From Logical to Distributional Models
This work addresses the integration of symbolic and statistical approaches in natural language processing, but it appears incremental as it builds on existing models without clear empirical validation.
The paper tackled the problem of connecting logical and distributional semantic models for natural language by transferring logic and reasoning from logical to distributional models, resulting in a method to compare sentence meanings word by word through vector operations.
The paper relates two variants of semantic models for natural language, logical functional models and compositional distributional vector space models, by transferring the logic and reasoning from the logical to the distributional models. The geometrical operations of quantum logic are reformulated as algebraic operations on vectors. A map from functional models to vector space models makes it possible to compare the meaning of sentences word by word.