Natural Language Processing and Hadoop
summary
Shows how to run NLP at scale by pairing classic pipeline steps — tokenization, parsing, feature extraction and model training — with Hadoop to process terabyte corpora, arguing more data beats fancier algorithms. Demonstrates concrete workflows to break down raw text, extract named entities and signals that feed search and recommender systems, and points out current systems are mainly information extraction vs semantic understanding, with a roadmap to move toward deeper language models.
AI-written summary · about
related talks
47mSolving Real & Big (Data) Problems Using Hadoop
29mThe Data Journey
21mMeeting complex data load and data preparation challenges
59mPolyglot Pipelines with Apache Nifi
09mMatthew Russell interviewed at OSCON 2010
56mResistance is Futile: What You Need to Know about Big Data
44mReal World Use Cases Hadoop and NoSQL in Production
53mBig Data: from mammoth to elephant