Semi-supervised multi-label classification using incomplete label information

Tan, Qiaoyu; Yu, Yongchao; Yu, Guoxian; Wang, Jun

doi:10.1016/j.neucom.2017.04.033

Cited by 48 publications

(40 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To avoid this limitation, researchers have resorted to develop semi-supervised multilabel classifiers [23], [24], [25], [26], [27], in which limited labeled samples as well as abundant unlabeled samples are jointly used for training. Besides, considering the fact that labeled data is tagged by human efforts, they might have some missing or noisly labels [28], [29], [30], [31], [32], [33], several approaches have been proposed to design multi-label classifiers under weak-label setting [28], [29], [34], [35] or with noisy labels [36], [32], [37], [38].…”

Section: A Multi-label Learningmentioning

confidence: 99%

Individuality- and Commonality-Based Multiview Multilabel Learning

Tan

Wang

et al. 2021

IEEE Trans. Cybern.

Self Cite

View full text Add to dashboard Cite

In multi-view multi-label learning, each object is represented by several heterogeneous feature representations and is also annotated with a set of discrete non-exclusive labels. Previous studies typically focus on capturing the shared latent patterns among multiple views, while do not sufficiently consider the diverse characteristics of individual views, which can cause performance degradation. In this paper, we propose a novel approach (ICM2L) to explicitly explore the individuality and commonality information of multi-label multiple view data in a unified model. Specifically, a common subspace is learned across different views to capture the shared patterns. Then, multiple individual classifiers are exploited to explore the characteristics of individual views. Next, an ensemble strategy is adopted to make prediction. Finally, we develop an alternative solution to jointly optimize our model, which can enhance the robustness of the proposed model towards rare labels and reinforce the reciprocal effects of individuality and commonality among heterogeneous views, and thus further improve the performance. Experiments on various real-word datasets validate the effectiveness of ICM2L against state-of-the-art solutions, and ICM2L can leverage the individuality and commonality information to achieve an improved performance as well as to enhance the robustness toward rare labels.

show abstract

Section: A Multi-label Learningmentioning

confidence: 99%

Individuality- and Commonality-Based Multiview Multilabel Learning

Tan

Wang

et al. 2021

IEEE Trans. Cybern.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Tan et al [16] developed the Semi-supervised Multilabel categorization via Incomplete Label Information (SMILE). Initially, label correlation was estimated from incompletely labeled features and their missed CLs were restored.…”

Section: A Review On Various Techniques For Multi Lable Laearningmentioning

confidence: 99%

A Survey and Analysis of Multi-Label Learning Techniques for Data Streams

Yallini¹,

Mukunthan²

2020

IJISRT

View full text Add to dashboard Cite

Multi-Label Learning (MLL) solves the challenge of characterizing every sample via a particular feature which relates to the group of labels at once. That is, a sample has manifold views where every view is symbolized through a Class Label (CL). In the past decades, significant number of researches has been prepared towards this promising machine learning concept. Such researches on MLL have been motivated on a pre-determined group of CLs. In most of the appliances, the configuration is dynamic and novel views might appear in a Data Stream (DS). In this scenario, a MLL technique should able to identify and categorize the features with evolving fresh labels for maintaining a better predictive performance. For this purpose, several MLL techniques were introduced in the earlier decades. This article aims to present a survey on this field with consequence on conventional MLL techniques. Initially, various MLL techniques proposed by many researchers are studied. Then, a comparative analysis is carried out in terms of merits and demerits of those techniques to conclude the survey and recommend the future enhancements on MLL techniques.

show abstract

“…Recently, hybrid learning methods that can learn from labeled and unlabeled data gained much interest hoping to construct better performing classifiers [5]. These methods can be categorized into three major approaches varying from fully supervised to unsupervised learning [6]: (1) Building an initial classifier based on a set of labeled data then use it for labeling the unlabeled data. Hereafter, a new classifier is built based on both the earlier and the new labeled data.…”

Section: Introductionmentioning

confidence: 99%

“…Thus, such facts definitely burden the deployment of semi-supervised learning in high-reliability real applications, which require more accurate performance compared with the existing supervised techniques. Therefore, it is crucial to build a safe semi-supervised classifier using unlabeled data without reducing the performance significantly [6]. The word safe means that the overall performance is not statistically considered worse than those using only labeled data.…”

Section: Introductionmentioning

confidence: 99%

Can People Really Do Nothing? Handling Annotation Gaps in ADL Sensor Data

Abdel-Hakim

Deabes

2019

Algorithms

View full text Add to dashboard Cite

In supervised Activities of Daily Living (ADL) recognition systems, annotating collected sensor readings is an essential, yet exhaustive, task. Readings are collected from activity-monitoring sensors in a 24/7 manner. The size of the produced dataset is so huge that it is almost impossible for a human annotator to give a certain label to every single instance in the dataset. This results in annotation gaps in the input data to the adopting learning system. The performance of the recognition system is negatively affected by these gaps. In this work, we propose and investigate three different paradigms to handle these gaps. In the first paradigm, the gaps are taken out by dropping all unlabeled readings. A single "Unknown" or "Do-Nothing" label is given to the unlabeled readings within the operation of the second paradigm. The last paradigm handles these gaps by giving every set of them a unique label identifying the encapsulating certain labels. Also, we propose a semantic preprocessing method of annotation gaps by constructing a hybrid combination of some of these paradigms for further performance improvement. The performance of the proposed three paradigms and their hybrid combination is evaluated using an ADL benchmark dataset containing more than 2.5 × 10 6 sensor readings that had been collected over more than nine months. The evaluation results emphasize the performance contrast under the operation of each paradigm and support a specific gap handling approach for better performance.

show abstract

Semi-supervised multi-label classification using incomplete label information

Cited by 48 publications

References 12 publications

Individuality- and Commonality-Based Multiview Multilabel Learning

Individuality- and Commonality-Based Multiview Multilabel Learning

A Survey and Analysis of Multi-Label Learning Techniques for Data Streams

Can People Really Do Nothing? Handling Annotation Gaps in ADL Sensor Data

Contact Info

Product

Resources

About