| 1 |
Author(s):
Ating Emmanuel A., Sikoki Francis D..
Page No : 1-9
|
Relative Yield and Sex Ratio of the West African Fiddler Crab Uca Tangeri in Mbo River, Niger Delta, Nigeria.
Abstract
The study of the relative yield and sex ratio of the West African fiddler crab (Uca tangeri) in Mbo river of Akwa Ibom State, Niger Delta, Nigeria, was conducted for 12 consecutive months. Uca tangeri exhibited sexual dimorphism with a sex ratio of 1.2:1.0, which was significantly biased in favour of males. The yield isopleths demonstrate the response of the crab to both variation in E (exploitation level) and Lc/L∞ (a proxy for mesh size). Yield contours with Lc/L∞ = 0.74 usually consist of four quadrants (Pauly and Soriano, 1986), each with its characteristics. The yield isopleths with Lc/L∞ = 0.74 and E = -1.62 belong to quadrant B, which implies that, large specimens were caught at high effort level.
| 2 |
Author(s):
Samuel O. Essang, Jackson E. Ante, Augustine O. Otobi, Stephen I. Okeke, Ubong D. Akpan, Runyi E. Francis, Jonathan T. Auta, Daniel. E. Essien, Sunday E. Fadugba, Olamide M. Kolawole, Edet E. Asanga, and Benedict I. Ita.
Page No : 10-26
|
Optimizing Neural Networks with Convex Hybrid Combination of Activation Functions: A New Approach to Enhance Gradient Flow and Learning Dynamics.
Abstract
Activation functions are crucial for the efficacy of neural networks as they introduce non-linearity and affect gradient propagation. Traditional activation functions, including Sigmoid, ReLU, Tanh, Leaky ReLU, and ELU, possess distinct advantages but also demonstrate limits such as vanishing gradients and inactive neurons. This research introduces an innovative method that integrates five activation functions using coefficients to formulate a new hybrid activation function. This integrated function seeks to harmonize the advantages of each element, alleviate their deficiencies, and enhance network training and generalization. Our mathematical study, graphical visualization, and hypothetical tests demonstrate that the combined activation function provides enhanced gradient flow in deeper layers, expedited convergence, and improved generalization relative to individual activation functions. Significant outcomes encompass the alleviation of disappearing gradient and inactive neuron issues, augmented gradient stability, and enhanced expressiveness in intricate neural networks. These findings indicate that mixed activation functions can enhance the learning dynamics of deep networks, offering an effective and resilient alternative to conventional activation functions.
| 3 |
Author(s):
Mansur Bala Safiyaniu, Abdulrazak Muhammad Idris, Suleiman Adamu, Auwalu Ibrahim Abba, Oyom Bright Bassey.
Page No : 27-36
|
Molecular Detection of Bovine Viral Diarrhoea Virus (BVDV) Infection Among Cattle in Daura Zone, Katsina State, Northwest, Nigeria.
Abstract
Bovine viral diarrheal virus (BVDV) infection causes a diverse range of clinical outcomes from being asymptomatic, or a transient mild disease, to producing severe cases of acute disease that leading to animal death. The infected animals may suffer from mild diarrhoea or respiratory symptoms or else show no signs of infection at all. BVDV is a small, enveloped single-stranded positive-sense RNA virus, measuring about 12.5kb that belongs to the Pestivirus genus and Flaviviridae family. Proper control of BVDV involved the removing of infected animals from the herd, this can achieved through proper detection of BVDV infected animals. This study aimed at molecular detection of BVDV infection among rearing cattle in Daura zone, Katsina State, Northwestern Nigeria. 125 blood samples were analyses for the presence of BVDV using RT-PCR according to manufacturer information. An overall prevalence rate of 10.4% was obtained, and diarrhoea, nasal and eye discharge remained a major sign of the infection. The study suggest the need for improving sanitary politics in the veterinary sector to prevent potential transmission the BVDV infection among the cattle and other domestic animals in the study area.
| 4 |
Author(s):
Chukwudi Anderson Ugomma, Kenneth Okwuosha.
Page No : 37-55
|
On Empirical Selection of Lognormal and Weibull Distributions: Application to Nigerian Stock Market.
Abstract
In this article, we empirically compared and selected the best performed distribution among Lognormal and Weibull distributions. The dataset for this study were monthly stock prices of BUA cement enlisted and trades on Lagos Stock Exchange, a subsidiary of Nigeria Stock Exchange (NSE). The dataset comprised of sample of fifty-eight (58) log-transformed monthly closed stock prices between 2nd January, 2020 and 1st November, 2024, obtained from https://ng.investing.com/equities/bua-cement-plc-historical-data .The Maximum Likelihood Estimator (MLE) was used to obtain the parameters of both Lognormal and Weibull distributions in a view of comparing and selecting the best distribution that fits our dataset. The Minimum Mean Squared Error (MSE) and Akaike Information Criterion (AIC) were used as selection criteria and the Weibull distribution was found to outperform the Lognormal distribution since it exhibited the least MSE and AIC. Also, the selected Weibull distribution was subjected to goodness-of-fit using Kolmogorov-Smirnov test and the empirical evience shows that BUACEM stock price follows a Weibull distribution with 5% level of significance, hence, making the Weibull distribution the right choice for fitting BUACEM stock prices on the Nigerian Stock market.
| 5 |
Author(s):
Fendoung Dejiang Guy Hector, Sieliechi Joseph, Ngassoum Martin Benoît.
