calibration of IoT devices in coffee production, as well as the establishment of a localized Good Agricultural Practices (GAP) reinforced by IoT-based microclimate conditions.
Coffee, IoT, Microclimate, Wireless Sensor.
Josaphat Rutaganda1, Mariantonietta Fiore2, Nino Adamashvili3, 1Department of Agronomy, Institut Polytechnique UniLaSalle, Rouen, France, 2, 3Department of Economics, the University of Foggia, Foggia, Italy
The paper presents a systematic literature reviewthat investigates the potential of integrating the Internet of Things (IoT), Artificial Intelligence (AI), and Blockchain technology (BCT) to improve the safety and sustainability of the wine supply chain. The study identifies the challenges and opportunities of these technologies and their potential for addressing issues such as food safety, quality control, and traceability in the wine industry. Additionally, the framework is proposed by authors in order to overcome the challenges associated with the implementation of these technologies in the wine supply chain. The findings suggest that the integration of IoT, AI, and BCT can lead to increased transparency, efficiency, and security in the wine production chain.The study concludes that the integration of these technologies can contribute to a more sustainable and safer wine supply chain.
BCT, IoT, AI, wine, supply chain.
Lakmali Karunarathne1, Swathi Ganesan2 and Dr Nalinda Somasiri3, 1Academic Associate, Department of Computer Science, York St John University, London EC1A 4JT, UK, 2Lecturer, Department of Computer Science, York St John University, London EC1A 4JT, UK, 3Head of Programme, Department of Computer Science, York St John University, London EC1A 4JT, UK
As a result of the development in computing technologies have begun to believe the human expectations on these needs in the different sort of components. The eSand Transport System with IOT (eSTSI) is a sand transport system designed to provide secure and accurate data such as gross weight with the sand and the truck, viewing the details of the owner when the RFID card is detected, sending alerts through the mobile application from the Firebase by interconnecting with the IOT device, viewing the schedule of the selected truck with the date and destination, and displaying the location once the truck is passed the checkpoint. The main functionalities of eSTSI are to identify the truck with the correct information via the RFID card that retrieves the data who has enrolled with the app and stores the data in the firebase. The expected services are aimed to provide by this system.
RFID, IOT, Mobile Application, Firebase, Vehicle, recognition, system, public.
Michelle H. Benavides, Mariam C. Salcedo, Estefanny Y. Cáceres, Ivan E.Mallque and Diana Chipana Continental University, Huancayo 12001, Peru
This research work has been carried out for the criminology unit of the National Police of Peru in Huancayo city in the year 2021, because it has a deficient people detection process. This work is based on stating that the proposed implementation of a tracking system based on facial recognition will improve the ef iciency of the PNPs people detection process. The methodology is engineering - cascade model, that includes four stages: requirements analysis, design, construction and testing. Regarding the approach, this is quantitative, since it measures the investigation dimensions: compliance of police personnel, degree of similarity and police relationship with the community. In addition, the population is the total of people who reside in Huancayo city, who are 18 years or older, and the sample was non-probabilistic, taking 15 participants who are of legal age, who reside in Huancayo city. Based on all the work carried out, the results of the investigation showed an increase in the average percentage of the degree of similarity, from 32% to 89.49%; in the same way, it was possible to show that there was an increase in the average percentage of the police relationship with the community, from 44% to 93%. According to everything detailed, it is confirmed that the configuration of the implementation proposal of a tracking system based on facial recognition ef iciently influences the improvement of the people detection process for the criminology unit of the PNP of Huancayo city, Peru in the year 2021.
Tracking, Facial recognition, Degree of similarity, Police relationship with the community.
Prasad Samudrala, Justin Jose, and Amit Kulkarni, Ph.D, Honeywell Building Technologies,Advanced Technology Group, Honeywell Inc
Trap antenna is well known method and has many applications. With this method, trap(s) are used on antenna to block currents of some frequencies and so electrically divide the antenna into multiple segments and thus one antenna can work on multiple frequencies. In this paper, trap antenna method is used to design a dual band Sub GHz printed F-antenna. The antenna is printed on FR4 board to achieve low cost solution. The two bands are 865-870 MHz and 902-928 MHz. The challenge of this design is that the frequency separation of the two bands is very small. In this case,and also the extra section for low frequency band is too small. Then, the influence of trap LC component variation due to tolerance to the two resonant frequencies is big, and so it is difficult to achieve good in band return loss within the LC tolerance. This is the main difficulty of this design. The problem is solved by placing the low band section away from the end of the antenna.
