Journal of Science & Technology (JST)
Not a member yet
967 research outputs found
Sort by
AI Based Detecting Deception in Online Interactions: An Analysis of the Dishonest Internet Users
Get PDFWith the widespread adoption of the internet, online interactions have become an integral part of modern communication. However, this surge in digital interactions has also brought about a significant rise in deceptive practices, ranging from misinformation and fraud to identity theft and cyberbullying. Detecting and mitigating these dishonest behaviors has become a critical concern for maintaining trust and integrity in digital communities. The primary challenge lies in developing a robust and automated system capable of identifying deceptive content amidst the vast volume of online interactions. In the absence of advanced AI-based systems, deception detection in online interactions has heavily relied on manual monitoring, keyword-based filters, and rule-based algorithms. These conventional methods are limited in their effectiveness, as they struggle to adapt to evolving deceptive tactics and often generate false positives or negatives. Therefore, the need for effective deception detection systems in online interactions has never been more pressing. The advent of social media, e-commerce, and various online forums has created an environment where deceptive practices can have far-reaching consequences. Ensuring the safety and trustworthiness of these platforms is imperative for user confidence, cybersecurity, and the overall well-being of online communities. Hence, by utilizing machine learning algorithms, advanced linguistic analysis, and behavioral pattern recognition, this research aims to develop a powerful tool capable of accurately discerning deceptive from genuine online interactions. Through the integration of multi-modal approaches and feature engineering, the proposed system promises to significantly enhance the accuracy and efficiency of deception detection in digital communities, ultimately fostering a safer and more trustworthy online environment.
 
MACHINE LEARNING FOR ROBOT NAVIGATION CLASSIFICATION USING ULTRASOUND SENSOR DATA
Get PDFRobot navigation is a crucial aspect of robotics, enabling autonomous robots to move safely and efficiently through their surroundings. Conventionally, engineers and programmers have relied on fixed rules and heuristics to guide robot movements. However, these rules are often specific to certain environments and struggle to adapt to new or changing conditions. For instance, simple obstacle avoidance techniques or path planning algorithms are commonly used. While effective in controlled settings, they lack the flexibility needed to handle diverse and unpredictable surroundings. In recent years, machine learning (ML) has emerged as a promising alternative. ML allows robots to learn from data and adjust their navigation strategies based on real-time sensory inputs. As a result, this project focuses on implementing ML for robot navigation classification, aiming to create more capable and versatile robotic systems. By utilizing this approach, robots can learn from their experiences and sensory data, improving their ability to navigate complex environments. This adaptive approach is especially valuable in scenarios where the environment undergoes frequent changes or presents diverse and challenging obstacles, beyond what traditional rule-based methods can handle. The utilization of ultrasound sensor data as input provides the robot with valuable distance information, enabling precise obstacle detection and avoidance. Furthermore, incorporating ML into robot navigation enhances their capability to handle complex real-world scenarios and dynamic environments. The use of ultrasound sensor data proves to be a valuable choice, providing crucial information for accurate obstacle detection and path planning. Ultimately, this proposed ML-based approach underscores the potential of ML techniques (i.e., logistic regression, and multilayer perceptron) in enhancing robot navigation capabilities, opening doors for more advanced and autonomous robotic systems capable of operating effectively in diverse and unpredictable environments.
 
A MACHINE LEARNING FRAMEWORK FOR BIOMETRIC AUTHENTICATION USING ELECTROCARDIOGRAM (ECG)
Get PDFThis paper presents a system for how with suitably embrace and modify AI (ML) techniques used to construct electrocardiogram (ECG)- based biometric authentication systems. The proposed system can assist agents and engineers in ECG-based biometric verification components to define the boundaries of required datasets and get preparing information with great quality. To determine the limits of datasets, a use case analysis is conducted. In light of different application scenarios for ECG-based verification, three distinct use cases (or validation classes) are developed. By providing more qualified preparing information given to corresponding AI models, the accuracy of ML-based ECG biometric authentication systems are expanded in result. The ECG time cutting method with the R-top mooring is utilized in this system to secure ML preparing information with great quality. In the proposed system, four new measurement metrics are acquainted with assess the quality of the ML training and testing data. Additionally, a Matlab toolkit, containing all proposed tools, metrics, and test data with exhibitions utilizing different ML techniques, is developed and made publicly available for further analysis. For developing ML-based ECG biometric authentication, the proposed system can guide experts to establish the appropriate ML solutions and the ML training datasets along with three identified user case scenarios. For analysts taking on ML techniques to design new systems in other research domains, the proposed framework
 
