FACULTY OF PHYSICAL SCIENCE

Understanding customer’s preference for product features is crucial for businesses seeking to optimize their offerings and increase sales. This study aims to identify key product features that significantly influence purchasing decisions. The research specif ically examines customer preference across different demographic groups, such as age and income levels, to determine variations in product feature importance. Using chi-square test for analysis, this study evaluates the association between demographic factors and product features preferences, providing insight into customer’s decision-making patterns. Based on the analysis it was found that the key product features that affect customers purchasing decision is quality and performance. The findings will help businesses tailor their marketing and product development strategies to better align with costumer expectations

This project implements a deep learning algorithm for image recognition, focusing ontrafficsign classification. Traditional machine learning methods struggle with manual featureextraction and dataset diversity. To address these limitations, a robust convolutional neural network (CNN) with residual blocks, dropout layers, and global average pooling is utilized. Preprocessing techniques like normalization and data augmentation enhance accuracyandgeneralization. Using TensorFlow and Keras, experiments were conducted on the German TrafficSignRecognition Benchmark (GTSRB) and the Chinese Traffic Sign Dataset. The model achieved99.54% validation accuracy and 94.95% test accuracy on GTSRB, but overfittingledto60.38% accuracy on the smaller Chinese dataset. The study highlights CNN effectiveness in pattern recognition, with strengths inGPUacceleration and modular architecture. Challenges like overfitting and computational constraints persist. Future research should explore transfer learning, ensemble methods, andreal-time optimization to enhance performance. This study advances deep learning-basedimage recognition for applications in autonomous driving and traffic management. This project implements a deep learning algorithm for image recognition, focusing ontrafficsign classification. Traditional machine learning methods struggle with manual featureextraction and dataset diversity. To address these limitations, a robust convolutional neural network (CNN) with residual blocks, dropout layers, and global average pooling is utilized. Preprocessing techniques like normalization and data augmentation enhance accuracyandgeneralization. Using TensorFlow and Keras, experiments were conducted on the German TrafficSignRecognition Benchmark (GTSRB) and the Chinese Traffic Sign Dataset. The model achieved99.54% validation accuracy and 94.95% test accuracy on GTSRB, but overfittingledto60.38% accuracy on the smaller Chinese dataset. The study highlights CNN effectiveness in pattern recognition, with strengths inGPUacceleration and modular architecture. Challenges like overfitting and computational constraints persist. Future research should explore transfer learning, ensemble methods, andreal-time optimization to enhance performance. This study advances deep learning-basedimage recognition for applications in autonomous driving and traffic management.

The web-based sectional result computation system abstracts the complexitiesof calculating and generating academic results in a University or educational institution. It provides a user-friendly interface that simplifies data management, computationalgorithms, result generation, security, access control, and integration withothersystems. By abstracting these components, the system streamlines the result computationprocess and ensures accuracy, reliability, and timeliness. It simplifies data storage, retrieval, and organization, automates grade calculations and performance indicators, and generates result sheets and reports in customizable formats. The systemalsoabstracts security measures, ensuring authorized access to student results, and integrateswith other institutional systems for seamless data synchronization. This abstractionallows users, such as faculty, staf , and administrators, to focus on result analysis, interpretation, and decision-making, rather than getting overwhelmed by technical details. The web-based nature of the system enables easy access fromanywhere withaninternet connection, enhancing accessibility and convenience. Overall, the result computation system's abstraction simplifies and automates result processing, improvesef iciency, reduces errors, and provides standardized and reliable frameworks forhandling academic outcomes. It empowers users with accurate and timely student results, facilitating informed decision-making and enhancing the overall academic experiencewithin the institution

This study discusses the pseudo random number generators, one of the categories used in generating random numbers. Random numbers are useful in various simulation processes, such as statistical and numerical analysis, gaming, cryptography, gambling, etc. It is therefore important to the study the process of generating random numbers. The purpose of this study is to observe the concept of the pseudo-random number generating techniques, some examples and required properties of the pseudo random number generators and a specific pseudo random number generating algorithm for the generation of sequence of random numbers

This research focuses on the design and implementation of an intelligent chatbotcourse adviser system for the Department of Computer Scienceat theUniversity of Benin. The study addresses the limitations of traditional manualcourse advisor methods by leveraging artificial intelligence, machine learning, and natural language processing technologies to create an automated, efficient, and user-friendly academic guidance system. The research employed a mixed-method approach, combining qualitativeandquantitative data collection techniques to ensure comprehensive systemdevelopment. The study included surveys of students and academicstaff, analysis of existing course adviser processes, and systematic evaluationoftechnological requirements. The implementation phase involved developinganintelligent chatbot system with advanced features including 24/7 availability, personalized course recommendations, real-time prerequisite verification, andautomated academic progress tracking. Results demonstrate significant improvements in academic adviser services, with the chatbot system providing immediate, accurate, and consistent courseguidance. The system successfully reduced administrative workload, minimizedadvisory errors, and enhanced student access to academic support. Useracceptance testing showed high satisfaction rates among students andstaff, xi validating the system's effectiveness in addressing traditional advisorychallenges. The research contributes to the growing body of knowledge in educationaltechnology and provides a practical framework for implementing AI-drivenacademic support systems. The findings suggest that intelligent chatbot systemscan significantly enhance academic adviser services, offeringpotentialapplications across various educational institutions. Recommendationsforfuture development and system optimization are provided based on the researchoutcomes.