Trap , SubGHz , printed F antenna, FR4 board , LC Component.
ADDA BOUALEM, DJAHIDA TAIBI AND AROUA AMMAR, Department of Science Computing, Ibn Khaldoun University, Tiaret, Algeria.
This paper Various studies cited in the literature deal with the classic problem of obstacle coverage, where the deployment environment, sensor nodes, and base stations have characteristics that are considered perfect but suffer from various flaws in the real world. This paper presents other barrier coverage types ranked in a new classification based on linear and nonlinear barrier coverage according to deterministic and insecure environments, and enumerates some of the different current and future challenges of these coverage types and connectivity in WSNs.
WSN, Barrier Coverage, Deterministic and Uncertain Linear Barrier Coverage, Deterministic and Uncertain non-linear Barrier coverage, Connectivity, Current and Future Challenges.
Nashwan Ghaleb Al-Thobhani1, 2 and Jamil Sultan 1, 2, 1Computer Network Engineering, Technology Department, Sana’a Community College, Sana’a, Yemen, 2Faculty of Engineering & IT, University of Modern Sciences, Sana’a, Yemen
Wireless Sensor Networks (WSNs) is one of network technology that have revolutionized the world of technology and to explain the importance of the networking major in the different majors, in this project we choose one of the important major in the life is the e-healthcare major. In thisresearch explain how can make system emergency to the elderly people and heart disease through collaborating between the WSNs technology and internet of things ( IOT )devices to make the emergency system use the heart sensor and IOT devices to send Short Message Services (SMS) connected with Global Position System (GPS) to determine the location of patient by click the link with availability Wireless Fidelity (Wi-Fi) or Third Generation (3G) to easily arrive for saving patients life in the short time and surest way. This technology is not used only in the healthcare major, it also used in different majors likes sports, entertainments, martially etc .
WBANs; IOT; GPS; diabetes;management; healthcare;Arduino; Wi-Fi; sensors; Wireless Sensor Networks(WSNs).
Zhongshun Rao1, Victor Phan2, 1Tarbut VTorah Community Day School, 5 Federation Way, Irvine, CA92603, 2Computer Science Department, California State Polytechnic University, Pomona, CA91768
This project explored a solution to help with club management . Club management has been a complex task formany people . The solution to this problem is to create a centralized management application that allows all management and communication to be done on it . The application is made with FlutterFlowand googlefirestore. It consists of two main parts, the communication and the management. Possible challenges are the users’ experiences in features like chat and task management, but these will be improved in the future through updates. The application can be applied to all kinds of clubs like sport club or school clubs, providing communicationandmanagement for members. It is hoped that this application could provide users a convenient and ef ective waytoorganize their clubs.
club management, organizational ef iciency, comprehensive platform, Mobile applications.
Chunwei Shi1, Andrew Park2, 1Northwood High School, 4515 Portola Pkwy, Irvine, CA 92620, 2Computer Science Department, California State Polytechnic University, Pomona, CA91768
There is an increasing need to provide incentives for people to create healthier habits as the spread of COVID-19caused many people since then to lead a more sedentary lifestyle. To get more people out and active, Runspirationseeks to incentivize people to move more through a gamified system that rewards people the more they move; userswill be required to move at least 2km or more with the amount they can earn increasing the farther they are abletorun. The application itself is created using a combination of Flutter and Firebase to handle and maintainuserprogress as they unlock and earn more coins and achievements to showcase to others. The use of randomness, asseen in the spinning wheel, also helps to increase the motivation of the user to do fitness. The primary challenge wehad to overcome was the lack of a solid tool within Flutter for accumulating and using health data in a timelymanner, requiring us to explore novel ways to access that information to use with the rest of the app. Theapplication also used the health API from Apple to monitor the user’s fitness status.
Flutter, HealthKit, Firebase, Running.
