Multi-source domain adaptation with joint learning for cross-domain sentiment classification

Zhao, Chuanjun; Wang, Suge; Li, Deyu

doi:10.1016/j.knosys.2019.105254

Cited by 43 publications

(17 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then, under the guidance of multigranulation probabilistic models, we utilize the model of INSs, MGRSs and PRSs to handle the above-mentioned challenges. Moreover, compared with existing popular nonlinear modeling approaches, such as formal concept analysis [33,34,58,[64][65][66][67], control systems [59,60] and sentiment analysis [61][62][63]68,69], which neither effectively handle indeterminate and incomplete information in complicated MAGDM problems, nor reasonably fuse and analyze multi-source information with incorrect and noisy data, it is necessary to combine INSs, MGRSs with PRSs to develop some meaningful hybrid models along with corresponding MAGDM approaches. In light of MAGDM procedures in the current section, we sum up the merits of the proposed MAGDM algorithm below:…”

Section: Discussionmentioning

confidence: 99%

“…Gil et al [60] studied a surrogate model based optimization of traffic lights cycles and green period ratios by means of microscopic simulation and fuzzy rule interpolation. Smarandache et al [61] explored word-level sentiment similarities in the context of NSs, and some other meaningful works on word-level sentiment analysis were also investigated recently [62,63,68,69].…”

Section: The Contributions Of the Researchmentioning

confidence: 99%

See 1 more Smart Citation

Multi-Attribute Group Decision Making Based on Multigranulation Probabilistic Models with Interval-Valued Neutrosophic Information

Zhang

Kang

et al. 2020

Mathematics

Self Cite

View full text Add to dashboard Cite

In plenty of realistic situations, multi-attribute group decision-making (MAGDM) is ubiquitous and significant in daily activities of individuals and organizations. Among diverse tools for coping with MAGDM, granular computing-based approaches constitute a series of viable and efficient theories by means of multi-view problem solving strategies. In this paper, in order to handle MAGDM issues with interval-valued neutrosophic (IN) information, we adopt one of the granular computing (GrC)-based approaches, known as multigranulation probabilistic models, to address IN MAGDM problems. More specifically, after revisiting the related fundamental knowledge, three types of IN multigranulation probabilistic models are designed at first. Then, some key properties of the developed theoretical models are explored. Afterwards, a MAGDM algorithm for merger and acquisition target selections (M&A TSs) with IN information is summed up. Finally, a real-life case study together with several detailed discussions is investigated to present the validity of the developed models.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: The Contributions Of the Researchmentioning

confidence: 99%

Multi-Attribute Group Decision Making Based on Multigranulation Probabilistic Models with Interval-Valued Neutrosophic Information

Zhang

Kang

et al. 2020

Mathematics

Self Cite

View full text Add to dashboard Cite

show abstract

“…Both models work well due to the weight reduction for non-sentiment parts from a sentence. Zhao et al [41] discuss multi-source domain adaptation for a cross-domain sentiment classification task. Their paper shows that joint learning for cross-domain tasks leads to good results and a greater generalisation capability, while at the same time enabling deep domain fusion.…”

Section: Deep Learning For Affective Classificationmentioning

confidence: 99%

Automatic Expansion of Domain-Specific Affective Models for Web Intelligence Applications

et al. 2021

View full text Add to dashboard Cite

Sentic computing relies on well-defined affective models of different complexity—polarity to distinguish positive and negative sentiment, for example, or more nuanced models to capture expressions of human emotions. When used to measure communication success, even the most granular affective model combined with sophisticated machine learning approaches may not fully capture an organisation’s strategic positioning goals. Such goals often deviate from the assumptions of standardised affective models. While certain emotions such as Joy and Trust typically represent desirable brand associations, specific communication goals formulated by marketing professionals often go beyond such standard dimensions. For instance, the brand manager of a television show may consider fear or sadness to be desired emotions for its audience. This article introduces expansion techniques for affective models, combining common and commonsense knowledge available in knowledge graphs with language models and affective reasoning, improving coverage and consistency as well as supporting domain-specific interpretations of emotions. An extensive evaluation compares the performance of different expansion techniques: (i) a quantitative evaluation based on the revisited Hourglass of Emotions model to assess performance on complex models that cover multiple affective categories, using manually compiled gold standard data, and (ii) a qualitative evaluation of a domain-specific affective model for television programme brands. The results of these evaluations demonstrate that the introduced techniques support a variety of embeddings and pre-trained models. The paper concludes with a discussion on applying this approach to other scenarios where affective model resources are scarce.

show abstract

“…In 2019, Zhao et al [18] have established a new approach with "multi-source domain adaptation for CDSC tasks". This model exploited CNN and "bidirectional gated recurrent units" for soft parameter sharing and deep feature extraction for transferring information among tasks.…”

Section: Literature Reviewmentioning

confidence: 99%

Multi-Source Domain Adaptation in Sentiment Analysis using Optimized Neural Network and Cross-Domain Semantic Library

Patel¹

2021

IJIES

View full text Add to dashboard Cite

Nowadays, the rapid growth of internet has led to the way for easiest data generation. Sentiment Analysis (SA) is necessary to scrutinize the user-generated data for classifying the sentiment polarity into anyone of the two classifications (negative and positive). SA is implemented by machine learning techniques and lexicon oriented techniques. However, the traditional SA models could not deal with missing labelled data and internet slang data. This paper aims to propose a new multi-source sentiment analysis model that includes six stages to perform the task of domain adaptation. For checking similarity between statements, higher order statistics based on text features is exploited. The higher order statistics based features extraction includes use of modified cross entropy measure. Remaining features are then given to proposed classifier that predicts the polarity of target domain in a precise way. For classification purpose Neural Network (NN) is exploited. Particularly, the weights of NN are tuned in an optimal manner using Improved Grey Wolf Optimization (IGWO) algorithm, which is the enhanced version of GWO algorithm. Finally, the performance of the proposed model is compared over several conventional methods of domain adaptation and gained an improvement of 28% to 6% in accuracy.

show abstract

Multi-source domain adaptation with joint learning for cross-domain sentiment classification

Cited by 43 publications

References 25 publications

Multi-Attribute Group Decision Making Based on Multigranulation Probabilistic Models with Interval-Valued Neutrosophic Information

Multi-Attribute Group Decision Making Based on Multigranulation Probabilistic Models with Interval-Valued Neutrosophic Information

Automatic Expansion of Domain-Specific Affective Models for Web Intelligence Applications

Multi-Source Domain Adaptation in Sentiment Analysis using Optimized Neural Network and Cross-Domain Semantic Library

Contact Info

Product

Resources

About