Vectorized Finite State Automata

András Kornai (IBM Almaden Research Center)

We present a technique of finite state parsing based on vectorization and describe the application of this technique to a well-known problem of natural language processing, that of extracting {\it relational information} from English text. We define Vectorized Finite State Automata, the theoretical model behind the applied system, and discuss their significance.


PS version (6 pages, 152k)

PDF version (6 pages, 176k)