Alessandro Rizzo, Independent Researcher, Brescia, Italy
TIn this paper, we present the Energy-Impulse-Information Tensor (EIIT), a unified framework for studying the interplay between energy, impulse, and information in physical systems. The EIIT combines classical concepts of energy and impulse with the more recent concept of information, allowing for a more comprehensive understanding of physical phenomena. We provide a detailed mathematical definition of the EIIT and demonstrate its applicability in several areas including fluid dynamics, quantum mechanics and cosmology. Our results show that the EIIT has potential to be a powerful tool for predicting behavior of physical systems and we encourage further research into its applications. By incorporating principles of quantum mechanics into this framework it allows for analysis and manipulation of complex systems through study of their energy, momentum and information properties. We demonstrate potential applications of EIIT in addressing fundamental problems in computer science, physics and mathematics. A formal mathematical definition of EIIT is presented along with its symmetries and properties. This framework has been applied to study four-color theorem with connections to golden ratio explored. Additionally we have investigated relationship between equation of continuity and EIIT framework emphasizing role determinant energy components play in ensuring stability and continuity in system. By framing the P vs NP problem in terms of efficient energy allocation using the EIIT tensor, we have shown that finding an efficient algorithm to solve an NP problem is equivalent to finding an efficient way to allocate energy in a system described by the EIIT tensor. We have proposed a modified version of the Karmarkar-Karp algorithm for solving the subset sum problem using the EIIT framework. However, it is important to note that the P vs NP problem remains unsolved as the conjecture of existence of an algorithm satisfying the equation of continuity with det E i = 0 for all NP problems has not been proven.
Energy-Impulse-Information Tensor (EIIT), Quantum Mechanics, General Relativity, P vs NP, Golden Section Method, Equation of Continuity, Energy Allocation, Unification, Subset Sum Problem, Karmarkar-Karp Algorithm.
Marisha Parikh, Kishori Telvekar, Department of Information Technology, Garware Institute of Career Education and Development, University of Mumbai, India.
The conventional shopping trolley system, once a cornerstone of retail shopping, has become archaic in the rapidly evolving landscape of technology-driven innovation. The demand for swift, ef icient, and cost-ef ective shopping experiences has culminated in the development of an automated self-checkout trolley system. This study underscores the pressing need to transform the traditional shopping trolley system into an automated self-checkout trolley system to enhance customer satisfaction. The proposed smart trolley system leverages cutting-edge computer vision technology to automatically identify products as customers place them in the cart, while also handling digital payment options, expediting the checkout process without the need for protracted queues at checkout counters. The innovative system of ers a centralised invoicing system and places a premium on customer time, encouraging customers to shop for even fewer items at the most competitive prices. The metamorphosis of the traditional shopping trolley system into an automated self-checkout trolley system has the potential to upend the retail industry, delivering a more streamlined and lucrative shopping experience for customers while also amplifying retailer profitability. The research provides a lucid understanding of the benefits and significance of implementing this pioneering system to shape the future of retail.
automated self-checkout trolley, computer vision, new generation, supermarkets, Raspberry Pi. customer satisfaction, ef iciency, productivity, investment costs, technology and retail industry.
Omar Hujran, Department of Analytics in the Digital Era, United Arab Emirates University, Al-Ain, United Arab Emirates.
Semiotic analysis serves as a tool for comprehending the ways in which individuals interpret and comprehend their surroundings by scrutinizing the interplay between signs, symbols, and their connotations. The efficacy of digital business platforms is heavily reliant on the creation of well-crafted, meticulously structured websites that facilitate effective communication with targeted customers. E-commerce websites extensively employ images and texts to enhance product awareness and cultural significance. Scholars contend that the utilization of e-commerce websites will expand if designers integrate the cultural norms of different user subsets. Thus, this study aimed to examine this issue through a literature review of semiotic analysis applied to e-commerce websites.
Semiotic analysis, Website design, E-commerce, Culture, Literature review.
Bejjam Vaila1 and Dr.S.Sudhakar Ilango2, 1School of Computer Science and Engineering VIT-AP University, Amaravathi, India, 2School of Computer Science and Engineering VIT-AP University, Amaravathi, India.
Climate Change and Global Warming are two important factors that are causing major disasters all over the globe. Especially sea-level rise due to rapid glacier melting causing severe coastal erosion in recent times. According to the recent scientific reports regarding coastal erosion is the global Land Lost (LL) is 27,860 km 2 and Land Gained (LG) is 13,810 km 2 during 1984-2015. As per NASA global sea-level report the sea-level raise is increasing very rapidly, where sea levels have risen on average 1.6 millimeters (0.063 inches) per year between 1900 and 2018. The study is an attempt to examine the problem of shoreline change in Kakinada with view to understand the causes of erosion and accretion. The present study aims to detect the Kakinada coastline changes in the past five years 1989,2017,2018,2019,2020 and 2021 satellite images using remote sensing data.