This study was carried out to investigate the occurrence and structural effects of bitumen in Agbabu community using integrated geophysical methods (aeromagnetic, aero-radiometric and electrical resistivity tomography (ERT). Two forms of datasets (Secondary and Primary) were used. These datasets provided useful information on the lithology and geological structures within the area. The geophysics data processing approach employed concentrated on enhancing the geophysical data quality and this aided in tracing accurate positioning of geological boundaries, the responses related to bituminous zone and geological structures that may be of vital economic importance The digitized geological map covering the study area was obtained for lithological information. The secondary datasets consist of an aeromagnetic and aero-radiometric basically for the reconnaissance study. Aeromagnetic and aero-radiometric map was obtained from the Nigerian Geological Survey Agency (NGSA) and processed using Oasis Montaj software to depict main lithology and structural features present in the Agbabu area. The primary data was acquired within the area suspected to have high potential for bitumen deposit using the wenner-schlumberger configuration. The potential difference produced was measured with the aid of PASI 16 GL-N Earth resistivity meter. The apparent resistivity
values obtained was processed using RES2DINV software which helped to automatically obtain the 2D inversion model of the subsurface. The results of aeromagnetic study show low TMI amplitude (-201.5 – 16.8nT) and high AS amplitude (0.053 – 0.172nT) at suspected bitumen deposit regions. The low TDR amplitude (- 1.4 to - 0.4nT) confirms a concealed basement depression hosting bitumen deposit. The 3D-Euler deconvolution helps to locate the sources of magnetic anomalies, it is deeper within the sedimentary terrain (841 – 1703m) and shallower (185 – 841m) within the basement. The interpretations of radiometric datasets revealed the spatial variation of potassium (K), thorium (Th) and uranium (U) radioelement concentrations as high as (0.6 – 2.5%), (7.9 – 28.0 ppm) and (1.9 – 5.1 ppm) respectively within the basement complex, but also low as (0.0 – 0.6%), (1.8 – 9.2ppm) and (0.3 – 1.9ppm) respectively within the sedimentary terrain. The ternary image shows very 16 low radiometric intensity that contained the bitumen structures, moderately high around lithological boundaries and concealed linear features reveal. This research has shown that the occurrence of bitumen was found between the depth of 13.4m and 9.93m for Traverses 1,2,3 and Traverses 4,5 respectively in 2- Dimensional electrical resistivity images which corroborated by boreholes with a depth of about 18m. The
results of this research indicated that bitumen has an average thickness of 11.67 m. The answers aid the exploitation of bitumen and made available for government and other relevant bodies in formulating policies for minerals development in the country. From the result of the coordinates of the airborne surveys, agreed perfectly with the values of Geophysical (ERT) traverses indicating the reliability of this type of joint geophysical investigation

Legendre polynomial is a second order ordinary differential equation as well as a type of Fourier Series written in the system of orthogonal polynomials with a vast number of mathematical properties and numerous applications. It is obtained through linear differential equation methods based on Sturm-Liouville theory.

Molecular dynamics and simulations are a vital tool used for the description of molecules in terms of their shapes, bonds, length, dihedrals and molecular structure in space. Molecules that exist tends to undergo some discrete changes when subjected to a particular environment. This changes in molecules over the time defines the final or ultimate state in which the molecule can or will exist. One of such molecules which could be simulated is Myosin. Myosin is a rotor protein molecule responsible for muscle contraction of bio-organisms. Considering its aid in kinetic activities, it is therefore a significant molecule to study and investigate under molecular dynamics and simulations. In this study, we are to carry out molecular dynamics simulation on myosin molecule using the Avogadro software, by importing the molecule, visualising, optimising its geometry and carrying out its energy minimisation. The results show that the minimised energy is the same for any time steps used. Also, the time steps used affects the time taken for the simulation to complete. Molecular dynamics simulation of myosin molecule is very paramount in aiding scientist on how drug and bio-supplement should be modelled to suite the biological systems they are to be used on.

Proteins are one of the most important families of biological macromolecules. Proteins can assume many different structures. Adopting different computational methods many protein functions and structure related problems can be explored. This thesis focuses on three different protein issues. The structural changes induced by high temperature on a large enzyme were investigated simulating the denaturation of glucose oxidase. Molecular dynamics (MD) simulations at different high temperatures were performed. The transition state of the denaturation process was found and the relative ensemble of structures characterized. Different protein properties were analyzed and found in agreement with experimental and theoretical data. Moreover the breaking points of the protein were localized and point mutations on the protein sequence were suggested. Antifreeze proteins (AFP) allow different organisms to survive in subzero environments. These proteins lower the freezing point of physiological fluids. MD simulations of the snow flea AFP (sfAFP) in water have shown the partial instability of the protein structure. When attached to different ice planes at the ice/water interface, the sfAFP induces local ice melting. AFPs are divided into two categories: hyperactive and moderately active depending on their antifreeze power. The water diffusion profile of ice/water systems containing one protein from each family were compared

Over the years, Nigeria has experienced a steady growth in the telecommunication industry, with over ten (10) mobile wireless telecommunication services providers, over fifteen thousand (15,000) Base Transceiver Stations (BTS) spreads across the country, and also over One Hundred and seventeen million lines actively connected. With this tremendous growth, the concern among Nigerians about the effects and possible health hazard of electromagnetic waves radiation from the BTS and the mobile handsets cannot be over emphasize. This study focuses on measurement of Electromagnetic waves Radiation (w/m2) radiated from Base Transceiver Stations (BTS). These measurements were carried out using the handheld GQ EMF 390 digital meter. Maximum radiation values were measured and recorded from (BTS) of a selected site in UNIBEN and the results were compared with the standard provided by the International Commission on Non-Ionizing Radiation Protection, ICNIRP, 1998 (4.5W/m2 for 900MHz and 9W/m2 for 18000MHz) and an agreement was obtained.