Emmanuel Chibuogu Asogwa

This study applies machine learning to investigate gendered power dynamics and women's socio-economic and political engagement in Nigeria (1991-2023, World Bank, UNDP, INEC data). We trained Random Forest, Support Vector Machine (SVM), and Neural Network models with k-fold cross-validation, evaluating performance with R2, RMSE, and MSE. The SVM model demonstrated superior performance (R2 = 0.998). Feature analysis revealed that women's industry participation positively correlates with population share and education, while rural residence diminishes their likelihood of being employers. Additionally, K-means clustering of 2023 voting data uncovered regional variations in women's political representation. This research highlights enduring socio-economic and spatial barriers, demonstrating how AI-based evidence can inform gender-sensitive policies for inclusive representation in Nigeria.

Breaking away from traditional methods, this study investigates the asymmetric effects of macroeconomic shocks on economic growth in Nigeria using a novel Machine Learning-Augmented Nonlinear Autoregressive Distributed Lag (NARDL) model. We utilized a Random Forest algorithm to data-driven feature selection, thus model optimization and its enhanced robustness against random selection of lags. The results confirm the presence of a long-run cointegrating relationship and show that positive and negative oil price shocks statistically differ in their effects on GDP. We also find inflation to have a strong negative long-run effect, and government capital expenditure is a significant driver of growth. Such embedding of machine learning in the NARDL model is a more empirically valid policy analysis tool that supplies key findings for policymakers in resource-dependent economies.

This study evaluates a hybrid model that integrates Long Short-Term Memory (LSTM) networks and Convolutional Neural Networks (CNN) to predict stock prices. The model leverages two datasets: historical Google stock data and sentiment data from Reddit comments. Sentiment analysis was performed using VADER from NLTK, which classified comments as negative, neutral, or positive, while a CNN model was trained to predict sentiment scores. Separately, an LSTM model was built using ten years of Google stock data from Yahoo Finance, with features scaled using MinMax normalization to improve learning and a dropout layer added to prevent overfitting. Model performance was evaluated using Root Mean Squared Error (RMSE) and Mean Squared Error (MSE). The LSTM model performed well on test data but showed lower accuracy on unseen data during forecasting. The hybrid model successfully combined the outputs of both the CNN and LSTM, demonstrating superior performance with lower RMSE and higher classification accuracy compared to the standalone models. This highlights the potential of integrating sentiment analysis with traditional stock prediction. The study acknowledges challenges in classifying neutral sentiments, suggesting that more comprehensive sentiment data is needed for future research.