Satellite imagery; DSAS; FCC.
Long Chen1, Andrew Park2, 1Northwood High school, 4515 Portola Pkwy, Irvine, CA 92620, 2Computer Science Department, California State Polytechnic University, Pomona, CA 91768
In light of the recent pandemic and the increasingly online lifestyles of general people, the need to maintain an active lifestyle is becoming more important to consider as “more than a quarter of all adults [are] not getting enough physical activity. This puts more than 1•4 billion adults at risk of developing or exacerbating diseases linked to inactivity” . HealthPet seeks to help tackle the issue by providing a place where users can earn and interact with virtual pets based on how active they are throughout the day. By using Unity and Firebase alongside native code, we are able to put together a package that provides a goal and incentive while also being able to preserve and maintain a users progress on their path toward a healthier day.
Apple Watch Fitness, Pets, Exercise, Motivation.
Zihao Lin1, Jonathan Sahagun2, 1Lower Merion High School, 315 E Montgomery Ave, Ardmore, PA 19003, 2Computer Science Department, California State Polytechnic University, Pomona, CA 91768
Serious games are video games designed for more than just pure entertainment purposes . Serious games developers combine traditional game mechanics and the ideas to educate, inform and facilitate social change . These games can be used in many occasions, such as education, healthcare and more. Serious games use simulations and scenarios to provide an immersive and interactive learning experience. They of er an environment to experiment with a variety of solutions to real-world problems, promoting critical thinking and decision making skills. These games can also improve knowledge retention, motivation, and engagement, as they provide instant feedback, rewards, and challenges. This application is like one of the many serious games, it provides a simulation of a highway, its primary purpose is to help to train juvenile’s knowledge on driving and logical thinking, and relax during the playthrough .
Educational, Serious game, Coding, Drivin.
Abel Varghese, Mahendher Marri and Sibi Chacko, School of Engineering and Physical Sciences, Heriot-Watt University, Edinburgh EH14 4AS, UK
The Philippines is known to be a country that values the agricultural sector. Agriculture is the backbone of the Philippine economy, contributing around 9% to its gross domestic product (GDP) and providing livelihood to millions of Filipinos. Local vegetables such as pechay, mustasa, sitaw, talong, and ampalaya are some of these essential agricultural crops, used in different famous dishes in the country. The emergence of technology helps individual and community improve their way of administering and managing crops,which is why it is very important to develop an innovative way to produce sustainable vegetable crops. This paper presents the development of an application that will classify the diseases, pests, and deficiencies of vegetable crops. This application usesdifferent Convolutional Neural Networks architectures such as ResNet, YOLO, and Faster R-CNN to dissect information from digital photographs. This application provides different information on diseases, pests, and deficiencies, and information on how to manage and administer the crops for sustainable production. The mobile application helps many vegetable growers identifythe problems and challenges of their crops. The used CNN architecture provides accurate detection, analysis, and interpretation of the content of digital photographs, and served as a way to provide information on the solutions. ResNet architecture provides a high accuracy rate among YOLO and Faster R-CNN in the detection and classification of diagnosis.
Local Agricultural Vegetables, Diagnosis, Deep Neural Networks, Classification ResNet, YOLO, Faster R-CNN .
Loyd S Echalar, Dr. Arnel C. Fajardo, Technological Institute of the Philippine –Quiapo Manila, Philippines
In this paper, a study of autonomous vehicles in MATLAB/Simulink® 2022 is carried out and is conducted using vehicles with three different speeds 40, 80, and 120 km/hr. A normal highway in UAE is considered for road modelling. All vehicles modelled are representative of that available in the UAE. In the model, lane following and lane-keeping assistance functions and Simulink block which are described using artificial neural networks are selected. Simulation is validated with existing published results of physical vehicle models. In the simulations, it is assumed that vehicles have minimal steering angles as the system is in an autonomous collision free environment, selected from MATLAB. Results are obtained as velocities, accelerations, and safe distance with respect to the preceding vehicle. The following results are critically analysed and validated.
Artificial Intelligence, Artificial Neural Networks, MATLAB, Autonomous Vehicle, Deep Neural Networks, Ackermann.
Xin Chen, Alex Reibman, Sanjay Arora, Ernst & Young LLP. US.
