Graph convolutional network (GCN) has been successfully applied to capture global non-consecutive and long-distance semantic information for text classification. However, while GCN-based methods have shown promising results in offline evaluations, they commonly follow a seen-token-seen-document paradigm by constructing a fixed document-token graph and cannot make inferences on new documents. It is a challenge to deploy them in online systems to infer steaming text data. In this work, we present a continual GCN model (ContGCN) to generalize inferences from observed documents to unobserved documents. Concretely, we propose a new all-token-any-document paradigm to dynamically update the document-token graph in every batch during both the training and testing phases of an online system. Moreover, we design an occurrence memory module and a self-supervised contrastive learning objective to update ContGCN in a label-free manner. A 3-month A/B test on Huawei public opinion analysis system shows ContGCN achieves 8.86% performance gain compared with state-of-the-art methods. Offline experiments on five public datasets also show ContGCN can improve inference quality. The source code will be released at https://github.com/Jyonn/ContGCN.

翻译：图卷积网络（GCN）已被成功应用于捕获文本分类中的全局非连续及长距离语义信息。然而，尽管基于GCN的方法在离线评估中展现出良好效果，但它们通常遵循"已见标记-已见文档"范式，通过构建固定的文档-标记图，无法对新文档进行推理。这对其在在线系统中处理流式文本数据构成了挑战。本文提出了一种持续图卷积模型（ContGCN），将推理能力从已观测文档推广至未观测文档。具体而言，我们提出了一种新的"全体标记-任意文档"范式，在在线系统的训练与测试阶段，每批次动态更新文档-标记图。此外，我们设计了出现记忆模块与自监督对比学习目标，以无标签方式更新ContGCN。在华为舆情分析系统上进行的为期3个月的A/B测试表明，ContGCN相较于最先进方法实现了8.86%的性能提升。在五个公开数据集上的离线实验同样显示ContGCN能够提高推理质量。源代码将在https://github.com/Jyonn/ContGCN 发布。