Data Scientist skilled in Python, Pandas for Data Analysis, Numpy, Scipy, Matplotlib, Power-BI, DAX, SQL, Scikit-learn, Machine Learning, Deep
Learning, Git & GitHub
@AbdulmalikAdeyemo
In this project, I investigated and analyzed absenteeism trends in a medium-sized company using Power BI and presented findings and recommendations to stakeholders.
This project demonstrates my ability to leverage data analytics tools and techniques to identify key insights and drive meaningful improvements in organizational performance.
In this project, I identified irregularities in the general ledger of a large company. Using a data-driven approach,
I applied analytical techniques (using Pandas) to detect outliers and anomalies in the journal entries posted to the ledger.
To do this, I performed a rule-based analysis that focused on identifying journal postings that did not net to zero,
which signaled that they were not business as usual.
By summing all the Amount column entries that belonged to a particular BAT_NAME ID and extracting all the journal postings whose amount did not net to zero,
I was able to identify anomalous transactions that required further investigation.
In this project, I implemented an anomaly detection algorithm to detect anomalous behavior in server on a network using Gaussian distribution.
The goal of this project is to create a robust anomaly detection algorithm that can accurately identify anomalous server behavior,
which can help prevent system failures and improve the overall performance of the servers.
In this project, I built 2 deep learning models, a binary-class ResNet-50 model trained on 85,056 fundus images to detect the presence or absence of diabetic retinopathy,
and a multi-class ResNet-18 model trained on 44,224 fundus images to grade the severity level diabetic retinopathy.
The goal of this project is to build an AI software that allows more patients to be diagnosed in less time.
In this project, I led a team of 20 interns in the development, evaluation, and deployment of a dementia prediction model based on a tabular dataset of 150 subjects aged 60 to 96.