Timeliness and contextual accuracy of recommendations are increasingly important when delivering contemporary digital marketing experiences. Conventional recommender systems (RS) suggest relevant but time-invariant items to users by accounting for their past purchases. These recommendations only map to customers’ general preferences rather than a customer’s specific needs immediately preceding a purchase. In contrast, RSs that consider the order of transactions, purchases, or experiences to measure evolving preferences can offer more salient and effective recommendations to customers: Sequential RSs not only benefit from a better behavioral understanding of a user’s current needs but also better predictive power. In this paper, we demonstrate and rank the effectiveness of a sequential recommendation system by utilizing a production dataset of over 2.7 million credit card transactions for 46K cardholders. The method first employs an autoencoder on raw transaction data and submits observed transaction encodings to a GRU-based sequential model. The sequential model produces a MAP@1 metric of 47% on the out-of-sample test set, in line with existing research. We also discuss implications for embedding real-time predictions using the sequential RS into Nexus, a scalable, low-latency, event-based digital experience architecture.
Sequential recommendation system, transaction data, ML architecture, sequential Neural Network, autoencoder, Information Retrieval.
Xichen Liu1, PatrickLe2, 1Crean Lutheran High School, 12500 Sand Canyon Ave, Irvine, CA 92618, 2Computer Science Department, California State Polytechnic University, Pomona, CA
This paper introduces PlantAssist, a system designed to address the challenges faced by residents in maintaining indoor vegetation. With an increasing number of people leaving their homes for extended periods, the need for a convenient and reliable solutionto water indoor plants has become crucial. Existing options such as hiring caretakers or relying on friends pose security risks and inconvenience. PlantAssist presents a hardware and software solution that emulates regular outside irrigation systems. The hardware is seamlessly integrated into the plants pot, allowing users to choose their preferred style without compromising the rooms aesthetic. The system utilizes smart AI prediction to determine when watering is necessary, ensuring plants receive optimal care. Through a user-friendly app, residents can effortlessly access information about water levels and schedule watering sessions. PlantAssist offers users a hassle-free lifestyle, ensuring their plants are watered without any effort required.
Indoor plant care, Automated watering system, Moisture control, AI prediction.
Purnima Das1, John F. Roddick1, Patricia A. H. Williams1 and Mehwish Nasim1, 2, 1College of Science and Engineering, Flinders University, Tonsley, SA 5042, 2The University of Western Australia, Perth, WA 6009
Association Rule Mining (ARM) has been recognised as a valuable and easy-to-interpret data mining technique in response to the exponential growth of big data. However, research on ARM techniques has mainly focused on enhancing computational efficiency while neglecting the automatic determination of threshold values for measuring the "interestingness" of items. Selecting appropriate threshold values (such as support, confidence, etc.) significantly affects the quality of the association rule mining outcomes. This study proposes an algorithm that utilises Particle Swarm Optimization (PSO) and ARM techniques to determine optimised threshold values in the health domain automatically. The algorithm was evaluated using the UCI machine learning medical database for heart disease. Results show that the proposed algorithm is capable of generating frequent itemsets and rules in an efficient manner and can detect the optimum threshold values. This research has practical implications for the health domains, as it can extract valuable results.
Association Rule Mining; Particle Swarm Optimisation; Health Data; Heart Disease; Optimised Frequent Itemsets.
Abel Varghese, Mahendher Marri and Sibi Chacko, School of Engineering and Physical Sciences, Heriot-Watt University, Edinburgh EH14 4AS, UK
In this paper,a study of autonomous vehicles in MATLAB/Simulink® 2022 is carried out and is conducted using vehicles with three different speeds 40, 80, and 120 km/hr. A normal highway in UAE is considered for road modelling. All vehicles modelled are representative of that available in the UAE.In the model, lane following and lane-keeping assistance functions and Simulink block which are described using artificial neural networks are selected. Simulation is validated with existing published results of physical vehicle models. In the simulations,it is assumed that vehicles have minimal steering angles as the system is in anautonomous collision free environment, selected from MATLAB.Results are obtained as velocities, accelerations, and safe distance with respect to the preceding vehicle. The followingresults are critically analysed and validated.
Artificial Intelligence, Artificial Neural Networks, MATLAB, Autonomous Vehicle, Deep Neural Networks, Ackermann.
