Online multi-label dependency topic models for text classification

Multi-label text classification is an increasingly important field as large amounts of text data are available and extracting relevant information is important in many application contexts. Probabilistic generative models are the basis of a number of popular text mining methods such as Naive Bayes or Latent Dirichlet Allocation. However, Bayesian models for multi-label text classification often are overly complicated to account for label dependencies and skewed label frequencies while at the same time preventing overfitting. To solve this problem we employ the same technique that contributed to the success of deep learning in recent years: greedy layer-wise training. Applying this technique in the supervised setting prevents overfitting and leads to better classification accuracy. The intuition behind this approach is to learn the labels first and subsequently add a more abstract layer to represent dependencies among the labels. This allows using a relatively simple hierarchical topic model which can easily be adapted to the online setting. We show that our method successfully models dependencies online for large-scale multi-label datasets with many labels and improves over the baseline method not modeling dependencies. The same strategy, layer-wise greedy training, also makes the batch variant competitive with existing more complex multi-label topic models.     

MSML-BERT模型的层级多标签文本分类方法研究

层级多标签文本分类相比普通的多标签文本分类更具有挑战性,因为文本的多个标签组织成树状的层次结构。当前方法使用相同的模型结构来预测不同层级的标签,忽略了它们之间的差异性和多样性。并且没有充分地建模层级依赖关系,造成各层级标签尤其是下层长尾标签的预测性能差,且会导致标签不一致性问题。为了解决以上问题,将多任务学习架构引入,提出了MSML-BERT模型。该模型将标签结构中每一层的标签分类网络视为一个学习任务,通过任务间知识的共享和传递,提高各层级任务的性能。基于此,设计了多尺度特征抽取模块,用于捕捉不同尺度和粒度的特征以形成不同层级需要的各种知识。进一步,设计了多层级信息传播模块,用于充分建模层级依赖,在不同层级之间传递知识,以帮助下层任务。在该模块中,设计了层次化门控机制,为了过滤不同层级任务之间的知识流动。在RCV1-V2、NYT和WOS数据集上进行了充分的实验,结果显示该模型的总体表现尤其是在下层长尾标签上的表现超过了其他主流模型,并且能维持较低的标签不一致比率。     

Incorporating label dependency into the binary relevance framework for multi-label classification

In multi-label classification, examples can be associated with multiple labels simultaneously. The task of learning from multi-label data can be addressed by methods that transform the multi-label classification problem into several single-label classification problems. The binary relevance approach is one of these methods, where the multi-label learning task is decomposed into several independent binary classification problems, one for each label in the set of labels, and the final labels for each example are determined by aggregating the predictions from all binary classifiers. However, this approach fails to consider any dependency among the labels. Aiming to accurately predict label combinations, in this paper we propose a simple approach that enables the binary classifiers to discover existing label dependency by themselves. An experimental study using decision trees, a kernel method as well as Naïve Bayes as base-learning techniques shows the potential of the proposed approach to improve the multi-label classification performance.     

Multi-label classification by exploiting label correlations

Nowadays, multi-label classification methods are of increasing interest in the areas such as text categorization, image annotation and protein function classification. Due to the correlation among the labels, traditional single-label classification methods are not directly applicable to the multi-label classification problem. This paper presents two novel multi-label classification algorithms based on the variable precision neighborhood rough sets, called multi-label classification using rough sets (MLRS) and MLRS using local correlation (MLRS-LC). The proposed algorithms consider two important factors that affect the accuracy of prediction, namely the correlation among the labels and the uncertainty that exists within the mapping between the feature space and the label space. MLRS provides a global view at the label correlation while MLRS-LC deals with the label correlation at the local level. Given a new instance, MLRS determines its location and then computes the probabilities of labels according to its location. The MLRS-LC first finds out its topic and then the probabilities of new instance belonging to each class is calculated in related topic. A series of experiments reported for seven multi-label datasets show that MLRS and MLRS-LC achieve promising performance when compared with some well-known multi-label learning algorithms.     

