摘要 开放意图分类是构建智能对话系统的重要任务之一,其目标是在识别已知意图的同时准确检测未知意图.然而,现有方法在建模复杂语义结构方面存在局限,难以刻画类别内部的多样性分布,容易导致类间混淆.为此,文中提出基于自适应粒球与纯净簇拆分的开放意图分类方法(Adaptive Granular-Ball and Pure Cluster Splitting for Open Intent Classification, AGPCS-OIC).首先,通过自适应粒球聚类构建反映数据真实分布的多中心子类结构,刻画类内异质性.然后,引入结构稀疏性驱动的纯净簇拆分策略,进一步划分边界松散但纯度较高的粒球,提升决策边界的表达能力和对未知类别的排斥能力.同时,结合粒球感知对比学习机制,以粒球中心为锚点构建结构级语义对,引导模型在特征空间中增强类内聚合性与类间可分性.实验表明,AGPCS-OIC在多个开放意图分类数据集上性能较优.
Abstract：Open intent classification is a critical task in building intelligent dialogue systems, and it is intended to detect unknown intents accurately while recognizing the known ones. However, existing methods are limited in modeling complex semantic structures and fail to represent the diversity within intent classes, resulting in inter-class confusion. To address this issue, a method for adaptive granular-ball and pure cluster splitting for open intent classification(AGPCS-OIC)is proposed. First, adaptive granular-ball clustering is applied to construct multi-center subclass structures reflecting the true data distribution, and thus intra-class heterogeneity is captured more effectively. Then, a structural sparsity-based pure cluster splitting strategy is introduced to further divide loosely bounded but high-purity granular-balls. The expressiveness of decision boundaries and the ability to reject the unknown intents are enhanced. Additionally, a granular-ball-aware contrastive learning mechanism is incorporated. Structural-level semantic pairs are built with granular-ball centers serving as anchors, to guide the model to improve intra-class compactness and inter-class separability in the feature space. Experiments show that AGPCS-OIC achieves strong performance on multiple open intent classification datasets.
王景凯, 李艳花, 刘家芬, 王向坤, 杨新. 基于自适应粒球与纯净簇拆分的开放意图分类[J]. 模式识别与人工智能, 2025, 38(8): 740-751.
WANG Jingkai, LI Yanhua, LIU Jiafen, WANG Xiangkun, YANG Xin. Adaptive Granular-Ball and Pure Cluster Splitting for Open Intent Classification. Pattern Recognition and Artificial Intelligence, 2025, 38(8): 740-751.
[1] ZHENG Y H, CHEN G Y, HUANG M L. Out-of-Domain Detection for Natural Language Understanding in Dialog Systems. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2020, 28: 1198-1209. [2] ABRO W A, QI G L, ALI Z, et al. Multi-turn Intent Determination and Slot Filling with Neural Networks and Regular Expressions. Knowledge-Based Systems, 2020, 208. DOI: 10.1016/j.knosys.2020.106428. [3] FIRDAUS M, KUMAR A, EKBAL A, et al. A Multi-task Hierarchical Approach for Intent Detection and Slot Filling. Knowledge-Based Systems, 2019, 183. DOI: 10.1016/j.knosys.2019.07.017. [4] YANG P, JI D, AI C M, et al. AISE: Attending to Intent and Slots Explicitly for Better Spoken Language Understanding. Knowledge-Based Systems, 2021, 211. DOI: 10.1016/j.knosys.2020.106537. [5] SHU L, XU H, LIU B. DOC: Deep Open Classification of Text Documents // Proc of the Conference on Empirical Methods in Natural Language Processing. Stroudsburg, USA: ACL, 2017: 2911-2916. [6] BREUNIG M M, KRIEGEL H P, NG R T, et al. LOF: Identifying Density-Based Local Outliers. ACM SIGMOD Record, 2000, 29(2): 93-104. [7] JAIN L P, SCHEIRER W J, BOULT T E, et al. Multiclass Open Set Recognition Using Probability of Inclusion // Proc of the 13th European Conference on Computer Vision. Berlin, Germany: Springer, 2014: 393-409. [8] HENDRYCKS D, GIMPEL K. A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks[C/OL]. [2025-05-15]. https://arxiv.org/pdf/1610.02136. [9] CHEN G Y, PENG P X, WANG X Q, et al. Adversarial Reciprocal Points