Learning from semantically dependent multi-tasks

Liu, Bin; Xu, Zenglin; Dai, Bo; Bai, Haoli; Fang, Xianghong; Ren, Yazhou; Zhe, Shandian

doi:10.1109/ijcnn.2017.7966296

Cited by 7 publications

(3 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our work is motivated by recent advances in multiview learning [28,27,26,29], multiple kernel learning [46,31], tensor analysis [56,36], graph learning [30], and adversarial learning [55,34]. Compared to previous works on multi-source sentiment analysis, our proposed framework focuses on transferring knowledge from multi-source domains with an ability of effective utilization of unlabeled data in an end-to-end way.…”

Section: Cross-domain Sentiment Analysismentioning

confidence: 99%

“…Nevertheless, the real situations are: (1) there often exist many domains of comments in the meanwhile; and (2) there is an adequate amount of labeled training data for every domain of interest is typically impractical, which makes the further study on how to handle with the limited-resource multiple domains of data necessary and worthwhile. To this end, a series of powerful multi-task models, which consider the system learning for each domain as an independent task so that each task can reinforce and complement each other, is established [52,38,39,12,11,21,59,61,13,37,38,49,36,54]. Among them, a group of representative studies employ a shared-private model [37,52,38,39,12,11,21,59,61], which introduces two feature spaces for any task: one is used to store task-dependent features, the other is used to capture shared features.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Unsupervised Sentiment Analysis by Transferring Multi-source Knowledge

Dai¹,

Liu²,

Zhang³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Background. Sentiment analysis (SA) is an important research area in cognitive computation-thus in-depth studies of patterns of sentiment analysis are necessary. At present, rich-resource data-based SA has been well-developed, while the more challenging and practical multi-source unsupervised SA (i.e. a target-domain SA by transferring from multiple source domains) is seldom studied. The challenges behind this problem mainly locate in the lack of supervision information, the semantic gaps among domains (i.e., domain shifts), and the loss of knowledge. However, existing methods either lack the distinguishable capacity of the semantic gaps among domains or lose private knowledge.Methods. To alleviate these problems, we propose a two-stage domain adaptation framework. In the first stage, a multi-task methodology-based shared-private architecture is employed to explicitly model the domain-common features and the domain-specific features for the labeled source domains. In the second stage, two elaborate mechanisms are embedded in the shared-private architecture to transfer knowledge from multiple source domains. The first mechanism is a selective domain adaptation (SDA) method, which transfers knowledge from the closest source domain. And the second mechanism is a target-oriented ensemble (TOE) method, in which knowledge is transferred through a well-designed ensemble method.Results and Conclusions. Extensive experiment evaluations verify that the performance of the proposed framework outperforms unsupervised state-of-the-art competitors. What can be concluded from the experiments is that transferring from very different distributed source domains may degrade the target-domain performance, and it is crucial to choose proper source domains to transfer from.

show abstract

Section: Cross-domain Sentiment Analysismentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Unsupervised Sentiment Analysis by Transferring Multi-source Knowledge

Dai¹,

Liu²,

Zhang³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…It has been shown that learning multiple related tasks simultaneously can be advantageous relative to learning these tasks independently [34,35]. Multi-task learning (MTL) methods are firstly developed for classification problems and can be classified into two different types: regularization-based learning and joint feature learning.…”

Section: Multi-task Clusteringmentioning

confidence: 99%

Self-paced multi-task clustering

et al. 2019

Self Cite

View full text Add to dashboard Cite

Multi-task clustering (MTC) has attracted a lot of research attentions in machine learning due to its ability in utilizing the relationship among different tasks. Despite the success of traditional MTC models, they are either easy to stuck into local optima, or sensitive to outliers and noisy data. To alleviate these problems, we propose a novel self-paced multi-task clustering (SPMTC) paradigm. In detail, SPMTC progressively selects data examples to train a series of MTC models with increasing complexity, thus highly decreases the risk of trapping into poor local optima. Furthermore, to reduce the negative influence of outliers and noisy data, we design a soft version of SPMTC to further improve the clustering performance. The corresponding SPMTC framework can be easily solved by an alternating optimization method. The proposed model is guaranteed to converge and experiments on real data sets have demonstrated its promising results compared with state-of-the-art multi-task clustering methods.

show abstract

Self-Paced Multi-Task Multi-View Capped-norm Clustering

Ren

Yan

et al. 2018

Neural Information Processing

View full text Add to dashboard Cite

Learning from semantically dependent multi-tasks

Cited by 7 publications

References 30 publications

Unsupervised Sentiment Analysis by Transferring Multi-source Knowledge

Unsupervised Sentiment Analysis by Transferring Multi-source Knowledge

Self-paced multi-task clustering

Self-Paced Multi-Task Multi-View Capped-norm Clustering

Contact Info

Product

Resources

About