Uber's Two-Tower model for Uber Eats restaurant recommendations | Eric Seufert posted on the topic | LinkedIn

This article was written by a Hotel Marketing Flipboard. Click here to read the original article

🚀 Excited to Share My Uber Ride Cancellation Prediction Project!
During my training at (National Telecommunication Institute (NTI)), I had the opportunity to work on a Machine Learning project that predicts whether an Uber ride will be canceled or completed based on different factors.
First, I’d like to sincerely thank my instructor Abdulrahman Bakeer for the constant support, valuable feedback, and clear guidance throughout this project 🙏.
I also want to thank my amazing friends and teammates who worked with me — this project wouldn’t have been possible without your collaboration and effort 💪.
@Youssef Khaled
Mohamed Ashraf
🧠 Project Overview
Our project, “Uber Ride Cancellation Prediction Model,” aims to help ride-sharing platforms like Uber minimize cancellations by predicting them early using data-driven insights.
We analyzed and preprocessed real-world ride data containing key features such as:
Booking Value 💵
Trip Distance 📏
Customer and Driver Ratings ⭐
Vehicle Type 🚗
Payment Method 💳
Time of the Day & Weekend Indicator ⏰
⚙️ Technical Details
We performed Exploratory Data Analysis (EDA) to detect trends and correlations (for example, cancellations were higher during weekends and at peak hours).
We handled missing values, removed outliers, and encoded categorical variables for model training.
Several models were trained and compared:
Logistic Regression
Decision Tree
Random Forest
XGBoost
The Random Forest Classifier achieved the best performance, with high accuracy and balanced precision-recall results.
💻 Deployment
To make the model accessible and interactive, we built a Streamlit web app.
Users can input ride details (like distance, booking value, driver rating, etc.) and instantly see whether the model predicts the ride will be “Canceled” or “Completed.”
We also included:
A dynamic background 🎨
Visual feedback with charts 📊
PDF report generation of predictions 📄
🎯 Key Learnings
This project strengthened my skills in:
Data preprocessing & feature engineering
Model evaluation & optimization
Streamlit app development
End-to-end machine learning deployment
I’m really proud of how far we’ve come as a team, and I look forward to applying these skills in real-world data projects and future challenges in the data science and AI field.
hashtag#NTI hashtag#MachineLearning hashtag#DataScience hashtag#Python hashtag#Streamlit hashtag#Uber hashtag#PredictionModel hashtag#EDA hashtag#XGBoost hashtag#RandomForest hashtag#Project hashtag#AI hashtag#Teamwork

Trending

Park Hyatt Celebrates Over Four Decades of Personalized Luxury with Landmark Global Brand Expansion Plans on the Horizon and Compelling Campaign

…more