Joint Multi-Label Attention Networks for Social Text Annotation

Dong, Hang; Wang, Wei; Huang, Kaizhu; Coenen, Frans

doi:10.18653/v1/n19-1136

Cited by 10 publications

(4 citation statements)

References 23 publications

(25 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The JMAN model, as shown in Fig. 2, is an extension to our previous work [44]. Instead of feeding the whole text sequence X into the neural network as in HAN [12], [17], JMAN takes as inputs the title, x t , and the content (in this work, the abstract of a document is treated as the content), x a , and processes them separately, where…”

Section: B Overall Designmentioning

confidence: 99%

Automated Social Text Annotation With Joint Multilabel Attention Networks

Dong

Wang

Huang

et al. 2021

IEEE Trans. Neural Netw. Learning Syst.

Self Cite

View full text Add to dashboard Cite

Automated social text annotation is the task of suggesting a set of tags for shared documents on social media platforms. The automated annotation process can reduce users' cognitive overhead in tagging and improve tag management for better search, browsing, and recommendation of documents. It can be formulated as a multilabel classification problem. We propose a novel deep learning-based method for this problem and design an attention-based neural network with semantic-based regularization, which can mimic users' reading and annotation behavior to formulate better document representation, leveraging the semantic relations among labels. The network separately models the title and the content of each document and injects an explicit, title-guided attention mechanism into each sentence. To exploit the correlation among labels, we propose two semantic-based loss regularizers, i.e., similarity and subsumption, which enforce the output of the network to conform to label semantics. The model with the semantic-based loss regularizers is referred to as the joint multilabel attention network (JMAN). We conducted a comprehensive evaluation study and compared JMAN to the state-of-the-art baseline models, using four large, real-world social media data sets. In terms of F 1 , JMAN significantly outperformed bidirectional gated recurrent unit (Bi-GRU) relatively by around 12.8%-78.6% and the hierarchical attention network (HAN) by around 3.9%-23.8%. The JMAN model demonstrates advantages in convergence and training speed. Further improvement of performance was observed against latent Dirichlet allocation (LDA) and support vector machine (SVM). When applying the semantic-based loss regularizers, the performance of HAN and Bi-GRU in terms of F 1 was also boosted. It is also found that dynamic update of the label semantic matrices (JMAN d) has

show abstract

Section: B Overall Designmentioning

confidence: 99%

Automated Social Text Annotation With Joint Multilabel Attention Networks

Dong

Wang

Huang

et al. 2021

IEEE Trans. Neural Netw. Learning Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Multi-Task Learning (MTL) is a learning paradigm in machine learning and its purpose is to take advantage of useful information contributed by multiple related tasks to improve the generalization performance of all the tasks [11]. MTL has shown significant advantage to single-task learning because of its ability to facilitate knowledge sharing between tasks [31], e.g., bioinformatics and health informatics [32,33], web applications [34,35] and remote sensing [36][37][38].…”

Section: Multi-task Learning In Human Activity Recognitionmentioning

confidence: 99%

Context-Aware Human Activity and Smartphone Position-Mining with Motion Sensors

et al. 2019

Self Cite

View full text Add to dashboard Cite

Today's smartphones are equipped with embedded sensors, such as accelerometers and gyroscopes, which have enabled a variety of measurements and recognition tasks. In this paper, we jointly investigate two types of recognition problems in a joint manner, e.g., human activity recognition and smartphone on-body position recognition, in order to enable more robust context-aware applications. So far, these two problems have been studied separately without considering the interactions between each other. In this study, by first applying a novel data preprocessing technique, we propose a joint recognition framework based on the multi-task learning strategy, which can reduce computational demand, better exploit complementary information between the two recognition tasks, and lead to higher recognition performance. We also extend the joint recognition framework so that additional information, such as user identification with biometric motion analysis, can be offered. We evaluate our work systematically and comprehensively on two datasets with real-world settings. Our joint recognition model achieves the promising performance of 0.9174 in terms of F1-score for user identification on the benchmark RealWorld Human Activity Recognition (HAR) dataset. On the other hand, in comparison with the conventional approach, the proposed joint model is shown to be able to improve human activity recognition and position recognition by 5.1% and 9.6% respectively.

show abstract

“…Recently, the attention mechanism has achieved a great success in many NLP tasks [34], [35], [36], [37], [38], [39], [40], [41], [42]. In the task of (T)ABSA, attention mechanism can help models effectively distinguishing the sentiment polarities of different aspects in the same sentence [9], [10], [11], [43], [44], [45], [46], [47], [48], [49].…”

Section: Introductionmentioning

confidence: 99%

Embedding Refinement Framework for Targeted Aspect-Based Sentiment Analysis

Liang

Yin

Liu

et al. 2023

IEEE Trans. Affective Comput.

View full text Add to dashboard Cite

The state-of-the-art approaches to Targeted Aspect-Based Sentiment Analysis (TABSA) are mostly built on deep neural networks with attention mechanisms. One problem is that embeddings of targets and aspects are either pre-trained from large external corpora or randomly initialized. We argue that affective commonsense knowledge and words indicative of sentiment could be used to learn better target and aspect embeddings. We therefore propose an embedding refinement framework called RAEC (Refining Affective Embedding from Context), in which sentiment concepts extracted from affective commonsense knowledge and word relative location information are incorporated to derive context-affective embeddings. Furthermore, a sparse coefficient vector is exploited in refining the embeddings of targets and aspects separately. In this way, embeddings of targets and aspects can capture the highly relevant affective words. Experimental results on two benchmark datasets show that our framework can be easily integrated with existing embedding-based TABSA models and achieves state-of-the-art results compared to models relying on pre-trained word embeddings or built on other embedding refinement methods.

show abstract

Joint Multi-Label Attention Networks for Social Text Annotation

Cited by 10 publications

References 23 publications

Automated Social Text Annotation With Joint Multilabel Attention Networks

Automated Social Text Annotation With Joint Multilabel Attention Networks

Context-Aware Human Activity and Smartphone Position-Mining with Motion Sensors

Embedding Refinement Framework for Targeted Aspect-Based Sentiment Analysis

Contact Info

Product

Resources

About