Time Series Forecasting Projects – Research-oriented Python portfolio exploring classical and modern forecasting techniques, from Prophet to LightGBM, with EDA, feature engineering, and model evaluation on real-world datasets.

A program to analyze crime and sports data in Philadelphia to determine if a correlation exists between whether a Philly team's win/loss to a given team has an effect on the city's violent crime rate. Conclusion--> If the Flyer's lose to the Redwings, stay out of Center City.

This project shows best practices for optimizing memory in Syncfusion JavaScript Pivot Table using virtual scrolling, paging, disposal, and server-side processing.

An analysis of the Home advantage in NBA games

Dealing with 100000000 rows of data in a table.

A modern, flexible and scalable Geographic Information System (GIS) designed for creating and editing geographic data, developed for seamless integration into web applications.

A web scraper designed to collect and analyze public resumes for research purposes.

A professional, web-based application designed specifically for handling and analyzing large CSV and text files that are too massive for conventional spreadsheet software. Built with Python and Flask, it provides a robust, browser-based interface for searching, previewing, and exporting data from files that can exceed several gigabytes in size.

Financial dataset flat file reconstruction, applying data structuring, ETL, and SQL RDBMS best practices.

Efficient Serving of Large-scale Vector Search with Sharded Indexes

OpenExcelLite is a lightweight, dependency-free Excel (XLSX) generation library built on the official Open XML SDK. It focuses on speed, schema correctness, streaming large datasets, and clean, easy-to-use APIs — without the heavy overhead of ClosedXML, EPPlus, or NPOI.

Ensemble Deep Random Vector Functional Link Neural Network Based on Fuzzy Inference System (Interpretable Tabular Model) - IEEE TFS 2025

This project used SparkSQL and PySpark to analyze home sales data, optimizing performance with caching and partitioning by build date.

Sensor monitoring and dashboarding platform for MATLAB and GNU Octave. Plot 100M+ points at 200+ FPS, define state-dependent thresholds, detect events in real time, and build interactive dashboards.

This repository contain the sample that proves that the Syncfusion Pivot table is capable of loading and displaying a million records instantly and perform scrolling without any performance lag.