Keystroke dynamics as signal for shallow syntactic parsing
This work addresses the challenge of enhancing natural language processing models for tasks like chunking and CCG supertagging, though it is incremental as it builds on existing multi-task learning methods.
The authors tackled the problem of improving shallow syntactic parsing by using keystroke dynamics as an auxiliary signal, resulting in models that significantly outperformed those trained on text annotations alone.
Keystroke dynamics have been extensively used in psycholinguistic and writing research to gain insights into cognitive processing. But do keystroke logs contain actual signal that can be used to learn better natural language processing models? We postulate that keystroke dynamics contain information about syntactic structure that can inform shallow syntactic parsing. To test this hypothesis, we explore labels derived from keystroke logs as auxiliary task in a multi-task bidirectional Long Short-Term Memory (bi-LSTM). Our results show promising results on two shallow syntactic parsing tasks, chunking and CCG supertagging. Our model is simple, has the advantage that data can come from distinct sources, and produces models that are significantly better than models trained on the text annotations alone.