Page No : 56-69
|
Effect of Landfill Leachate Salinity on the Performance of Hybrid Vertical Flow Treatment Wetlands under Equatorial Climate.
Abstract
In this study, vertical flow treatment wetlands performances on landfill leachate operated under equatorial climate have been assayed during 2 years. Two pilots planted with Echinochoa Pyramidalys Lam have been operated. Effluent electrical conductivities (EC) measured at each of the four different seasons with the values of 2.5±0.5; 6±1; 7±2 and 9±1 mS/cm for long rain, short rain, short dry and long dry respectively; were used to determined COD and TN treatment performances. According to the experimental results, EC, pH, COD and TN decrease at different treatment stage. The evaluation of treatment system Removal Rate (RR) showed highest value of 66±6% for RRCOD and 79±6% for RRTN both obtained during long rain season. The highest RMCOD (171±9 gm-2d-1) and RMTN (4±1gm-2d-1) was obtained during short rain season. This work will help to design a full scale TW for LL under equatorial climate in Africa.
| 6 |
Author(s):
Okoh Jophet Ewere, Owoyi Mildred Chiyeaka, Okoh Jophet Ewere.
Page No : 70-84
|
Comparative Study of Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA) and Support Vector Machine (SVM) In Dataset.
Abstract
Classification techniques is an important factor in data analysis. Over the years, different classification method have been proposed for classification of dataset. In this paper, we compared three classifiers (LDA, QDA and SVM) in three imbalanced datasets (Iris, Pima and Glass data) and misclassification rate of the three classification method were compared. The experiments concentrated on analyzing the average misclassification rate among classifiers across the three dataset studied using the misforest imputation method to balance the dataset respectively. The results reveal that for the glass dataset, the QDA classifies the dataset better than the two other classification method studied, while for the iris and glass datasets, the LDA outperformed the other two classifiers studied. The conclusion in this study is that LDA have the least average misclassification error, followed by the QDA and then the SVM with an average misclassification rate of 0.2863.
| 7 |
Author(s):
Okafor S. E., Aronu C. O..
Page No : 85-93
|
Evaluating the Performance of Laplace and Its Variants in Modelling Economic Data.
Abstract
The Laplace distribution and its extensions have been widely utilized in statistical modelling due to their ability to capture real-world data characteristics such as skewness and heavy tails. This study evaluates the performance of the classical Laplace (L) distribution against three of its variants: the Transmuted Laplace (TL), Alternative Laplace (AL), and Asymmetric Laplace (ASL) distributions. While these extensions introduce additional parameters to enhance flexibility, their empirical performance remains a subject of interest. Using three datasets Rent prices, Voltage Drop, and Nigeria’s Unemployment Rate. This study assesses model fit based on the Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and Mean Squared Error (MSE). Findings reveal that the standard Laplace (L) distribution consistently outperforms its counterparts. In the Rent dataset, it achieves the lowest AIC (613.636), BIC (609.2266), and a reasonable MSE (2343.761), whereas the TL and AL distributions yield significantly higher AIC and BIC values, and the ASL distribution demonstrates an extremely high MSE (9.34 × 10¹²), indicating poor fit. A similar trend is observed in the Voltage Drop dataset, where the L distribution records the lowest AIC (201.1564), BIC (197.7293), and MSE (132.7978), while TL and ASL show excessive model instability. In the Unemployment Rate dataset, the L distribution again provides the best fit, with an AIC of 349.7985, a BIC of 345.896, and a moderate MSE of 186.4666. On average, across all datasets, the L distribution remains the most robust model, with the lowest AIC (388.197), BIC (384.284), and MSE (887.6751). The AL distribution follows closely with an MSE of 888.9518 but exhibits significantly higher AIC (2426.027) and BIC (2424.071). The ASL distribution, while demonstrating moderate AIC (1443.016) and BIC (1448.885), suffers from poor predictive accuracy with an extremely high MSE (3.19E+12). The TL distribution performs the worst, with the highest AIC (34,686.77), BIC (20,112.08), and an MSE of 76,038.22, highlighting its instability. In conclusion, this study establishes that the standard Laplace (L) distribution provides the most reliable and accurate fit across diverse datasets. While alternative forms introduce additional flexibility, their increased complexity does not necessarily yield superior model performance. Future research should explore modifications to improve the parameter stability of Laplace extensions and investigate alternative estimation techniques to enhance predictive accuracy in real-world applications.
| 8 |
Author(s):
Owuamanam M. C., Nwawuike I. M..
Page No : 94-114
|
Temporal Dynamics of Soil Chemical Properties During Incubation with Biochar from Diverse Biomass Sources.