Evaluation of Microbial Pathogens and Effect of Time on the Quality of Fermented African Oil Bean Seed
Get PDFThe African oil bean seed (ugba) is a fermented food kept and sold for a period of time by vendors. The study was carried out to determine the presence of pathogens in the african oil bean seed (ugba) and determine the effect of time on the samples. A total of 12 samples were obtained from vendors in three different markets within Enugu metropolis, Enugu state, Nigeria. The already fermented samples were further kept for 4 days and their microbial count determined. Cultures were done on macconkey agar, potato dextrose agar, salmonella shigella agar, eosin methylene blue agar and nutrient agar. The isolates were characterized and identified by standard microbiological methods and antibiotic susceptibility pattern were identified by disk diffusion method. The values for the microbial counts were carried out in triplicate and results reported as mean ± standard error, the data obtained were subjected to one-way ANOVA using statistical package for social science (SPSS) for windows evaluation. There was statistical difference at p<0.05 between the microbial count from the first to the fourth day in the samples from the various markets. The values ranged from (1.61 ± 0.03) in Garriki, (1.37 ± 0.02) in Mayor and (1.52 ± 0.08) in Agbani market in the first day while values of the fourth day were; (1.45 ± 0.09) in Garriki, (8.60 ± 0.28) in Mayor and (1.29 ± 0.03) in Agbani market. Organoleptic changes in texture and colour were seen to be a factor of time due to storage. E.coli, Salmonella sp and Bacillus sp were isolated from samples in all the market. A decline in growth of E.coli and Salmonella sp by the fourth day of storage was observed for all samples.From the results, ugba
spoilage is primarily a result of the continued activity of fermentative organism; Bacillus sp and other spoilage organisms; E. coli and Salmonella sp.
 
A ROUTING PROTOCOL FOR EFFICIENTLY MANAGING ENERGY IN UNDERWATER WIRELESS SENSOR NETWORKS WITH DEPTH CONTROL
Get PDFIn Underwater wireless sensor network attracted massive attention from researchers. In underwater wireless sensor network, ma ny sensor nodes are distributed at different depths in the sea. Due to its complex nature, updating their location or adding new de vices is pretty challenging. Due to the constraints on energy storage of underwater wireless sensor network end devices and the comple xity of repairing or recharging the device underwater, this is highly significant to strengthen the energy performance of underwater wireless sensor network. An imbalance in power usage can cause poor performance and a limited network lifetime. To overcome these issues, we propose a depth controlled with energy-balanced routing protocol, which will be able to adjust the depth of lower energy nodes and be able to swap the lower energy nodes with higher energy nodes to ensure consistent energy utilization. The proposed energy -efficient routing protocol is based on an Genetically updated Algorithm and data fusion technique. In the proposed energy-efficient routing protocol, an existing genetic algorithm is enhanced by adding an encoding strategy, a crossover procedure, and an improved mutation ope ration that helps determine the nodes. The pro- posed model also utilized an enhanced back propagation neural network for data fusion operation, which is based on multi-hop system and also operates a highly optimized momentum technique, which helps to choose only optimum energy nodes and avoid duplicate selections that help to improve the overall energy and further reduce the quantity of data transmission. In the remaining energy and directions of each participating node. In the simulation, the proposed model achieves 86.7% packet delivery ratio,keeping the energy usage at 12.6% and 10.5% packet loss rati
Performance Evaluation of Approximate (8; 2) Compressor for Multipliers in Error-Resilient Image Processing Applications
Get PDFAbstract
Approximate computing improves performance in error-resilient applications like image and video processing. Multipliers are part of its computing unit, which frequently requires a large number of resources. This study compares an approximation (8; 2) compressor to other known models in terms of quality, power consumption, delay, and circuit area. The proposed approximation compressor is implemented in 8 × 8 and 16 × 16 multipliers. To demonstrate the quality of the suggested compressor, an 8 × 8 approximation multiplier was utilized to multiply two images in MATLAB tools. Qualitative measures such as SSIM and PSNR were examined and acceptable results were obtained. The suggested 8 × 8 multiplier circuit produces an acceptable error rate, as indicated by the MED and NED accuracy criteria. Finally, we used a Synopsys Design Compiler to synthesize the proposed approximation compressor and multiplier designs. The suggested 16 × 16 multiplier improves latency, area, and power delay products by 5%, 17%, and 8%, respectively, compared to similar current approximate multipliers.
 
