Zero-shot User Intent Detection via Capsule Neural Networks

By Congying Xia, Chenwei Zhang, Xiaohui Yan, Yi Chang, Philip S. Yu. 2018

224 citations recorded by the source corpus.

User intent detection plays a critical role in question-answering and dialog systems. Most previous works treat intent detection as a classification problem where utterances are labeled with predefined intents. However, it is labor-intensive and time-consuming to label users’ utterances as intents are diversely expressed and novel intents will continually be involved. Instead, we study the zero-shot intent detection problem, which aims to detect emerging user intents where no labeled utterances are currently available. We propose two capsule-based architectures: IntentCapsNet that extracts semantic features from utterances and aggregates them to discriminate existing intents, and IntentCapsNet-ZSL which gives IntentCapsNet the zero-shot learning ability to discriminate emerging intents via knowledge transfer from existing intents. Experiments on two real-world datasets show that our model not only can better discriminate diversely expressed existing intents, but is also able to discriminate emerging intents when no labeled utterances are available.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Erik Cambria
2. Philip S. Yu
3. Abhinav Rastogi
4. Wenpeng Yin
5. Xiao-Dan Zhu
6. Chenliang Li
7. Dan Roth
8. 刘晗

In Computer Vision and Pattern Recognition

1. 刘晗
2. Philip S. Yu
3. Lei Li
4. Wenpeng Yin
5. Erik Cambria
6. Raghav Gupta
7. Xiao-Dan Zhu
8. Haiyun Peng

Citations in the corpus

• Incremental Few-shot Text Classification with Multi-round New Classes: Formulation, Dataset and System (2021, background): “One is intent detection that aims at understanding the intents under user queries (Liu and Lane, 2016; Xia et al., 2018).”
• Towards Open Intent Discovery for Conversational Text (2019, background): “[15, 31, 48, 51]) since these are formulated as classification problems, and classify intents into limited, pre-defined categories.”
• Towards Scalable Multi-domain Conversational Agents: The Schema-Guided Dialogue Dataset (2019, background): “As virtual assistants incorporate diverse domains, recent work has focused on zero-shot modeling (Bapna et al. 2017; Xia et al. 2018; Shah et al. 2019), domain adaptation and transfer learning techniques (Yang, Salakhutdinov, and Cohen 2017; Rastogi, Hakkani-Tür, and Heck 2017; Zhu and Yu 2018).”
• Benchmarking Zero-shot Text Classification: Datasets, Evaluation and Entailment Approach (2019, background): “Xia et al. (2018) study the zero-shot intent detection problem.”
• Reconstructing Capsule Networks for Zero-shot Intent Classification (2019, methodology): “IntentCapsNet (Xia et al., 2018) is the first work to employ capsule networks for zero-shot intent clas-sification.”
• Meta-learning for Few-shot Natural Language Processing: A Survey (2020, background): “Both (Xia et al., 2018, 2020) used two intents as few-shot classes and other intents for training.”
• Induction Networks for Few-Shot Text Classification (2019, methodology): “Xia et al. (2018) reused the supervised model similar to that of Yang et al. (2018) for intent classification, in which a capsule-based architecture is extended to compute similarity between the target intents and source intents.”
• A Capsule Network for Recommendation and Explaining What You Like and Dislike (2019, methodology): “ly aggregated to form high-level features [22]. Recently, this notion has been applied to some NLP tasks, including relation extraction [36], text classification [35], zero-shot user intent detection [33] and multi-task learning [34]. In this work, we exploit the capsule-based architecture with a ne”
• Towards Scalable and Reliable Capsule Networks for Challenging NLP Applications (2019, background): “Xia et al. (2018) studied the zero-shot intent detection problem with capsule networks, which aims to detect emerging user intents in an unsupervised manner.”
• Pre-trained Language Models Can be Fully Zero-Shot Learners (2022, background): “Transferring knowledge from seen classes to unseen ones requires accurate and discriminative descriptions of all classes (Liu et al., 2019a; Xia et al., 2018) or joint embeddings of categories and documents (Nam et al., 2016).”
• Unknown Intent Detection Using Gaussian Mixture Model with an Application to Zero-shot Intent Classification (2020, background): “(2) How to better utilize these prior knowledge to extract more informative semantic representations, such as data augmentation and hierarchical representations learned by capsule networks (Xia et al., 2018).”
• Composed Variational Natural Language Generation for Few-shot Intents (2020, unknown): “(9) We choose the harmonic mean as our evaluation criteria instead of the arithmetic mean because the overall results are significantly affected by the many-shot class accuracy accm over the few-shot classes accf in arithmetic mean (Xian et al., 2017).”