Combining Convolution Neural Network and Bidirectional Gated Recurrent Unit for Sentence Semantic Classification

Zhang, Dejun; Tian, Long; Hong, Mingbo; Han, Fei; Ren, Yafeng; Chen, Yilin

doi:10.1109/access.2018.2882878

Cited by 49 publications

(32 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although this study combined BiGRU and CNN, it could not offer promising results. Contrary to BiGRU+CNN, our model carries solely a BiLSTM network and provided 1.2 percentage points higher accuracy than [36]. Finally, Usama et al [37] presented various models by encountering multilevel and multitype fusion of features along with the combination of LSTM, GRU, and CNN models.…”

Section: A Mr Resultsmentioning

confidence: 99%

“…Conversely, our approach used a single-layered BiLSTM network and achieved 0.5 and 0.6 percentage points greater accuracy than ALE-LSTM and WALE-LSTM, respectively. Also, Zhang et al [36] suggested a novel architecture by leveraging the series combination of bidirectional gated recurrent unit (BiGRU) and CNN (BiGRU+CNN) and achieved an accuracy of 78.30% on MR dataset. Although this study combined BiGRU and CNN, it could not offer promising results.…”

Section: A Mr Resultsmentioning

confidence: 99%

“…To this end, we focused on the simplicity and stability of the architecture, by taking into account only the long-term dependency in a text. We concluded that using a pre-trained embedding with a single-layer BiLSTM along with global pooling concept could produce competitive results, compared to recently presented approaches [36]- [41]. In the following subsections, we introduced the LSTM network along with its associated mathematical details.…”

Section: B Deep Learning Approachmentioning

confidence: 99%

See 2 more Smart Citations

Sentiment Classification Using a Single-Layered BiLSTM Model

2020

View full text Add to dashboard Cite

This study presents a computationally efficient deep learning model for binary sentiment classification, which aims to decide the sentiment polarity of people's opinions, attitudes, and emotions expressed in written text. To achieve this, we exploited three widely practiced datasets based on public opinions about movies. We utilized merely one bidirectional long short-term memory (BiLSTM) layer along with a global pooling mechanism and achieved an accuracy of 80.500%, 85.780%, and 90.585% on MR, SST2 and IMDb datasets, respectively. We concluded that the performance metrics of our proposed approach are competitive with the recently published models, having comparatively complex architectures. Also, it is inferred that the proposed single-layered BiLSTM based architecture is computationally efficient and can be recommended for real-time applications in the field of sentiment analysis. INDEX TERMS Bidirectional long short-term memory, deep learning, long-term dependencies, natural language processing, sentiment analysis.

show abstract

Section: A Mr Resultsmentioning

confidence: 99%

Section: A Mr Resultsmentioning

confidence: 99%

Section: B Deep Learning Approachmentioning

confidence: 99%

See 1 more Smart Citation

Sentiment Classification Using a Single-Layered BiLSTM Model

2020

View full text Add to dashboard Cite

show abstract

“…To obtain and filter the features of the relationship between the two parts, the BiLSTM model is applied, and it can filter the features of the discourse and obtain a low-dimensional vector to reduce the complexity of the model. Other RNN models, such as BGRU [31], can also extract text features in the discourse semantic analysis task. However, in order to compare our results with the baseline models, which applied LSTM or BiLSTM as their main network, we choose BiLSTM as the feature extraction network of our model.…”

Section: B Feature Fusionmentioning

confidence: 99%

“…where i represents the number of the current layer. Lin et al [31] added a modulating factor (1 − p i ) γ to the cross entropy loss, with a tunable focusing parameter γ ≥ 0, and α i ∈ [0, 1] balances the importance of positive and negative examples. Because the data distribution features of each child node are similar, the feature screening is the same as in (5).…”

Section: Improved Loss Functionmentioning

confidence: 99%

Tree Framework With BERT Word Embedding for the Recognition of Chinese Implicit Discourse Relations

Jiang

2020

IEEE Access

View full text Add to dashboard Cite

Currently, discourse relation recognition (DRR), which is not directly marked with connectives, is a challenging task. Traditional approaches for implicit DRR in Chinese have focused on exploring the concepts and features of words; however, these approaches have only yielded slow progress. Moreover, the lack of Chinese labeled data makes it more difficult to complete this task with high accuracy. To address this issue, we propose a novel hybrid DRR model combining a pretrained language model, namely bidirectional encoder representations from transformers (BERT), with recurrent neural networks. We use BERT as a text representation and pretraining model. In addition, we apply a tree structure to the implicit DRR in Chinese to produce hierarchical classes. The 19-class F1 score of our proposed method can reach 74.47% on the HIT-CIR Chinese discourse relation corpus. The attained results showed that the use of BERT and the proposed tree structure forms a novel and precise method that can automatically recognize the implicit relations of Chinese discourse.

show abstract

Intelligent pig‐raising knowledge question‐answering system based on neural network schemes

Kung

Chen

et al. 2021

Agronomy Journal

View full text Add to dashboard Cite

The most important parts of Taiwan's agriculture are animal husbandry and the pig industry, of which the output value reached NT$75.6 billion in 2017. Taiwan has a high technical level of pig raising. However, practical pig-raising skills rely mainly on the inheritance of mentoring experience. The livestock and pig breeding industry in Taiwan has no relevant pig breeding knowledge management information system and no intelligent knowledge question-answering system. Therefore, this study designs and implements an intelligent knowledge question-and-answer system for pig farming. To identify intelligent questions and answers for raising livestock pigs, this study addresses the following issues: (a) to determine the semantic meaning of a sentence, the system needs to accurately interpret the meaning of a question and to identify the expression of a knowledge entity. Therefore, this study applies the lattice long-shortterm memory (LSTM) and structured perceptron methods to parse sentences accurately and correctly perform word segmentation. A set of stop words for pig raising in Taiwan is initially established to ensure accurate sentence parsing; (b) to understand the intent of each sentence, the bidirectional gated recurrent init (bi-GRU) method is adopted to realize the knowledge extraction of the livestock in the question and complete the intent detection and slot filling. The bi-GRU extracts the correct livestock knowledge and classifies it into suitable topics; (c) the wide range of knowledge sources in questions often leads to unrelated vocabulary, repeated words, and structural loss in potential answers. To establish an effective knowledge search graph, the method infers implicit knowledge from the existing knowledge and conducts related knowledge retrieval based on the inference results. The knowledge data model is

show abstract

Combining Convolution Neural Network and Bidirectional Gated Recurrent Unit for Sentence Semantic Classification

Cited by 49 publications

References 22 publications

Sentiment Classification Using a Single-Layered BiLSTM Model

Sentiment Classification Using a Single-Layered BiLSTM Model

Tree Framework With BERT Word Embedding for the Recognition of Chinese Implicit Discourse Relations

Intelligent pig‐raising knowledge question‐answering system based on neural network schemes

Contact Info

Product

Resources

About