A text classification method based on a convolutional and bidirectional long short-term memory model

Huan, Hai; Guo, Zelin; Cai, Tingting; He, Zongjian

doi:10.1080/09540091.2022.2098926

Cited by 27 publications

(10 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Applying LSTM works well with long sequence dependencies. As property vectors, we used 100-element pre-learned global property vectors developed according to the GloVe method, an unsupervised learning algorithm for learning vector representations of various terms [16]. Since our text corpus was relatively small compared to several other parameters used in artificial neural network models, we trained our model using the Stanford Amazon Dataset service [17], which contains 34,686,770 reviews of product and user information, ratings, and plaintext reviews.…”

Section: The Sentiment-based Approach In Body Image Analysismentioning

confidence: 99%

The Use of Natural Language Processing Elements for Computer-Aided Diagnostics and Monitoring of Body Image Perception in Enterally Fed Patients with Head and Neck or Upper Gastrointestinal Tract Cancers

Różańska,

Gliwska,

Barańska

et al. 2024

Cancers

View full text Add to dashboard Cite

Background: Psycho-oncology care has emerged as a significant concern in contemporary oncology practice, given its profound impact on patient psychological well-being. Patients undergoing treatment for head–neck or upper gastrointestinal tract cancers often experience complex emotional and psychological challenges, necessitating specialized support and intervention. Traditional approaches to psycho-oncological care may be limited in their ability to comprehensively assess and address patients’ needs. Therefore, exploring innovative methodologies, such as leveraging natural language processing (NLP) elements, is crucial to enhancing the effectiveness of psycho-oncological interventions. Methods: In this study, we developed a method utilizing natural language processing (NLP) elements to augment psycho-oncological care for patients with head–neck or upper gastrointestinal tract cancers. The method aimed to facilitate vocabulary, sentiment, and intensity analysis of five basic emotions (happiness, sadness, anger, disgust, and fear), as well as to explore potential areas of difficulty such as body image, pain, and self-esteem. We conducted research involving 50 patients across three treatment stages. Results: Our method facilitated the identification of characteristic features at each treatment stage, aiding in the tailoring of appropriate therapies to individual patient needs. The results offer insights valuable to psychologists and psychiatrists for expedited diagnosis and intervention, potentially influencing therapy outcomes. Additionally, the data may inform treatment decisions by addressing patient-specific concerns. Furthermore, our method holds promise for optimizing the allocation of psychological care resources, particularly at the initial stages of patient contact. Limitations: The main problem in the research was the fairly wide age range of participants, which explains the potential diversity of vocabulary. Conclusion: In conclusion, our study demonstrates the potential utility of integrating natural language processing (NLP) elements into psycho-oncological care for patients with head–neck or upper gastrointestinal tract cancers. The developed method offers a novel approach to comprehensively assessing patients’ emotional states and areas of difficulty, thereby facilitating tailored interventions and treatment planning. These findings underscore the importance of continued research and innovation in psycho-oncology to enhance patient care and outcomes.

show abstract

Section: The Sentiment-based Approach In Body Image Analysismentioning

confidence: 99%

The Use of Natural Language Processing Elements for Computer-Aided Diagnostics and Monitoring of Body Image Perception in Enterally Fed Patients with Head and Neck or Upper Gastrointestinal Tract Cancers

Różańska,

Gliwska,

Barańska

et al. 2024

Cancers

View full text Add to dashboard Cite

show abstract

“…Liang et al 27 gave an improved ensemble model for Chinese text classification based on CNN and RNN structure. Guo et al 28 propose a text classification method based on the Convolutional and Bi-LSTM Model (CBM), which can extract both the local shallow semantic features and the global deep semantic features. Liu et al 29 combined Chinese syntactic dependency tree with graph convolution and proposed a new sentiment classification model (dependent tree graph convolution Network, DTGCN), which improved the performance of sentiment classification in this paper.…”

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Feature-enhanced text-inception model for Chinese long text classification

