Computer Science

In recent years, advancements in Natural Language Processing (NLP) techniques have revolutionized the field of accessibility and exclusivity of testing, particularly for visually impaired individuals. CBT has shown in years back its relevance in terms of administering exams electronically, making the test process easier, providing quicker and more accurate results, and offering greater flexibility and accessibility for candidates. Yet, its relevance was not felt by the visually impaired students as they cannot access printed documents. Hence, in this paper, we present an NLP-driven Computer-Based Test guide for visually impaired students where the NLP-driven Computer-Based Test guide employs state-of-the-art machine learning algorithms to provide real-time assistance and support to visually impaired students. The system utilizes optical character recognition (OCR) technology to convert the text-based questions and the options and also uses automatic speech recognition (ASR) technology to convert spoken words into text-based in a machinereadable format. Subsequently, the NLP model processes the converted text, enabling the system to comprehend and analyze the content. The system uses a pre-trained model to interpret the spoken answers and provides instant feedback to the students, validating their responses and guiding them through the test. The methodology adopted for this system is Object Oriented Analysis and Design Methodology (OOADM) where Objects are discussed and built by modeling realworld instances. To validate that the system is not perverse, the system is further evaluated to test for accuracy using sample labels (A, B, C, D, E, F, G) to compare with the voice recordings obtained from 20 visually impaired students which is been predicted by the system to attain values for precision, recall, and F1-scores. These metrics are used to assess the performance of the model and have indicated that this model is proficient enough to give its better performance to the evaluated system.

Key words: 1. Introduction In this paper, we propose a novel artificial intelligence (AI) approach to the price volatility problem (PVP). Presented is a decision support system (DSS), with the predictive classification ability of deviant learning algorithm (DLA) used in analyzing a Premium Motor Spirit PMS) pump price dataset. The results show that classification accuracy of above 75% can be attained if the learning extent is increased beyond a certain threshold value or is set at a maximum value. The approach can be a useful tool in detecting periods a state in Nigeria may have high likelihood of hike in the pump price of PMS. (Artificial Intelligence, deviant learning, learning extent, predictive classification, price volatility Problem, Premium Motor Spirit, Decision Support System. PMS popularly known as 'Petrol', is one of the major byproduct of the Oil and Gas industry (Onyinye and Nkechi, 2015), in Nigeria that serves the dual purpose of fuel for automobiles and small power generating sets. The product is a volatile liquid with a characteristic odor that depends on the quality of its refinement. It is also referred to as 'gasoline' and its use stems from the fact that it is well suited for most combustible engines as well as its low price when compared to other fuels such as diesel, aviation fuel and natural gas. Due to its popular use, measures have been put in place by both the Federal and State Governments to regulate the pump price of PMS and its distribution nationwide. However, due to market forces, price volatility may arise. Price volatility is one pertinent issue faced by Government and the citizens, which if not handled properly, could lead to major unrest in the country. This is further compounded by the complex nature of the Nigerian state and the PMS distribution bottleneck across the many states of the Federation. The price volatility problem (PVP) is the result of variability in PMS pump price by Government and petroleum regulatory agencies. Even though measures have been put in place to enforce the various price regimes, there is still a clear evidence of hike in the pump price of PMS. Thus, when there is a hike in PMS in a location, a chain reaction of price hikes follow in other locations and this may prove very difficult for the regulatory agencies to monitor and control. The price hike normally results in a phenomenon refer to as the 'local neighbor effect' which describes the condition of a contagious hike from one nearby petrol pump filling station to another. In order to implement a PMS pump price hike detection/prediction scheme, several approaches have been put forward by some researchers. The use of statistical analysis for forecasting petrol price in the Australian market has been investigated in (Pekol, 2010). This analysis include the identification and use of some crude-oil related variables such as fuel price, population size, and 52