Samra Urooj Khan1, N.S.A.M Taujuddin1, Sundas Naqeeb Khan2, Zoya Khan3, 1Faculty of Electrical and Electronic Engineering, Universiti Tun Hussein Onn Malaysia (UTHM), 86400, Parit Raja, Johor, Malaysia, 2Faculty of Computer Science and Information Technology, Universiti Tun Hussein Onn Malaysia (UTHM), 86400, Parit Raja, Johor, Malaysia, 3Lisbon Synopsys, Portugal
The condition of the propagation environment is essential in the design and execution of any transmission medium. As a result, mathematical modeling of transport route has been a focus of study for centuries. Geometrical channel modeling, as demonstrated by researchers and theorists, is best suited for mobile-to-mobile (M2M) communication settings. Several hollow cylindrical geometrical systematic collections have been thoroughly studied in this research. According to the literature, an elliptical modeling technique could more centralized national the transmission channel. Furthermore, the influence of different channel coefficients across multiple-in-multiple-out (MIMO) resonators has been illustrated utilizing geometrical models. Moreover, the velocity of a mobile station (MS) inside the M2M presenter has still not been assessed among the MIMO resonators. For 5G communications networks, a study of several mobile station (MS) variables would be given.
M2M, MIMO, MS, Geometrical modeling, Transmission, 5G, Communication
Aysun Karamustafaoglu1, Morgan Glisson1, Matthew Ferraro1, Seref Recep Keskin2, Gulustan Dogan1, Jennifer Lozano1, John Knox1, 1Nokia Networks 1University North Carolina of Wilmington, U.S, 2Logiwa WMS, U.S
The purpose of this Natural Language Processing (NLP) paper is to discover, learn, and help others to better understand and disclose the sentiment of fictional and blended, non-fictional stories during the period of the Wilmington Riots of 1898 utilizing the literature Hanover; or the Persecution of the Lowly. In addition, the goal of the project is to process the events and sentiments of the characters to derive feeling from the story. We analyzed each sentence and designated a specific sentiment (neutral, positive, negative) and emotion to it. After processing those sentences, we conducted several machine learning processes to try and predict and/or classify the sentiment of a sentence from the text. Although NLP balancing techniques were attempted, our models were not able to predict the sentiment of the sentences at a decent accuracy level. Future work may involve conducting a sentiment analysis on Hanover at a document-level or phrase-level analysis.
Machine learning, Natural Language Processing (NLP), Natural Language Processing, NLP, Sentiment Analysis, Black history, Wilmington Massacre 1898, Wilmington 1898, racism, Wilmington history, David Bryant Fulton, Jack Thorne.
Nelson Ndugu ,Rashmi Margani, Makerere University, Uganda
This paper focuses on critical problems in NLP related to the linguistic diversity and variation across the African continent, specifically with regards to African local dialects and Arabic dialects that have received little attention. The proposed approach features combination of several word embedding techniques with LightGBM to address the challenge of accurately capturing the context expressed in TUNIZI. We evaluated our various approaches, demonstrating its effectiveness while highlighting the potential impact of the proposed approach on businesses seeking to improve customer experience and product development in African local dialects.The idea of using the model as a teaching tool for product-based instruction is interesting, as it could potentially stimulate interest in learners and trigger techno entrepreneurship. Overall, our modified approach offers a promising analysis to the challenges of dealing in African local dialects. Particularly Arabic dialects, which could have a significant impact on businesses seeking to improve customer experience and product development.
Machine Learning, NLP, Computational Linquistic Modeling, Product-based learning.
Manar Khalid Ibraheem, Mbarka Belhaj Mohamed and Ahmed Fakhfakh, Laboratory of signals, systems, artificial intelligence and networks (SM@RTS), Digital Research Center of Sfax (CRNS)
Forest fire applications face a significant challenge in handling a large volume of sensitive data that requires transmission, immediate storage, and processing. The acquisition of information related to natural phenomena is crucial and necessitates efficient tracking and timely processing. However, this task poses constraints in terms of storage space, computing time, and power consumption, particularly within wireless sensor networks (WSNs) that comprise low-cost sensor nodes with limited size and power capabilities. To optimize sensor lifetime and overall bandwidth utilization, it becomes imperative to reduce the transmitted data. This research focuses on investigating the impact of energy consumption in WSNs on forest fire applications by incorporating sleep-based concepts and analysing the disparity in power consumption between sleep mode and active mode. The simulation and analysis of results will be conducted using the Omnet++ program.
Forest fires, Wireless sensor networks (WSNs), Energy consumption, Sleep mode& Omnet++.