From Algorithms to Assets: Data Science Meets Real Estate
summary
Describes building automated valuation models (AVMs) that use historical transaction and listing data with a market‑approach (not forecasting) to price properties at a point in time; explains how US iBuyer programs (Zillow, Redfin, Opendoor, Offerpad) scaled purchases, why Zillow’s 2021 misstep leads to a portfolio glut and ~2,000 layoffs, and why data quality, transparency, pipelines, regulation and market maturity matter; proposes recommendation systems, LLMs for listings/management and graph neural networks for geospatial transaction graphs.
AI-written summary · about
chapters (12)
- 0:00Data science in real estate
- 1:32Introduction
- 3:21I buyer program and automated valuation models
- 8:54Zillow’s AVM failure
- 10:38How AVMs work
- 12:28Recommendation systems
- 14:56LLMs for real estate agents
- 16:19AI adoption challenges
- 21:03Data governance
- 22:06Regulation and standardization of AVMs
- 23:32Emerging AI techniques
- 24:15Q&A
related talks
16mUnique Challenges and Opportunities in Working with Residential Real Estate Data
32mData as the Driving Force of Real Estate Industry
45mInnovating with Data
49mBuilding Machine Learning Pipeline Using Aerosolve
19mUnique Challenges and Opportunities in Working with Residential Real Estate Data
01h 01mYou don't need AI; You need to know what your data is.
35mBuilding Data-Intensive Teams
39mLessons Learned: Insights from 30 Years in Cloud and AI/ML Integration