A PRELIMINARY STUDY ON DISINFECTION AND IN VITRO PROPAGATION OF CELOSIA TRIGYNA LINN.
Get PDFCelosia trigyna is an underutilized indigenous vegetable in Nigeria with numerous medicinal uses. Information is sparse on any attempt to conserve it. In other to conserve it through in vitro propagation, a disinfection protocol was developed using Clorox (NaClO) and Mancozeb (C8H12MnN4S8Zn) (Treatment T1 to T9) and a preliminary investigation testing its responsiveness in vitro. Nodal explants of Celosia trigyna were collected from the mother plants at Obafemi Awolowo University Estate, Ile – Ife (Latitude 7⁰32’N and Longitude 4⁰31’E). The nodal explants of C. trigyna were cultured on Murashige and Skoog (MS) medium supplemented with various levels of plant growth regulators (PGRs). The clean cultures obtained were monitored for degree of callus formation, morphology of callus, shoot and root regeneration. Treatment T6 (5% v/v Clorox for 9 minutes and 1% w/v mancozeb for 5 minutes) was best for disinfecting nodal explants of C. trigyna resulting into 81.47 ± 2.14a % aseptic cultures. Nodal explants cultured on MS with 0.5mg/L 2, 4 –D produced callus. This study established that using 5% v/v Clorox for 9 minutes followed by 1% w/v mancozeb for 5 minutes is effective for producing clean cultures of nodal explants and that C. trigyna can be propagated in vitro
City Guard: Raspberry Pi 4 Driven Accident Surveillance System with MEMS, GPS, GSM, and Integrated Applications
Get PDFThis study is focused on the concept of the Internet of Things (IoT)-based smart city model on Rasp- berry Pi. It assesses the design and implementation of a smart city based on the Internet of Thingsusing Raspberry Pi. The main aim is to develop an urban IoT system that will aid in the realizationof a smart city while also resolving domestic issues using Raspberry Pi. Raspberry Pi is an essential component of the smart city model’s implemented system. The concept of the IoT helps analyze the functions used to define the functions of Raspberry Pi. By considering the developing country such as India, IoT plays an essential role in building a smart city model. In this work, the different low-cost operations have been analyzed based on the functions of Raspberry Pi for the smart city model.
 
Review Paper on Inverted Brayton Cycles
Get PDFThe exhaust gas from an internal combustion engine contains approximately 30% of the thermal energy of combustion. Waste heat recovery (WHR) systems aim to reclaim a proportion of this energy in a bottoming thermodynamic cycle to raise the overall system thermal efficiency. One of promising heat recovery approaches is to employ an inverted Brayton cycle (IBC) immediately downstream of the primary cycle. However, it is a little-studied approach as a potential exhaust-gas heat-recovery system, especially when applied to small automotive power-plants.The experiments of the IBC prototype were conducted in the gas stand. The correlated IBC model can be utilized for the further development of the IBC system. Researchers were reviewed core paper on Inverted Brayton Cycles (IBC) and concluded that there were possibility of heat recovery system in that for changing different mechanical components
Milk Processing: A Review
Get PDFNutritious milk needs to be processed after milking to improve its shelf life and make it healthy for human consumption. One important factor Microbial contamination is responsible for milk spoilage. There is some internal anti-microbial defense mechanism in freshly drawn milk, but for longer, to avoid or decontaminate microbial growth, some other external anti-microbial preservation measures need to be initiated for storage. This review article provides some information about milk processing