A Comparative Analysis of Knowledge-Intensive and Data-Intensive Semantic Parsers
This provides insights for semantic parsing researchers by systematically comparing two dominant approaches and identifying complementary error patterns.
The authors compared knowledge-intensive and data-intensive semantic parsers, finding they achieve comparable overall performance but make different types of errors that align with their theoretical properties, leading to new development directions.
We present a phenomenon-oriented comparative analysis of the two dominant approaches in task-independent semantic parsing: classic, knowledge-intensive and neural, data-intensive models. To reflect state-of-the-art neural NLP technologies, we introduce a new target structure-centric parser that can produce semantic graphs much more accurately than previous data-driven parsers. We then show that, in spite of comparable performance overall, knowledge- and data-intensive models produce different types of errors, in a way that can be explained by their theoretical properties. This analysis leads to new directions for parser development.