融合标签关系的法律文本多标签分类方法

随着大数据技术的快速发展,多标签文本分类在司法领域也催生出诸多应用.在法律文本中通常存在多个要素标签,标签之间往往具有相互依赖性或相关性,准确识别这些标签需要多标签分类方法的支持.因此,文中提出融合标签关系的法律文本多标签分类方法.方法构建标签的共现矩阵,利用图卷积网络捕捉标签之间的依赖关系,并结合标签注意力机制,计算法律文本和标签每个词的相关程度,得到特定标签的法律文本语义表示.最后,融合标签图构建的依赖关系和特定标签的法律文本语义表示,对文本进行综合表示,实现文本的多标签分类.在法律数据集上的实验表明,文中方法获得较好的分类精度和稳定性.     

Stacking model of multi-label classification based on pruning strategies

Exploiting dependencies between the labels is the key of improving the performance of multi-label classification. In this paper, we divide the utilizing methods of label dependence into two groups from the perspective of different ways of problem transformation: label grouping method and feature space extending method. As to the feature space extending method, we find that the common problem is how to measure the dependencies between labels and to select proper labels to add to the original feature space. Therefore, we propose a ReliefF-based pruning model for multi-label classification (ReliefF-based stacking, RFS). RFS measures the dependencies between labels in a feature selection perspective and then selects the more relative labels into the original feature space. Experimental results of 9 multi-label benchmark datasets shows that RFS is more effective compared to other advanced multi-label classification algorithms.

     

面向ICD疾病分类的深度学习方法研究

国际疾病分类（ICD）是用于临床目的和健康管理的分类工具,是卫生统计数据的建立基础,在其庞大的分类体系中,含有与疾病健康问题和临床治疗相关的分类和对应的代码。针对在国际疾病分类的庞大标签空间中的多标签分类问题,提出一种端到端的深度学习方法。采用改进的图注意力网络对标签空间进行建模,基于注意力重构的多标签分类器进行分类。在标签空间建模中,结合国际疾病分类中手术与操作分类的层次结构,构建出三种不同的图结构,利用图注意力网络将标签空间的结构信息融入到模型中,从而利用标签之间的依赖关系进行多标签文本分类。所提出的方法与实际应用场景有着紧密联系。实验表明,在临床国际疾病分类数据集上,相比于传统文本分类和其他标签空间建模方法,所提方法在分类性能上有明显的提升。     

Exploiting label dependencies for improved sample complexity

Multi-label classification exhibits several challenges not present in the binary case. The labels may be interdependent, so that the presence of a certain label affects the probability of other labels’ presence. Thus, exploiting dependencies among the labels could be beneficial for the classifier’s predictive performance. Surprisingly, only a few of the existing algorithms address this issue directly by identifying dependent labels explicitly from the dataset. In this paper we propose new approaches for identifying and modeling existing dependencies between labels. One principal contribution of this work is a theoretical confirmation of the reduction in sample complexity that is gained from unconditional dependence. Additionally, we develop methods for identifying conditionally and unconditionally dependent label pairs; clustering them into several mutually exclusive subsets; and finally, performing multi-label classification incorporating the discovered dependencies. We compare these two notions of label dependence (conditional and unconditional) and evaluate their performance on various benchmark and artificial datasets. We also compare and analyze labels identified as dependent by each of the methods. Moreover, we define an ensemble framework for the new methods and compare it to existing ensemble methods. An empirical comparison of the new approaches to existing base-line and state-of-the-art methods on 12 various benchmark datasets demonstrates that in many cases the proposed single-classifier and ensemble methods outperform many multi-label classification algorithms. Perhaps surprisingly, we discover that the weaker notion of unconditional dependence plays the decisive role.     