Learning for Open Set Recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021, 44(11): 8065-8081. [10] XU K Y, REN T Z, ZHANG S K, et al. Unsupervised Out-of-Domain Detection via Pre-trained Transformers // Proc of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing. Stroudsburg, USA: ACL, 2021: 1052-1061. [11] ZHANG H L, XU H, LIN T E. Deep Open Intent Classification with Adaptive Decision Boundary. Proceedings of the AAAI Conference on Artificial Intelligence, 2021, 35(16): 14374-14382. [12] ZHANG H L, XU H, ZHAO S J, et al. Learning Discriminative Representations and Decision Boundaries for Open Intent Detection. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2023, 31: 1611-1623. [13] LIU X K, LI J Q, MU J J,et al. Effective Open Intent Classification with K-Center Contrastive Learning and Adjustable Decision Boundary. Proceedings of the AAAI Conference on Artificial Inte-lligence, 2023, 37(11): 13291-13299. [14] LI Y H, OUYANG X C, PAN C F, et al. Multi-granularity Open Intent Classification via Adaptive Granular-Ball Decision Boundary. Proceedings of the AAAI Conference on Artificial Intelligence, 2025, 39(23): 24512-24520. [15] BAI H Y, MING Y F, KATZ-SAMUELS J, et al. HYPO: Hyperspherical Out-of-Distribution Generalization[C/OL].[2025-05-15]. https://arxiv.org/pdf/2402.07785. [16] XIA S Y, LIU Y S, DING X, et al. Granular Ball Computing Classifiers for Efficient, Scalable and Robust Learning. Information Sciences, 2019, 483: 136-152. [17] XIA S Y, ZHANG H, LI W H, et al. GBNRS: A Novel Rough Set Algorithm for Fast Adaptive Attribute Reduction in Classification. IEEE Transactions on Knowledge and Data Engineering, 2020, 34(3): 1231-1242. [18] BENDALE A, BOULT T E. Towards Open Set Deep Networks // Proc of the IEEE Conference on Computer Vision and Pattern Re-cognition. Washington, USA: IEEE, 2016: 1563-1572. [19] CHEN L. Topological Structure in Visual Perception. Science, 1982, 218(4573): 699-700. [20] XIA S Y, ZHENG S Y, WANG G Y, et al. Granular Ball Sampling for Noisy Label Classification or Imbalanced Classification. IEEE Transactions on Neural Networks and Learning Systems, 2023, 34(4): 2144-2155. [21] CASANUEVA I, TEMČINAS T, GERZ D, et al. Efficient Intent Detection with Dual Sentence Encoders // Proc of the 2nd Workshop on Natural Language Processing for Conversational AI. Stroudsburg, USA: ACL, 2020: 38-45. [22] LARSON S, MAHENDRAN A, PEPER J J, et al. An Evaluation Dataset for Intent Classification and Out-of-Scope Prediction // Proc of the Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing. Stroudsburg, USA: ACL, 2019: 1311-1316. [23] LIU X K, ESHGHI A, SWIETOJANSKI P, et al. Benchmarking Natural Language Understanding Services for Building Conversational Agents // Proc of the 10th International Workshop on Spoken Dialogue Systems. Berlin, Germany: Springer, 2021: 165-183. [24] COUCKE A, SAADE A, BALL A, et al. Snips Voice Platform: An Embedded Spoken Language Understanding System for Privateby-Design Voice Interfaces[C/OL].[2025-05-15]. https://arxiv.org/pdf/1805.10190. [25] XU J M, WANG P, TIAN G H, et al. Short Text Clustering via Convolutional Neural Networks // Proc of the 1st Workshop on Vector Space Modeling for Natural Language Processing. Stroudsburg, USA: ACL, 2015: 62-69. [26] ZHOU Y H, WANG P Y, LIU P J, et al. The Open-World Lottery Ticket Hypothesis for OOD Intent Classification // Proc of the Joint International Conference on Computational Linguistics, Language Resources and Evaluation. Stroudsburg, USA: ACL, 2024: 15988-15999.
2025, Vol. 38 
