A Stabilized Dense Network Approach for High-Dimensional Prediction

Yang, Wenjing; Yang, Yuhong

doi:10.1109/ijcnn52387.2021.9533554

Cited by 3 publications

(1 citation statement)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Conventional methods of sentiment classification are mostly based on statistical machine learning [16,21,22]. The representative methods of this line are structure based and spectral clustering methods.…”

Section: Conventional Cross-domain Sentiment Classificationmentioning

confidence: 99%

Big Data Based Transfer Learning for Sentiment Classification with Multiple Source Domains

Li¹,

Wang²,

Yang³

2022

Proceedings of the 2022 3rd International Conference on Big Data and Informatization Education (ICBDIE 2022)

View full text Add to dashboard Cite

Sentiment classification, served as a curial technology in natural language processing and computational linguistics, has drawn a lot of attentions from researchers. However, due to the high cost of manual labeling in the era of big data, conventional methods of sentiment classification are unqualified to be employed in a new domain directly. Hence, in this paper, we explore big data based transfer learning for sentiment classification with multiple source domains. To solve the problem of inherent domain gap, we propose a novel framework Adversarial crossdomain sentiment classification with weighted domain-dependent fEAture learning dubbed AdEA. Specifically, AdEA involves an individual domain-invariant feature extractor and several domain-dependent feature extractors. To obtain the domain-invariant feature, we use a reversed discriminator loss for these extractors. Furthermore, we propose a weighted learning module to reinforce the relationship of domain-dependent features between source and target domains. Integrated with these two domain-related features, AdEA is able to achieve better capability of cross-domain sentiment classification. Experimental results show the effectiveness of our proposed method.

show abstract

Section: Conventional Cross-domain Sentiment Classificationmentioning

confidence: 99%