Abstract
Soil chemical properties degradation is a very serious challenge facing food production in the tropics. To ensure sustainability, there is need for soil improvement. This study aimed at investigating the temporal dynamics of soil chemical properties during incubation with biochar from diverse biomass sources. An incubation experiment was carried out at the Soil Science Laboratory Faculty of Agriculture, Imo State University. The soil used was for the experiment was taken at the depth from 0 - 20 cm from Imo State Teaching and Research farm Owerri. The collected soil was air-dried, crushed, and sieved using 2 mm sieve and characterized before treatments application. Three types of biomass used for biochar production were saw dust, poultry manure and pig dung. The biomasses selected were air-dried before pyrolysis. Prior to the insertion of the biomasses, the biochar machine was first be heated for 10 minutes. 1kg of each agricultural wastes residue was inserted into the modified gas biochar kiln and charred at 300oC for 60 minutes. 500g soil each was placed in plastic container with a lid. The soil was mixed with 5g of the produced biochar each at the rate of 20t ha-1. The detailed treatments used for this experiment were as follows: T1: Saw dust biochar (20 t ha-1), T2: Poultry manure biochar (20 t ha−1), T3: Pig dung biochar (20 t ha−1), T4: 10 t ha−1 of Saw dust biochar + 10 t ha−1 of Poultry manure biochar, T5: 10 t ha−1 of Saw dust biochar + 10 t ha−1 of Pig dung biochar, T6: Control (No biochar amendment). 200ml of deionized water was added for three consecutive days to ensure field capacity is obtained. The containers were cover with lids and strongly tied with black waterproof to maintain dark condition. The incubation experiment was arranged in completely randomized design. Soil samples were collected on days 30, 45, 60 and 75 to analyze its chemical properties The result of the chemical composition of the produced biochar showed that all the produced biochars had an elevated pH and high exchangeable cations with its highest on PMB. The carbon content of the animal based biochars (PMB and PDB) was higher than that found on the plant based biochar (SDB). The nitrogen content followed a reverse trend as seen in carbon content. The results of the biochar application on the chemical properties of the soil showed increase in pH, SOM, TN, Avail P, Exch. Cations and Exch. H with a reduction in Exch. Al. The highest impact on pH, Avail P, Ca, Mg, K and CEC were on soils treated with PMB @ 20t/ha. Soils with SDB @ 20t/ha gave the highest SOM but its combination with PMB (SDB @10t/ha and PMB @10t/ha) gave the highest TN with control exhibiting higher Exch. Al. Despite the increase observed, dynamic changes were observed across the incubation intervals in all the evaluated soil chemical properties.The interaction between biochar from different biomass sources and incubation time underscore the importance of tailoring biochar use to specific biomass and environmental conditions. PMB@20t/ha stand out as the only produced biochar in this study, which not just impact positively on almost all the soil chemical properties evaluated but also maintain a long-lasting effect on the soil. Although, the results from the study showed a clear effect of biochar on soil chemical properties, conducting comprehensive field trials is recommended to ensure that local farmers benefit from this research finding.
| 9 |
Author(s):
Idris Aliyu Kankara.
Page No : 115-128
|
Artificial Neural Networks for Detection of Diabetes Mellitus by Selecting Significant Risk Factors Using Backward Stepwise Feature Selection Method.
Abstract
Diabetes Mellitus (DM) is a chronic disorder affecting more than 300 million people world-wide and it needs urgent attention. Selecting of significant risk factors (SRFs) and their contributions to the risk of the disease is the key for early detection of the disease. The aim of this paper is to use Backward Stepwise Feature Selection Method (BSFSM) to select the SRFs and their contribution to the risk of DM and Kappa statistic value (KSV) to evaluate the model performance. Dataset consists of 400 patients with demographic, clinical, lifestyle and dietary risk factors were collected from General Hospital Kaura Namoda, Zamfara State, Nigeria from 2019 to 2023 by checking file of patients suffering from DM. The results obtained revealed that BSFSM retained twelve (12) SRFs namely Blood Glucose level (BGL), High Body Mass Index (BMI), Family History of DM (FHDM), Preference for Sweet Food (PSWF), Age, Lack of Physical Activity (LPA), Blood Pressure (BP), Red Meat (RM), Refined Carbs (RC), Energy Drink (ED), White Rice (WR) and Processed Meat (PM) and remove two (2) Non-SRFs Sex and Preference for Salty Food (PSF). The SRFs contributed 85.40%, 51.34%, 55.72%, 68.23%, 57.50% 29.96% , 66.18%, 41.42%, 12.20%, 18.65%, 29.76% and 10.11% to the risk of DM respectively. Similarly, the Non-SRFs contributed 0.98% and 1.16% to the risk of DM. The MLP model detected 98.6% DM patients in the training set, 96.3% in the validation set and 92.9% for test set. 97.8% Non-DM patients in the training set, 93.9% in the validation set and 93.8% for the test set. The KSV of the model was 0.94 and it was capable of distinguishing between DM and Non-DM patients. This paper demonstrated that BSFSM was capable of selecting the SRFs and their contributions and KSV adequately evaluate the performance of the model
