Intelligent fake reviews detection based on aspect extraction and analysis using deep learning

Bathla, Gourav; Singh, Pardeep; Singh, Robert P.; Wang, Zhaoxia; Tiwari, Rajeev

doi:10.1007/s00521-022-07531-8

Cited by 46 publications

(13 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The aspects' importance may vary in the same domain. Noun/noun phrases are worked as an aspect that is explained in many researches 12 . There are numbers of methods are available for aspect extraction such as POS tagging, n‐gram, and deep learning approaches.…”

Section: Proposed Approachmentioning

confidence: 99%

“…Noun/noun phrases are worked as an aspect that is explained in many researches. 12 There are numbers of methods are available for aspect extraction such as POS tagging, n-gram, and deep learning approaches. In order to extract all the aspects, we utilize the framework which is proposed in Reference 56.…”

Section: Problem Description and Input Embedding Layermentioning

confidence: 99%

“…10 Deep learning models have been effectively employed for numerous issues of NLP tasks and obtained improvement in the average accuracy. 11,12 The deep learning models for the NLP tasks utilize a pretrained word vector at the initial phase of the model and the remaining phase of the model is trained on task-oriented data along with recurrent and/or convolutional neural networks. The popular pretrained word vectors are word2vec, 13 GloVe, 14 and FastText.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Cross‐domain sentiment classification using decoding‐enhanced bidirectional encoder representations from transformers with disentangled attention

Singh

Sachan

Patel

2022

Concurrency and Computation

View full text Add to dashboard Cite

Cross-domain sentiment classification is a significant task of sentiment analysis that objectives to predict the opinion orientation of text documents in the target domain by using the source domain's learned classifier. Most of the existing approaches of domain-adaptation in sentiment classification focus on sharing low-dimensional features across the domain using domain independent and specific features to mitigate the gap between domains. Earlier cross-domain sentiment classification approaches mainly focused on document level and sentence level, they cannot consider the full impact of aspect words, position of the words, and long-term dependencies. To address this concern, we propose a model for cross-domain sentiment classification, which is based on decoding-enhanced BERT with disentangled attention (DeBERTa). DeBERTa is a pretrained language model based on transformer architecture. In this article, we perform sentence and aspect embedding to mine wordpiece information from text document. DeBERTa language-model utilize disentangled attention mechanism and an enhanced mask decoder to understand the expression features. Disentangled attention mechanism is used to encode each word into two vectors (i.e., content and position vector). In order to predict the masked tokens during model pretraining, an enhanced mask decoder is employed, which incorporates absolute positions in the decoding layer.Finally, experiments are conducted on the benchmark dataset that demonstrates the superiority of fine-tuned DeBERTa model for cross-domain sentiment classification tasks.

show abstract

Section: Proposed Approachmentioning

confidence: 99%

Section: Problem Description and Input Embedding Layermentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Cross‐domain sentiment classification using decoding‐enhanced bidirectional encoder representations from transformers with disentangled attention

Singh

Sachan

Patel

2022

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…(Chen et al (2013); Zeng et al (2014); Wang et al (2014)). At present, many foreign scholars study the identification of Internet water armies for false comments (Jabeur et al (2023); Bathla et al (2022); Vidanagama et al (2022); He et al (2022); Lee et al (2022)); domestic scholars mainly study the identification of Internet water armies and the governance of rumors (He et al (2023); Li et al (2014); Peng et al (2023); Zhang et al (2019); Yan et al (2023); Chen and Du (2023); Zhang et al (2023)). However, most of these researches use different algorithms to accurately identify Internet water armies, and there is a lack of research on how Internet water armies promote the development of Internet public opinion communication.…”

Section: Introductionmentioning

confidence: 99%

A study on the propagation of online public opinion by internet water army

Nian,

Wang,

Zhang

et al. 2024

Soc. Netw. Anal. Min.

View full text Add to dashboard Cite

In the dissemination of information on the Internet, internet water armies have gradually become an important influencing factor in the direction of public opinion. This paper takes Weibo as the research background, obtains real data, establishes a communication dynamics model, and quantitatively researches the influence of internet water armies on the dissemination of public opinion in social networks. Firstly, this paper obtains real hot events in Weibo, analyzes the number of retweets and comments under the same hot topics, and finds the users of the internet water armies. Secondly, this paper establishes a new social network SIwIcS propagation model based on the propagation mechanism among users in Weibo, combining with the SIS infectious disease model, and introducing the impulse effect, the heat effect, and the herd effect, to realize the research of public opinion propagation in Weibo. After that, the SIwIcS propagation model is applied to small-world and scale-free networks to simulate the propagation of public opinion in social networks and analyze the influence of internet water armies on the propagation. The experimental results summarize the influence of network water armies on the propagation of public opinion. Finally, the acquisition of real data is utilized to verify the rationality and validity of the model. This paper quantitatively considers the driving nature of internet water armies on the trend of public opinion in the communication dynamics model, which provides a reasonable decision-making basis for the dissemination and control of public opinion in social networks, and has important theoretical and practical significance.

show abstract

“…The rise of online commerce has revolutionized how people buy and sell goods and services [1]. Customers increasingly rely on these platforms, sharing their thoughts and experiences through reviews after making purchases [2].…”

Section: Introductionmentioning

confidence: 99%

Transformer-Based Language Deep Learning Detection of Fake Reviews on Online Products

Maysara Mazin Badr Alsaad

2024

jes

View full text Add to dashboard Cite

In e-commerce, spotting fake reviews is vital for ensuring trust among consumers. However, identifying them poses a challenge because fake reviews are often crafted to seem genuine, and the sheer volume makes thorough checks difficult. Prior methods involve basic strategies like grammar checks or pattern analysis, but they fell short due to fake reviews generation methods are becoming increasingly sophisticated. Even machine learning, while helpful, struggle to pinpoint subtle fake reviews accurately. This has led to a shift toward deep learning algorithms, which show promise in handling the complexities that traditional methods cannot manage accurately. Specifically, transformer models like BERT, RoBERTa, and XLNet have emerged as potential solutions. This study evaluates the effectiveness of these models in distinguishing between human-generated and computer-generated reviews. RoBERTa displays high accuracy but requires longer learning periods, while BERT and XLNet offer decent accuracy with varying error rates. The investigation delves into these deep learning models to ascertain their capability to spot fake reviews across different scenarios within online platforms. RoBERTa achieves the highest accuracy among the models, reaching 97.1%. It also demonstrates a lower Type I error rate at 2.2%, although its Type II error remains at a moderate level.

show abstract

Intelligent fake reviews detection based on aspect extraction and analysis using deep learning

Cited by 46 publications

References 43 publications

Cross‐domain sentiment classification using decoding‐enhanced bidirectional encoder representations from transformers with disentangled attention

Cross‐domain sentiment classification using decoding‐enhanced bidirectional encoder representations from transformers with disentangled attention

A study on the propagation of online public opinion by internet water army

Transformer-Based Language Deep Learning Detection of Fake Reviews on Online Products

Contact Info

Product

Resources

About