AMAP-ML

AMAP-ML / IntTravel

Public

IntTravel: A Real-World Dataset and Generative Framework for Integrated Multi-Task Travel Recommendation

55
0
100% credibility
Found Feb 13, 2026 at 23 stars 2x -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

IntTravel offers a massive real-world dataset of travel interactions and user-friendly tools to prepare it for building multi-task recommendation systems.

How It Works

1
🗺️ Discover IntTravel

You stumble upon a huge collection of real-world travel stories from millions of people exploring cities.

2
📥 Gather travel data

You collect simple files with user profiles, places of interest, and their actions like visits and routes.

3
✨ Organize into smart sequences

With easy steps, you turn messy travel logs into neat patterns ready for recommendation magic.

4
🔧 Polish the insights

You refine the organized data to highlight key details like locations, weather, and preferences.

🚀 Power up travel suggestions

Now your system can generate personalized travel ideas, just like the one serving millions daily!

Sign up to see the full architecture

3 more

Sign Up Free

Star Growth

See how this repo grew from 23 to 55 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is IntTravel?

IntTravel delivers a massive real-world dataset with 4.1 billion interactions from 163 million users and 7.3 million POIs, pulled from a major Chinese navigation service, plus a Python-based data processing pipeline to prep it for generative multi-task travel recommendation models. Developers get scripts to transform raw user actions, profiles, and POI data into sequenced inputs ready for decoder-only models handling tasks like POI prediction, travel mode, and route planning. It solves the scarcity of integrated travel datasets by enabling end-to-end training on production-scale data.

Why is it gaining traction?

Unlike sparse recsys benchmarks, IntTravel offers authentic, geo-aware interactions with weather, timestamps, and via-points, fueling multi-task generative frameworks that scale to 80 layers without degradation. The pipeline handles negative sampling and label inference out-of-the-box, outputting model-ready CSVs for quick experimentation. Its Amap deployment proves real-world viability, drawing devs chasing integrated recommendation over siloed tasks.

Who should use this?

ML engineers at mapping or travel apps like Amap clones building next-POI predictors or route optimizers. Recsys researchers testing generative multi-task setups on huge datasets. Python teams needing a baseline for location-based rec without fabricating data.

Verdict

Grab the Hugging Face dataset and run the two-step Python pipeline if you're in travel recommendation—it's a rare real-world gem despite 19 stars and 1.0% credibility signaling early maturity. Docs cover usage, but expect tweaks for custom scales; solid for prototypes, watch for model code next.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.