picture collection
Cynthia Uwakwe is an articulated research writer that specializes on writing SEO articles and research technical papers, currently working with Beloga blogs on articles censored around entrepreneurship and business analysis. Focused on solving problems and analysing research works, creating hypotheses and meeting the needs and collective goal of team members and organizations through her meticulous analysis, detailing, great communication and organising skill which has helped in gaining more audience and content visibility from search engine by infusing keywords and information analysis which has increased the blogs traffic and search for relevant contents. This has boost the organisation portfolio and high demand increase in relatable contents.
My Elevator Pitch
My article on religion extremism
Chronic kidney Disease is strongly influenced by age, diabetes, hypertensions and kidney markers such as Glomerula filteration rate(GFR), creatine and blood urea nitrogen(BUN).
The dataset contains 5000 simulated with Kidney markers, cormobid conditions, medications categories and age_group.
i analysed CKD prevalenceusing clinically meaningful aggregation, examined age_related risk patterns, evaluated cormorbid conditions and explored relationships between kidney markers and CKD status.
CKD prevalence increases with age and is higher among individuals with diabetes and hypertension, and is associated with reduced GFR and elevated creatine level. Medication and hydration pattern also show variation across age group.
A Pitch video of our Digital platform to the srespective stakeholder.
This project explores the famous Titanic dataset from Kaggle and applies machine learning techniques to predict passenger survival. The goal was to classify whether a passenger survived or did not survive based on characteristics such as age, gender, passenger class, and fare.
This project represents my first complete end-to-end machine learning workflow — from data cleaning and preprocessing to model training, evaluation, and competition submission.
Can we predict whether a passenger survived the Titanic disaster using available passenger information?
Dataset Source: Kaggle Titanic Competition
Files used:
train.csv
test.csv
The training dataset included survival labels, while the test dataset was used for final predictions.
Handled missing values in Age and Embarked
Dropped columns with limited usefulness or excessive missing values:
Name
Cabin
Ticket
Converted categorical variables into numerical values:
Sex → encoded to numerical values
Embarked → encoded to numerical values
Logistic Regression using scikit-learn
Data preprocessing
Train-validation split
Model training
Prediction
Performance evaluation
Kaggle submission generation
Validation Accuracy: 0.79
Kaggle Leaderboard Score: 0.75
The model correctly predicted survival outcomes for a large portion of validation data and produced a competitive beginner baseline score.
During this project I encountered and resolved:
Object datatype errors
Feature encoding issues
File path and dataset loading challenges
Submission formatting errors
These debugging experiences improved my understanding of real-world machine learning workflows.
Machine learning models require numerical input
Data preprocessing significantly impacts model performance
Validation scores may differ from leaderboard scores
Debugging is a critical part of the ML process
Add FamilySize feature
Extract passenger titles (Mr, Mrs, Miss)
Experiment with feature scaling
Test additional algorithms such as Random Forest
Python
Pandas
scikit-learn
Jupyter Notebook
Kaggle
GitHub
Cynthia Uwakwe
Aspiring Data Analyst | Data Science Learner | Content Writer