多粒度信息关系增强的多标签文本分类

基于深度学习的多标签文本分类方法存在两个主要缺陷:缺乏对文本信息多粒度的学习,以及对标签间约束性关系的利用.针对这些问题,提出一种多粒度信息关系增强的多标签文本分类方法.首先,通过联合嵌入的方式将文本与标签嵌入到同一空间,并利用BERT预训练模型获得文本和标签的隐向量特征表示.然后,构建3个多粒度信息关系增强模块:文档级信息浅层标签注意力分类模块、词级信息深层标签注意力分类模块和标签约束性关系匹配辅助模块.其中,前两个模块针对共享特征表示进行多粒度学习:文档级文本信息与标签信息浅层交互学习,以及词级文本信息与标签信息深层交互学习.辅助模块通过学习标签间关系来提升分类性能.最后,所提方法在3个代表性数据集上,与当前主流的多标签文本分类算法进行了比较.结果表明,在主要指标Micro-F1、MacroF1、nDCG@k、P@k上均达到了最佳效果.     

Labelset topic model for multi-label document classification

It has recently been suggested that assuming independence between labels is not suitable for real-world multi-label classification. To account for label dependencies, this paper proposes a supervised topic modeling algorithm, namely labelset topic model (LsTM). Our algorithm uses two labelset layers to capture label dependencies. LsTM offers two major advantages over existing supervised topic modeling algorithms: it is straightforward to interpret and it allows words to be assigned to combinations of labels, rather than a single label. We have performed extensive experiments on several well-known multi-label datasets. Experimental results indicate that the proposed model achieves performance on par with and often exceeding that of state-of-the-art methods both qualitatively and quantitatively.     

Multi-label fault recognition framework using deep reinforcement learning and curriculum learning mechanism

In the actual working site, the equipment often works in different working conditions while the manufacturing system is rather complicated. However, traditional multi-label learning methods need to use the pre-defined label sequence or synchronously predict all labels of the input sample in the fault diagnosis domain. Deep reinforcement learning (DRL) combines the perception ability of deep learning and the decision-making ability of reinforcement learning. Moreover, the curriculum learning mechanism follows the learning approach of humans from easy to complex. Consequently, an improved proximal policy optimization (PPO) method, which is a typical algorithm in DRL, is proposed as a novel method on multi-label classification in this paper. The improved PPO method could build a relationship between several predicted labels of input sample because of designing an action history vector, which encodes all history actions selected by the agent at current time step. In two rolling bearing experiments, the diagnostic results demonstrate that the proposed method provides a higher accuracy than traditional multi-label methods on fault recognition under complicated working conditions. Besides, the proposed method could distinguish the multiple labels of input samples following the curriculum mechanism from easy to complex, compared with the same network using the pre-defined label sequence.     

Missing multi-label learning with non-equilibrium based on two-level autoencoder

For a multi-label learning framework, each instance may belong to multiple labels simultaneously. The classification accuracy can be improved significantly by exploiting various correlations, such as label correlations, feature correlations, or the correlations between features and labels. There are few studies on how to combine the feature and label correlations, and they deal more with complete data sets. However, missing labels or other phenomena often occur because of the cost or technical limitations in the data acquisition process. A few label completion algorithms currently suitable for missing multi-label learning, ignore the noise interference of the feature space. At the same time, the threshold of the discriminant function often affects the classification results, especially those of the labels near the threshold. All these factors pose considerable difficulties in dealing with missing labels using label correlations. Therefore, we propose a missing multi-label learning algorithm with non-equilibrium based on a two-level autoencoder. First, label density is introduced to enlarge the classification margin of the label space. Then, a new supplementary label matrix is augmented from the missing label matrix with the non-equilibrium label completion method. Finally, considering feature space noise, a two-level kernel extreme learning machine autoencoder is constructed to implement the information feature and label correlation. The effectiveness of the proposed algorithm is verified by many experiments on both missing and complete label data sets. A statistical analysis of hypothesis validates our approach.

