Large-Scale Time-Series Trend & Anomaly Tracker (1M+ Rows)

Project Description

Project Title: High-Volume Telecom Network Anomaly & Trend Tracker (1M+ Rows)

The Challenge: Analyzing long-term network downtime and subscriber trend data across millions of rows was exceeding Excel's memory limits, causing frequent crashes.

The Solution (Python Pandas + Streamlit/Excel): Processed 1M+ rows using Python’s optimized memory management to detect downtime anomalies and churn trends. The aggregated, lightweight summary was then exported directly to an interactive Excel Pivot & Dynamic Chart interface.

The Impact: Enabled seamless analysis of 1M+ row datasets without memory overhead, detecting 15% more network downtime spikes proactively.