Data Analysis and Software Developer

From scattered numbers to analytical data

Created 12 statistical reports and a presentation based on predictive models using Hadoop and MySQL databases to improve financial management for the upcoming year.

From numbers to visuals

Purveyed two years' worth of monthly financial reports and developed a predictive model focusing on deficits, earnings, and losses. The data was then visualized using Tableau’s dashboards.

From presentations to present-day changes

Presented the findings, identifying causes of certain losses and deficits, and proposed two alternative decision-making routes using predictive models, which improved the accountancy’s earnings by 15% .

What the trends tell us

Developed analytical models to track social media trends for a campaign addressing social media addiction among children aged 9-13.

How to analyze the trends

Analyzed data from three mobile social media platforms to monitor trend popularity over a month utilizing Django to store data.

What we did with the data

Created data visualizations using Tableau that were integral to an educational packet distributed to over 20 schools, impacting approximately 700 classrooms.

Real-Time Analysis of Network Traffic

Engineered a comprehensive system, integrating a full-stack framework, database, and machine learning algorithms, to efficiently analyze real-time network traffic.

Efficient means of analyzing ports

Leveraged NMAP and Wireshark for in-depth network analysis, identifying open ports and their associated URLs. Applied advanced machine learning models, including Decision Trees, K-Means clustering, and Convolutional Neural Networks (CNNs), to detect and filter malicious URLs.

Cyber Threats Detected before they could act!

Achieved 99% accuracy in cyber threat detection using Decision Tree models, significantly improving threat detection, anomaly prevention, and network latency reduction.

Demonstrates what computer vision can achieve

The Image Repair and Restoration Machine Learning project was published and presented at the AIST 2022 Conference which was dedicated to the field of Artificial Intelligence and my paper was nominated for the Best Presentation

Debuted at a Professional IEEE-affiliated Conference

It debuted during the Computer Images Panel and was a part of the series dedicated to showcasing the diversity of research featured in the fields of computer graphics, image processing, and computer vision

Read it now

The publication can be read at this link: Read more

Datasets

Employing Python libraries, Cancer Imaging Archives datasets, DICOM, and advanced deep learning methodologies, I developed and deployed two proficient deep learning models for precisely detecting brain tumors in MRI scans.

The models

Multiple deep learning models, including a Convolutional Neural Network (CNN) and a Support Vector Machine (SVM), underwent comprehensive training and testing with diverse variants to achieve optimal performance.

Computer Vision in Action

By operationalizing cutting-edge computer vision and machine-learning techniques, a robust neural network was created, exhibiting exceptional proficiency in accurately identifying anomalies within a dataset comprising 1000 images.

Libraries without end

Developed a Personal Library app using the OpenLibrary API to build a personal database of books. Implemented data retrieval and integration techniques to ensure accurate and up-to-date information from external sources.

Tables instead of shelves

Designed and optimized the app using MySQL for database management and Flask for the backend framework. Conducted performance testing to improve speed and efficiency, increasing productivity for personal book database management by 20% through query optimization and efficient data storage practices.

To view this continuously updated projectedt

The General Design and Coding Architecture can be found on my github