Daily run sheet is not daily manual work anymore
summary
Describes an automated system for generating daily run sheets from messy Indian customer address free-text using a three-step algorithm: address structuring (custom cleaning and training because standard English NLP stopwords/tools aren’t sufficient), address identification via a trained model, and clustering of identified addresses to produce delivery groupings. Mentions practical challenges in cleaning/training address data and selecting algorithms for address identification before clustering.
AI-written summary · about
chapters (6)
- 0:00Start
- 7:01Address Structuring-How?
- 13:53Named Entity Recognition
- 19:22Geo-coding Geocode
- 20:44Clustering
- 23:34Shoot your questions!
related talks
24mMachine learning techniques for data cleaning
48mCleaning Messy Data at Scale: APIs, LLMs, and Custom NLP Pipelines
02h 00mData Science life cycle from a technical perspective
19mUnique Challenges and Opportunities in Working with Residential Real Estate Data
27mScratch, Dent and Data: Engineering A Used Appliance Information Advantage
43mImplementing AI Document Processing at Scale: Syngenta
46mRobotization of Unstructured Data Processing
32mAutomate customer ticket handling: from an idea to a production ready model