Zaidibeth Ramos Software Engineer

Hierarchical ACO-Firefly Hybridization for Genomic Feature Selection in Breast Cancer

Universidad Centroccidental Lisandro Alvarado

Feb 2026 - Apr 2026 (3 months)

Overview: Designed and implemented a novel hierarchical computational approach to solve the "curse of dimensionality" in high-dimensional oncology genomic data. By hybridizing Ant Colony Optimization (ACO) and the Firefly algorithm, we created an advanced feature selection pipeline capable of isolating critical breast cancer biomarkers from massive genomic datasets. Architecture & Execution: Pipeline Design: Engineered a two-stage evaluation system: a statistical pre-filtering layer utilizing Mutual Information, followed by a robust Wrapper evaluation layer. Swarm Orchestration: Utilized ACO for global exploration to identify promising genetic subspaces, while deploying the Firefly algorithm for aggressive local exploitation to prune red

SOM Neural Network for Bank Customer Segmentation

Universidad Centroccidental Lisandro Alvarado

Jan 2026 - Mar 2026 (3 months)

This analysis addresses the problem of unsupervised bank customer segmentation using a Self-Organizing Map (SOM), also known as a Kohonen map, applied to the high-dimensional Bank Marketing dataset. The main objective was to project 29 customer features (derived from 12 original variables like age, balance, and marital status) onto a 10x10 grid to reveal natural clusters and inform strategic marketing decisions. The SOM model successfully converged, reducing the quantification error by 29.12% over 200 epochs. The study identified five distinct customer profiles, including: young, highly educated singles; married clients with high financial burden (mortgages/loans); and, most strategically, a segment of clients with a historically successful

R

RBF Neural Network for S&P 500 Stock Closing Price Prediction

Dec 2025 - Feb 2026 (3 months)

The project focuses on designing and implementing a Radial Basis Function (RBF) Neural Network for the prediction of the daily closing price of an S&P 500 stock (specifically, Apple Inc. or AAPL). Objective and Methodology: The work addresses a supervised regression problem using five input variables (Open, High, Low, Close, and Volume) to predict the next day's closing price. The RBF Network was chosen for its capacity to model non-linear relationships, utilizing a two-phase training process: non-supervised (K-Means to determine 50 centers) and supervised (weight adjustment using the Adam optimizer). Results and Conclusion: The convergence analysis showed severe overfitting (the validation error was nearly 8 times greater than the final tr

I

Implementation and Evaluation of a Simple Binary Genetic Algorithm on Six-Hump Camel Back and Schwefel Functions

Nov 2025 - Dec 2025 (2 months)

F

Fuzzy Controller for Weather Prediction

Oct 2025 - Nov 2025 (2 months)

The document details a Fuzzy Controller for Weather Prediction that uses fuzzy logic to model and predict two key variables: the change in temperature (Delta T) and the level of precipitation (P). The system relies on three linguistic input variables (Temperature, Humidity, and Wind Speed) and employs the Mamdani inference method (Min-Max) and Centroid defuzzification. The rule set is justified by three underlying physical principles: Physical Block: Ensures no precipitation occurs if humidity is low or normal. Latent Cooling: Models temperature drop and intense precipitation under conditions of saturation and strong wind. Thermal Inertia: Assumes the weather will remain stable in the absence of significant perturbations (Persistence Method

Application of Basic Neural Computation Models: Perceptron and ADALINE

Universidad Centroccidental Lisandro Alvarado

Jun 2025 - Jul 2025 (2 months)

This work involves the implementation, training, and evaluation of two fundamental neural network models: the Perceptron and the Adaptive Linear Neural Network (ADALINE). The objective was to analyze the performance of these models in pattern recognition and time series forecasting tasks. The Perceptron used a single-layer architecture to classify numerical patterns, with its convergence and robustness to noise evaluated using different learning rates. ADALINE, a linear model, was implemented to predict values in a time series, similarly analyzing the impact of learning rate and noise. Results indicate that the Perceptron perfectly classifies linearly separable patterns without noise, but its performance severely degrades (accuracy ≈19%) wi

A

Application of Cellular Automata in Forest Fire Propagation

May 2025 - Jul 2025 (3 months)

Mobile Robots: An Easy Way to Program Robotic Vacuums

Universidad Centroccidental Lisandro Alvarado

Apr 2025 - Jun 2025 (3 months)

This report details the design, implementation, and evaluation of two behavior-based control algorithms—reactive control and spatial knowledge control—for a mobile robot tackling the "Robot Vacuum" challenge on the VEXcode VR platform. The study compared a purely reactive algorithm, which proved simple and robust for basic evasion, against a hybrid deliberative-reactive algorithm incorporating a simple form of spatial and temporal knowledge. Experimental results showed that while the reactive algorithm achieved 67% coverage, the spatial knowledge algorithm reached 82% coverage, representing a 22.4% improvement in efficiency and a 47% reduction in revisiting previously explored areas. The conclusion is that the hybrid deliberative-reactive a

Zaidibeth Ramos

About

Detail

Timeline

Résumé