Project: Predict Estimated Time Up/Down and Leave Base Fine Prediction

From CLIENTOCLARIFY.AI PVT LTD (Jun 2019 - Dec 2021) [cite: 235, 238]

Business Problem Statement

The challenge involved predicting up/downtime based on fine prediction processes and providing recommendations for visitors at TTD, utilizing real-time data from local on-premises APIs. [cite: 246, 247, 251]

Solution

• Collected 6.7TB of valid real-time data from TTD local on-premises data API to identify the most visited cars and persons. [cite: 246]
• Predicted up/downtime based on the fine prediction process with integration with the TTD dashboard. [cite: 247]
• Developed code for Geo-distance with IoT Sensors (used to find the shortest distance between two points in Ellipsoidal Plane). [cite: 248]
• Built Artificial Neural Networks on the data and achieved an accuracy of 90% from 65%. [cite: 249]
• Implemented Predictive Models using Ensemble methods like XGBoost, Random Forest. [cite: 250]
• Provided recommendations to people regarding the nearest place to visit, parking, rooms, etc. [cite: 251]

Tools & Frameworks Used

• Artificial Neural Networks [cite: 249]
• Ensemble methods (XGBoost, Random Forest) [cite: 250]
• IoT Sensors [cite: 248]
• TTD dashboard (integration) [cite: 247]
• Python [cite: 338]
• NumPy, Scikit-learn, TensorFlow, PyTorch, Keras (general AI/ML libraries) [cite: 295, 351]