Joint Slot Filling and Intent Detection via Capsule Neural Networks

ACL 2019 Chenwei ZhangYaliang LiNan DuWei FanPhilip S. Yu

Being able to recognize words as slots and detect the intent of an utterance has been a keen issue in natural language understanding. The existing works either treat slot filling and intent detection separately in a pipeline manner, or adopt joint models which sequentially label slots while summarizing the utterance-level intent without explicitly preserving the hierarchical relationship among words, slots, and intents... (read more)

PDF Abstract

Evaluation Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK COMPARE
Intent Detection ATIS Capsule-NLU Accuracy 0.950 # 2
Slot Filling ATIS Capsule-NLU F1 0.952 # 2
Slot Filling SNIPS Capsule-NLU F1 0.918 # 2
Intent Detection SNIPS Capsule-NLU Accuracy 0.977 # 1