Semantic and syntactic analysis in learning representation based on a sentiment analysis model

Vo, Anh-Dung; Nguyen, Quang-Phuoc; Ock, Cheol-Young

doi:10.1007/s10489-019-01540-2

Cited by 29 publications

(6 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Principle of the BRFP Model. In the study of sentence embedding, syntactic information is an important aspect [36] because, from a grammatical point of view, the overall semantics of a sentence not only depends on the independent semantics of each word in the sentence [37] but also generates new semantics, even opposite semantics under the action of syntax.…”

Section: Methodsmentioning

confidence: 99%

BRFP: An Efficient and Universal Sentence Embedding Learning Model Method Supporting Fused Syntax Combined with Graph Embedding Representation Algorithm

Shen

2022

Journal of Sensors

View full text Add to dashboard Cite

Due to the rapidly growing volume of data on the Internet, the methods of efficiently and accurately processing massive text information have been the focus of research. In natural language processing theory, sentence embedding representation is an important method. This paper proposes a new sentence embedding learning model called BRFP (Factorization Process with Bidirectional Restraints) that fuses syntactic information, uses matrix decomposition to learn syntactic information, and fuses and calculates with word vectors to obtain the embedded representation of sentences. In the experimental chapter, text similarity experiments are conducted to verify the rationality and effectiveness of the model and analyzed experimental results on Chinese and English texts with the current mainstream learning methods, and potential improvement directions are summarized. The experimental results on Chinese and English datasets, including STS, AFQMC, and LCQMC, show that the model proposed in this paper outperforms the CNN method in terms of accuracy and F1 value by 7.6% and 4.8. The comparison experiment with the word vector weighted model shows that when the sentence length is longer, or the corresponding syntactic structure is complex, the model’s advantages in this paper are more prominent than TF-IDF and SIF methods. Compared with the TF-IDF method, the effect improved by 14.4%. Compared with the SIF method, it has a maximum advantage of 7.9%, and the overall improvement in each comparative experimental task is between 4 and 6 percentage points. In the neural network model comparison experiment, the model in this paper compared the CNN, RNN, LSTM, ST, QT, and InferSent models, and the effect significantly improved on the 14’OnWN, 14’Tweet-news, and 15’Ans.-forum datasets. For example, in the 14’OnWN dataset, the BRFP method has a 10.9% improvement over the ST method. The 14’Tweet-news dataset has a 22.9% advantage over the LSTM method, and the 15’Ans.-forum dataset has a 24.07% improvement over the RNN method. The article also demonstrates the generality of the model, proving that the model proposed in this paper is also a universal learning framework.

show abstract

Section: Methodsmentioning

confidence: 99%

BRFP: An Efficient and Universal Sentence Embedding Learning Model Method Supporting Fused Syntax Combined with Graph Embedding Representation Algorithm

Shen

2022

Journal of Sensors

View full text Add to dashboard Cite

show abstract

“…Suppose that the dataset of the news images has x classes, each class has y graphs, and the feature dimension is Z. Then the standard deviations of class a and the dimension b are shown in equation ( 4) [24,25].…”

Section: Feature Extractionmentioning

confidence: 99%

Automatic Semantic Annotation of News Images in Mobile Internet of Things and Construction of Semantic Internet of Things System

Wang

Ling

Chen

2022

Preprint

View full text Add to dashboard Cite

With the continuous development of mobile and semantic IoT (Internet of Things), the number of news images is increasing sharply. Automatic semantic annotation technology is an important tool to manage and retrieve news images. Therefore, the study is focused on the automatic semantic annotation of news images. First, the problems of the existing automatic annotation methods of news images are analyzed and studied, and then an automatic semantic annotation method of news images based on the fusion of weight and features is proposed. The proposed method extracts the colour features, texture features, and shape features of news images to determine their stability by using the standard deviation of features, and then calculates the weighting coefficient to realize automatic annotation of the news images. The results of the study in the corel5k dataset show that the accuracy of the weight-feature fusion annotation model is 60% for the colour feature, 62% for the texture feature, and 67% for the mixture of colour, shape, and texture. The accuracy of the fusion of weight and feature annotation method is 7% higher than that of the single feature annotation method. The maximum accuracy of the fusion of weight and feature annotation method is 91.2%. The study provides a reference for the research of automatic annotation of news images.

show abstract

“…Chen and Huang 7 researched aspect-level sentiment classification by incorporating external knowledge into a neural network and proposed a novel framework for model aspect-opinion pair identification. Vo et al 8 studied a knowledge representation approach that centers on aspect rating and weighting and proposed a novel model that utilizes semantic and syntactic components to capture semantic and sentimental information. Sentiment words present the most sentiment information in text, so a sentiment dictionary is critical for explicit sentiment analysis.…”

Section: Sentiment Analysismentioning

confidence: 99%

Implicit sentiment analysis based on graph attention neural network

Yang

Xing

et al. 2021

Engineering Reports

View full text Add to dashboard Cite

Sentiment analysis is one of the crucial tasks in the field of natural language processing. Implicit sentiment suffers a significant challenge because the sentence does not include explicit emotional words and emotional expression is vague. This paper proposed a novel implicit sentiment analysis model based on graph attention convolutional neural network. A graph convolutional neural network is used to propagate semantic information. The attention mechanism is employed to compute the contribution to the emotional expression of words.In order to solve the problem of multiple attention preserving repeated information, orthogonal attention constraint was used to make different attention store different emotional information; given the uneven distribution of emotional information, score attention constraint was proposed to make the model focus on a limited number of essential words. The performance of the proposed model was verified on implicit sentiment datasets. The F value reached 88.16%, which is higher than the benchmark model in the literature. The attention mechanism is analyzed to verify the effectiveness of orthogonal constraint and score constraint.

show abstract

Semantic and syntactic analysis in learning representation based on a sentiment analysis model

Cited by 29 publications

References 44 publications

BRFP: An Efficient and Universal Sentence Embedding Learning Model Method Supporting Fused Syntax Combined with Graph Embedding Representation Algorithm

BRFP: An Efficient and Universal Sentence Embedding Learning Model Method Supporting Fused Syntax Combined with Graph Embedding Representation Algorithm

Automatic Semantic Annotation of News Images in Mobile Internet of Things and Construction of Semantic Internet of Things System

Implicit sentiment analysis based on graph attention neural network

Contact Info

Product

Resources

About