Scholarly Publication
Permanent URI for this collection
Browse
Recent Submissions
- ItemA Neuro-Fuzzy-based Approach to Detect Liver Diseases(Pakistan Journal of Engineering and Technology, 2024) Ronke Seyi Babatunde; Akinbowale Nathaniel Babatunde; Shuaib Babatunde MohammedThe liver is a crucial organ in the human body and performs vital functions essential for overall health, including metabolism, immunity, digestion, detoxification, and vitamin storage. Detecting liver diseases at an early stage poses challenges due to the liver's ability to function adequately despite partial damage. Early detection is crucial as liver diseases have significant clinical and socio-economic impacts, affecting other organ systems and requiring timely intervention to improve patient survival rates. Classical diagnostic methods for liver disorders may not always produce better results, thus necessitating more advanced and accurate diagnostic systems. Intelligent systems like predictive modeling and decision support systems, have shown promising results in recent years in disease detection and are aiding medical practitioners. In this research, a neuro fuzzy-based system integrating neural networks and fuzzy logic (FL) was implemented. Based on the risk factors present in the dataset that was used to benchmark the algorithm, this system offered a classification accuracy of 97% which is comparable with existing systems in the literature. This study creates a neuro-fuzzy system for early liver disease identification, solving diagnostic issues and offering healthcare improvements. The proposed system is validated by presenting the simulation results
- ItemA COMPARATIVE ANALYSIS OF DEEP LEARNING TECHNIQUES FOR INSIDER THREAT DETECTION(FUW Trends in Science&Technology Journal, 2024-06-28) A.O.Akinlolu; E.R. Jimoh; I.A.Alabere; O.EkundayoAn insider threat refers to any malicious activities carried out by employees, contractors, or vendors who have authorised access to an organisationโs IT assets, resulting in significant negative impacts on its data and information resources. Existing literature reviews have identified shortcomings in utilizing diverse domains (such as system log files, file processes, logon records, HTTP, email, external drives, and the Lightweight Directory Access Protocol (LDAP)) to develop techniques capable of identifying insiders posing a threat to an organisation with minimal false positives. In contrast, this study opts for a more robust domain, specifically LAN data packets, to assess the activities of users within the organization's Local Area Network. It monitors deviations from the normal flow of data packets on the network, thereby classifying these anomalies as either malicious or benign. The synthetic dataset KDDCUPโ99, evaluated in this study, is widely recognized as one of the few publicly accessible datasets for network-based anomaly detection systems. The proposed stacked ensemble model demonstrates superior predictive performance, achieving an accuracy of 98%, compared to the 91.88% and 98.58% accuracy of the individual classifiers Naรฏve Bayes and KNN, respectively. The model can be improved by incorporating additional user behaviours such as email communication, browser activity, and file access to enhance accuracy and applicability.
- ItemDiacritic Restoration for Yoruba Text with under dot and Diacritic Marks Based on LSTM(FUOYE Journal of Engineering and Technology, 2023-09-18) Kingsley L.Ogheneruemu, Jumoke F. Ajao, Abdulrafiu M. Isiaka, Franklin O.Asahiah and Olumide K. OrimogunjeYoruba is a tonal language spoken primarily in Nigeria, some West African countries, and other parts of the world by over 40 million people. Many Yoruba texts written online lack tone marks, which can be confusing, ambiguous, and difficult for Natural Language Processing. This paper presents a method, which combines syllable-based approach and long short-term memory (LSTM) for diacritics restoration of standard Yoruba text. By enhancing the built-in varnishing gradient of RNN, the aim is intended to recover lost diacritics in Yoruba text for both characters that carry diacritic signs and underdot and return it with the proper diacritics. Data were acquired from Yoglobavoice, BBC Yoruba new and Yoruba words collected from literate indigenous writers. 27050 Yoglobalvoice datasets, 2000 Yoruba words extracted from BBC Yoruba news, and 1470 Yoruba words collected from a Yoruba language teacher. In addition, syllabic module was developed to group the tokenized word into different syllables. The output of the syllabication algorithm was fed into the Long Short-Term Memory (LSTM) module for training, the LSTM model was trained using 70% of the dataset and validated using 30% of the dataset. The result obtained showed 96% accuracy. From the result, it was observed that the use of LSTM for restoring diacritic gave an improved restoration of both character with under dot and character that contains tone-marks.
- ItemEnhancing Short-Term Load Forecasting Precision with Hybridized RNN and LSTM Model(Wellspring University Journal of Science and Computing, 2024-12-20) Foluso, A. A., Ajao, J. F., Musa, A., Okunade, O. A., and Ekundayo, O.This study evaluates the efficiency of various machine learning models for short-term load forecasting (STLF) at the 132KV/33KV Bida Transmission Station. Using load outage data from the Kutigi and Lemu feeder stations and meteorological data from NIMET, the research compares models such as RNN, GRU, LSTM, and a hybrid RNN-LSTM model. The hybrid RNN-LSTM model demonstrated superior performance, achieving low error metrics (MAE: 0.101, RMSE: 0.138) and a near-perfect coefficient of determination (Rยฒ: 0.999). Pearson's correlation analysis was pivotal in feature selection, enhancing the model's predictive accuracy. These findings highlight the hybrid RNN-LSTM model's ability to effectively capture complex temporal patterns, making it a valuable tool for precise STLF. The outcomes have significant implications for improving grid reliability and operational efficiency in the energy sector.
- ItemAn Enhanced Message Encryption Approach Using Residue Number System(Kasu Journal of Computer Science, 2024) Akeem Femi Kadri; Kafayat Odunayo Tajudeen; Ayisat Wuraola Asaju-GbolagadeSafe transmission of messages over the internet is ensured by data security. The data should be protected against unauthorized access and transmitted to the legal recipient safely and formally called cryptography. The well-known cryptography types are asymmetric and symmetric and different symmetric encryption algorithms such as Advanced Encryption Standard (AES), Data Encryption Standard (DES), Blowfish were proposed to protect the confidentiality of the transmitted and stored data. However, most of these symmetric techniques are vulnerable to brute-force attack because the cipher text remains unintelligible for the original data to be found. Consequently, this challenge prompted researchers to introduce Residue Number System for encryption. RNS helps to withstand the vulnerability of the encryption algorithms in brute-force attacks making transmitted and stored data more secure. Hence, this paper introduced encryption by using RNS to improve the security of transmitted data. Traditional moduli set set {๐1=2๐โ1,๐2= 2๐ ๐๐๐ ๐3=2๐+1} and Chinese Remainder Theorem (CRT) were used to encrypt and decrypt the data based on information exchange. Furthermore, this is done by representing each character in the plaintext data is assigned a unique number value and which is done by converting each character into its ASCII value and the ASCII value is then converted to Excess-3 value. Thus, anytime an attacker tries to access the data, the RNS algorithm will generate residues data, this helps to discourage an attacker from further threat. The result showed that the proposed system was able to resist brute-force attacks compared to others systems. Values of n for the moduli set were dynamic against encryption and decryption of data.