Deep Open Intent Classification with Adaptive Decision Boundary
About
Open intent classification is a challenging task in dialogue systems. On the one hand, it should ensure the quality of known intent identification. On the other hand, it needs to detect the open (unknown) intent without prior knowledge. Current models are limited in finding the appropriate decision boundary to balance the performances of both known intents and the open intent. In this paper, we propose a post-processing method to learn the adaptive decision boundary (ADB) for open intent classification. We first utilize the labeled known intent samples to pre-train the model. Then, we automatically learn the adaptive spherical decision boundary for each known class with the aid of well-trained features. Specifically, we propose a new loss function to balance both the empirical risk and the open space risk. Our method does not need open intent samples and is free from modifying the model architecture. Moreover, our approach is surprisingly insensitive with less labeled data and fewer known intents. Extensive experiments on three benchmark datasets show that our method yields significant improvements compared with the state-of-the-art methods. The codes are released at https://github.com/thuiar/Adaptive-Decision-Boundary.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Open Intent Detection | Banking_CG (test) | F1 (IND)64.3 | 24 | |
| Open Intent Detection | StackOverflow_CG (test) | F1 (IND)76.13 | 24 | |
| Open Intent Detection | OOS_CG (test) | F1 (IND)53.87 | 24 | |
| Unknown Intent Detection | ATIS (test) | Macro F183.73 | 20 | |
| Unknown Intent Detection | StackOverflow 50% seen classes (test) | Accuracy86.4 | 11 | |
| Open Intent Classification | BANKING 50% known classes (test) | Accuracy86.54 | 10 | |
| OOD Intent Detection | Persian-ATIS standard (test) | Macro F142.15 | 10 | |
| OOD Intent Detection | SNIPS standard (test) | Macro F171.93 | 10 | |
| Out-of-Domain Detection | BANKING77 25% known ratio | Accuracy78.85 | 8 | |
| Out-of-Domain Detection | BANKING77 50% known ratio | Accuracy78.86 | 8 |