     

Separate or joint? Estimation of multiple labels from crowdsourced annotations

Artificial intelligence techniques aimed at more naturally simulating human comprehension fit the paradigm of multi-label classification. Generally, an enormous amount of high-quality multi-label data is needed to form a multi-label classifier. The creation of such datasets is usually expensive and time-consuming. A lower cost way to obtain multi-label datasets for use with such comprehension–simulation techniques is to use noisy crowdsourced annotations. We propose incorporating label dependency into the label-generation process to estimate the multiple true labels for each instance given crowdsourced multi-label annotations. Three statistical quality control models based on the work of Dawid and Skene are proposed. The label-dependent DS (D-DS) model simply incorporates dependency relationships among all labels. The label pairwise DS (P-DS) model groups labels into pairs to prevent interference from uncorrelated labels. The Bayesian network label-dependent DS (ND-DS) model compactly represents label dependency using conditional independence properties to overcome the data sparsity problem. Results of two experiments, “affect annotation for lines in story” and “intention annotation for tweets”, show that (1) the ND-DS model most effectively handles the multi-label estimation problem with annotations provided by only about five workers per instance and that (2) the P-DS model is best if there are pairwise comparison relationships among the labels. To sum up, flexibly using label dependency to obtain multi-label datasets is a promising way to reduce the cost of data collection for future applications with minimal degradation in the quality of the results.     

基于标签相关性的K近邻多标签分类方法

针对K近邻多标签(ML-KNN)分类算法中未考虑标签相关性的问题,提出了一种基于标签相关性的K近邻多标签分类(CML-KNN)算法。首先,计算出标签集合中每对标签间的条件概率;其次,对于即将被预测的标签,将其与已经预测的标签间的条件概率进行排序,求出最大值;最后,将最大值跟对应标签值相乘同时结合最大化后验概率(MAP)来构造多标签分类模型,对新标签进行预测。实验结果表明,所提算法在Emotions数据集上的分类性能均优于ML-KNN、AdaboostMH、RAkEL、BPMLL这4种算法;在Yeast、Enron数据集上仅在1~2个评价指标上低于ML-KNN与RAkEL算法。由实验分析可知,该算法取得了较好的分类效果。     

A hybrid algorithm for Bayesian network structure learning with application to multi-label learning

We present a novel hybrid algorithm for Bayesian network structure learning, called H2PC. It first reconstructs the skeleton of a Bayesian network and then performs a Bayesian-scoring greedy hill-climbing search to orient the edges. The algorithm is based on divide-and-conquer constraint-based subroutines to learn the local structure around a target variable. We conduct two series of experimental comparisons of H2PC against Max–Min Hill-Climbing (MMHC), which is currently the most powerful state-of-the-art algorithm for Bayesian network structure learning. First, we use eight well-known Bayesian network benchmarks with various data sizes to assess the quality of the learned structure returned by the algorithms. Our extensive experiments show that H2PC outperforms MMHC in terms of goodness of fit to new data and quality of the network structure with respect to the true dependence structure of the data. Second, we investigate H2PC’s ability to solve the multi-label learning problem. We provide theoretical results to characterize and identify graphically the so-called minimal label powersets that appear as irreducible factors in the joint distribution under the faithfulness condition. The multi-label learning problem is then decomposed into a series of multi-class classification problems, where each multi-class variable encodes a label powerset. H2PC is shown to compare favorably to MMHC in terms of global classification accuracy over ten multi-label data sets covering different application domains. Overall, our experiments support the conclusions that local structural learning with H2PC in the form of local neighborhood induction is a theoretically well-motivated and empirically effective learning framework that is well suited to multi-label learning. The source code (in R) of H2PC as well as all data sets used for the empirical tests are publicly available.     

