Authors: Venkatesh C, Harshit Oberoi, Anil Goyal, Nikhil Sikka

Published on: April 25, 2024

Impact Score: 7.6

Arxiv code: Arxiv:2404.16553

Summary

• What is new: A real-estate recommendation system, RE-RecSys, that categorizes users into distinct categories for personalized property suggestions, with a novel weighing scheme for conversion rates.
• Why this is important: Improving property recommendations for different types of users in the real estate industry.
• What the research proposes: A multi-faceted recommendation system that tailors property suggestions based on user categorization and interaction history.
• Results: Deployable in real-world settings with an average latency of 40 ms serving 1000 rpm, demonstrating efficiency and scalability.

Technical Details

Technological frameworks used: End-to-end real estate recommendation system

Models used: Rule-based engine, content-filtering, combination of content and collaborative filtering

Data used: Real-world property and clickstream dataset from a leading real-estate platform in India

Potential Impact

Real estate platforms and companies looking to enhance user experience and engagement through tailored property recommendations

Want to implement this idea in a business?

We have generated a startup concept here: PropTune.