Information Extraction with Character-level Neural Networks and Free Noisy Supervision
This work improves information extraction for financial language processing at Bloomberg, though it appears incremental as it builds on an existing parser.
The authors tackled information extraction from financial text by augmenting an existing parser with a character-level neural network trained using noisy supervision from database consistency, which led to large improvements over Bloomberg's mature production system.
We present an architecture for information extraction from text that augments an existing parser with a character-level neural network. The network is trained using a measure of consistency of extracted data with existing databases as a form of noisy supervision. Our architecture combines the ability of constraint-based information extraction systems to easily incorporate domain knowledge and constraints with the ability of deep neural networks to leverage large amounts of data to learn complex features. Boosting the existing parser's precision, the system led to large improvements over a mature and highly tuned constraint-based production information extraction system used at Bloomberg for financial language text.