Guo

Jiayu

et al. 2023

Sci Rep

View full text Add to dashboard Cite

To solve the problem regarding unbalanced distribution of multi-category Chinese long texts and improve the classification accuracy thereof, a data enhancement method was proposed. Combined with this method, a feature-enhanced text-inception model for Chinese long text classification was proposed. First, the model used a novel text-inception module to extract important shallow features of the text. Meanwhile, the bidirectional gated recurrent unit (Bi-GRU) and the capsule neural network were employed to form a deep feature extraction module to understand the semantic information in the text; K-MaxPooling was then used to reduce the dimension of its shallow and deep features and enhance the overall features. Finally, the Softmax function was used for classification. By comparing the classification effects with a variety of models, the results show that the model can significantly improve the accuracy of long Chinese text classification and has a strong ability to recognize long Chinese text features. The accuracy of the model is 93.97% when applied to an experimental dataset.

show abstract

“…Many studies combine the advantages of RNN and CNN networks to integrate sequence information and local interaction information to improve text classification performance. Huan [17] et al proposed a model based on CNN and LSTM to extract shallow local semantic features and deep global semantic features of complex semantic texts. To highlight the role of key information in the text, an attention mechanism has been added to the model by some studies.…”

Section: Related Workmentioning

confidence: 99%

Using Recurrent Neural Network Structure and Multi-Head Attention with Convolution for Fraudulent Phone Text Recognition

Zhou¹,

Xu²,

Zhang³

et al. 2023

Computer Systems Science and Engineering

View full text Add to dashboard Cite

Fraud cases have been a risk in society and people's property security has been greatly threatened. In recent studies, many promising algorithms have been developed for social media offensive text recognition as well as sentiment analysis. These algorithms are also suitable for fraudulent phone text recognition. Compared to these tasks, the semantics of fraudulent words are more complex and more difficult to distinguish. Recurrent Neural Networks (RNN), the variants of RNN, Convolutional Neural Networks (CNN), and hybrid neural networks to extract text features are used by most text classification research. However, a single network or a simple network combination cannot obtain rich characteristic knowledge of fraudulent phone texts relatively. Therefore, a new model is proposed in this paper. In the fraudulent phone text, the knowledge that can be learned by the model includes the sequence structure of sentences, the correlation between words, the correlation of contextual semantics, the feature of keywords in sentences, etc. The new model combines a bidirectional Long-Short Term Memory Neural Network (BiLSTM) or a bidirectional Gate Recurrent United (BiGRU) and a Multi-Head attention mechanism module with convolution. A normalization layer is added after the output of the final hidden layer. BiLSTM or BiGRU is used to build the encoding and decoding layer. Multi-head attention mechanism module with convolution (MHAC) enhances the ability of the model to learn global interaction information and multi-granularity local interaction information in fraudulent sentences. A fraudulent phone text dataset is produced by us in this paper. The THUCNews data sets and fraudulent phone text data sets are used in experiments. Experiment results show that compared with the baseline model, the proposed model (LMHACL) has the best experiment results in terms of Accuracy, Precision, Recall, and F1 score on the two data sets. And the performance indexes on fraudulent phone text data sets are all above 0.94.

show abstract

A text classification method based on a convolutional and bidirectional long short-term memory model

Cited by 27 publications

References 24 publications

The Use of Natural Language Processing Elements for Computer-Aided Diagnostics and Monitoring of Body Image Perception in Enterally Fed Patients with Head and Neck or Upper Gastrointestinal Tract Cancers

The Use of Natural Language Processing Elements for Computer-Aided Diagnostics and Monitoring of Body Image Perception in Enterally Fed Patients with Head and Neck or Upper Gastrointestinal Tract Cancers

Feature-enhanced text-inception model for Chinese long text classification

Using Recurrent Neural Network Structure and Multi-Head Attention with Convolution for Fraudulent Phone Text Recognition

Contact Info

Product

Resources

About