An Improved Self Attention Mechanism Based on Optimized BERT-BiLSTM Model for Accurate Polarity Prediction

Shobana, J.; Murali, M.

doi:10.1093/comjnl/bxac013

Cited by 15 publications

(9 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This study collects a subset of tourist attraction reviews from Ctrip.com, serving as the primary data source. The use of web crawling technology, a contemporary method for acquiring experimental data in text mining, not only streamlines data collection but also establishes the foundation for subsequent research efforts [36][37][38]. The process of extracting tourist attraction review information is illustrated in Figure 6.…”

Section: Construction Of Smart Tourism Text Classification Modelmentioning

confidence: 99%

The Convolutional Neural Network Text Classification Algorithm in the Information Management of Smart Tourism Based on Internet of Things

Meng

2024

IEEE Access

View full text Add to dashboard Cite

The relentless progression of advanced technologies has driven the seamless integration of Internet of Things (IoT) services into the fundamental framework of contemporary tourism enterprises. In the quest for valuable insights from the vast reservoir of tourism data, this study employs a Convolutional Neural Network (CNN) as its primary instrument, culminating in the development of a Convolutional Neural Network-Bidirectional Long Short-Term Memory (CNN-BiLSTM) model designed for the classification of textual data in the smart tourism domain. The primary function of this model is to conduct sentiment analysis on smart tourism data, specifically by categorizing review data into either positive or negative sentiments. The incorporation of review ratings further aids in the accurate labeling of data according to their respective sentiment categories, thus streamlining the process of effective data annotation. The empirical findings reveal distinctive performance trajectories for the three models concerning positive comments across various evaluation metrics. Remarkably, in terms of precision, the CNN-BiLSTM model leads with an impressive 87.9%, closely followed by the BiLSTM model at 87.4%, while the Text Convolutional Neural Network (TextCNN) model trails slightly at 85.3%. Similarly, the recall highlights the CNN-BiLSTM model's excellence, achieving an impressive 88.34%, compared to 78.5% for BiLSTM and 77.6% for TextCNN models. In the realm of accuracy, the CNN-BiLSTM model maintains its dominance at 85.1%, while the BiLSTM and TextCNN models achieve 82.7% and 81.4%, respectively. Notably, the CNN-BiLSTM model outperforms the trio in terms of the F1 value, securing a robust 85.41%. In summary, the CNN-BiLSTM model consistently demonstrates excellent performance across a range of metrics, including precision, recall, accuracy, and F1 value, signifying its supremacy in this classification task. This study presents a systematic solution for enhancing smart tourism services, thereby providing a strong foundation for the growth and advancement of tourism enterprises.

show abstract

Section: Construction Of Smart Tourism Text Classification Modelmentioning

confidence: 99%

The Convolutional Neural Network Text Classification Algorithm in the Information Management of Smart Tourism Based on Internet of Things

Meng

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“… In the data classification stage, the Bi-LSTM algorithm [14] was used for text classification. Based on the emotional polarity, the UGR data were divided into positive data and negative data.…”

Section: Experimental Processmentioning

confidence: 99%

An Analysis and Modeling Method for Design Information Based on User Generated Reviews Mining and Extenics

Duan,

Zhang,

Bai

et al. 2024

Frontiers in Artificial Intelligence and Applications

View full text Add to dashboard Cite

To address the inapplicability of data classification results in industrial design, and the lack of theoretical research on design information analysis methods, a design information analysis method based on user experience data was proposed. User experience positive and negative user experience datasets were established through user experience data mining and applying text classification algorithms to perform initial processing of the data. Design features are obtained by extracting effective information from the data, and a design feature analysis is established by combining with the extension class element theory model. Effective information about users and products in the data is classified and analyzed; a conjugate relation between user information and product information is studied, and a design information correlation transformation model is established based on the extension compound theory. Finally, user and product information correlation analyses are combined to obtain an effective product design strategy. The effectiveness of this method is verified by taking electric scooters used by the elderly as an example.

show abstract

“…By combining different parallel blocks of a single-layer deep convolutional neural network (CNN) with different kernel sizes and filters with BERT, a BERT-based deep learning method (FakeBERT) was proposed for the detection of fake news in social media [21]. Shobana et al (2022) improved self-attention mechanism is added with BiLSTM for focusing on significant words in the context. To enhance the performance of Bidirectional Long Short-Term Memory, the weight parameters of Bi-directional LSTM are optimally selected by using APSO algorithm [22].…”

Section: Related Researchmentioning

confidence: 99%

“…Shobana et al (2022) improved self-attention mechanism is added with BiLSTM for focusing on significant words in the context. To enhance the performance of Bidirectional Long Short-Term Memory, the weight parameters of Bi-directional LSTM are optimally selected by using APSO algorithm [22].…”

Section: Related Researchmentioning

confidence: 99%

Sentiment analysis of hotel online reviews using the BERT model and ERNIE model—Data from China

Yu¹,

Liang

Zhu

2023

PLoS ONE

View full text Add to dashboard Cite

The emotion analysis of hotel online reviews is discussed by using the neural network model BERT, which proves that this method can not only help hotel network platforms fully understand customer needs but also help customers find suitable hotels according to their needs and affordability and help hotel recommendations be more intelligent. Therefore, using the pretraining BERT model, a number of emotion analytical experiments were carried out through fine-tuning, and a model with high classification accuracy was obtained by frequently adjusting the parameters during the experiment. The BERT layer was taken as a word vector layer, and the input text sequence was used as the input to the BERT layer for vector transformation. The output vectors of BERT passed through the corresponding neural network and were then classified by the softmax activation function. ERNIE is an enhancement of the BERT layer. Both models can lead to good classification results, but the latter performs better. ERNIE exhibits stronger classification and stability than BERT, which provides a promising research direction for the field of tourism and hotels.

show abstract

An Improved Self Attention Mechanism Based on Optimized BERT-BiLSTM Model for Accurate Polarity Prediction

Cited by 15 publications

References 20 publications

The Convolutional Neural Network Text Classification Algorithm in the Information Management of Smart Tourism Based on Internet of Things

The Convolutional Neural Network Text Classification Algorithm in the Information Management of Smart Tourism Based on Internet of Things

An Analysis and Modeling Method for Design Information Based on User Generated Reviews Mining and Extenics

Sentiment analysis of hotel online reviews using the BERT model and ERNIE model—Data from China

Contact Info

Product

Resources

About