一种基于标签相关性的类属属性多标签分类算法

多标签学习中一个样本可同时属于多个类别标签,每个标签都可能拥有反映该标签特定特点的特征,即类属属性,目前已经出现了基于类属属性的多标签分类算法LIFT。针对LIFT算法中未考虑标签之间相互关系的问题,提出一种基于标签相关性的类属属性多标签分类算法CLLIFT。该算法使用标签距离度量标签之间的相关性,通过在类属属性空间附加相关标签的方式完成标签相关性的引入,以达到提升分类性能的目的。在四个多标签数据集上的实验结果表明,所提算法与LIFT算法相比在多个多标签评价指标上平均提升21.1%。     

Incremental Multi-Label Learning with Active Queries

In multi-label learning,it is rather expensive to label instances since they are simultaneously associated with multiple labels.Therefore,active learning,which reduces the labeling cost by actively querying the labels of the most valuable data,becomes particularly important for multi-label learning.A good multi-label active learning algorithm usually consists of two crucial elements:a reasonable criterion to evaluate the gain of querying the label for an instance,and an effective classification model,based on whose prediction the criterion can be accurately computed.In this paper,we first introduce an effective multi-label classification model by combining label ranking with threshold learning,which is incrementally trained to avoid retraining from scratch after every query.Based on this model,we then propose to exploit both uncertainty and diversity in the instance space as well as the label space,and actively query the instance-label pairs which can improve the classification model most.Extensive experiments on 20 datasets demonstrate the superiority of the proposed approach to state-of-the-art methods.     

Air pollution prediction via multi-label classification

A Bayesian network classifier can be used to estimate the probability of an air pollutant overcoming a certain threshold. Yet multiple predictions are typically required regarding variables which are stochastically dependent, such as ozone measured in multiple stations or assessed according to by different indicators. The common practice (independent approach) is to devise an independent classifier for each class variable being predicted; yet this approach overlooks the dependencies among the class variables. By appropriately modeling such dependencies one can improve the accuracy of the forecasts. We address this problem by designing a multi-label classifier, which simultaneously predict multiple air pollution variables. To this end we design a multi-label classifier based on Bayesian networks and learn its structure through structural learning. We present experiments in three different case studies regarding the prediction of PM2.5 and ozone. The multi-label classifier outperforms the independent approach, allowing to take better decisions.     

The use of data-derived label hierarchies in multi-label classification

Instead of traditional (multi-class) learning approaches that assume label independency, multi-label learning approaches must deal with the existing label dependencies and relations. Many approaches try to model these dependencies in the process of learning and integrate them in the final predictive model, without making a clear difference between the learning process and the process of modeling the label dependencies. Also, the label relations incorporated in the learned model are not directly visible and can not be (re)used in conjunction with other learning approaches. In this paper, we investigate the use of label hierarchies in multi-label classification, constructed in a data-driven manner. We first consider flat label sets and construct label hierarchies from the label sets that appear in the annotations of the training data by using a hierarchical clustering approach. The obtained hierarchies are then used in conjunction with hierarchical multi-label classification (HMC) approaches (two local model approaches for HMC, based on SVMs and PCTs, and two global model approaches, based on PCTs for HMC and ensembles thereof). The experimental results reveal that the use of the data-derived label hierarchy can significantly improve the performance of single predictive models in multi-label classification as compared to the use of a flat label set, while this is not preserved for the ensemble models.     

基于负相关性增强的不平衡多标签学习算法

由于标签空间过大,标签分布不平衡问题在多标签数据集中广泛存在,解决该问题在一定程度上可以提高多标签学习的分类性能。通过标签相关性提升分类性能是解决该问题的一种最常见的有效策略,众多学者进行了大量研究,然而这些研究更多地是采用基于正相关性策略提升性能。在实际问题中,除了正相关性外,标签的负相关性也可能存在,如果在考虑正相关性的同时,兼顾负相关性,无疑能够进一步改善分类器的性能。基于此,提出了一种基于负相关性增强的不平衡多标签学习算法——MLNCE,旨在解决多标签不平衡问题的同时,兼顾标签间的正负相关性,从而提高多标签分类器的分类性能。首先利用标签密度信息改造标签空间;然后在密度标签空间中探究标签真实的正反相关性信息,并添加到分类器目标函数中;最后利用加速梯度下降法求解输出权重以得到预测结果。在11个多标签标准数据集上与其他6种多标签学习算法进行对比实验,结果表明MLNCE算法可以有效提高分类精度。