Challenges and Issues in Sentiment Analysis: A Comprehensive Survey

Raghunathan, Nilaa; Saravanakumar, Kandasamy

doi:10.1109/access.2023.3293041

Cited by 11 publications

(2 citation statements)

References 139 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Domain dependency was revealed by this examination, which is crucial for diagnosing sentiment problems. Raghunathan and Saravanakumar [15], explored lexicon and ML-based methods for sentiment evaluation. They looked at the difficulties of four distinct kinds of sentiment categorization in this study: cross-lingual, cross-domain, small-scale, and short-term applications.…”

Section: Literature Surveymentioning

confidence: 99%

Aspect term extraction from multi-source domain using enhanced latent Dirichlet allocation

Dhanal,

Ghorpade

2024

IJEECS

View full text Add to dashboard Cite

This study presents a comprehensive exploration of sentiment analysis across diverse domains through the introduction of a multi-source domain dataset encompassing hospitals, laptops, restaurants, cell phones, and electronics. Leveraging this extensive dataset, an enhanced latent Dirichlet allocation (E-LDA) model is proposed for topic modeling and aspect extraction, demonstrating superior performance with a remarkable coherence score of 0.5727. Comparative analyses with traditional LDA and other existing models showcase the efficacy of E-LDA in capturing sentiments and specific attributes within different domains. The extracted topics and aspects reveal valuable insights into domain-specific sentiments and aspects, contributing to the advancement of sentiment analysis methodologies. The findings underscore the significance of considering multi-source datasets for a more holistic understanding of sentiment in diverse text corpora.

show abstract

Section: Literature Surveymentioning

confidence: 99%

Aspect term extraction from multi-source domain using enhanced latent Dirichlet allocation

Dhanal,

Ghorpade

2024

IJEECS

View full text Add to dashboard Cite

show abstract

“…Currently, sentiment analysis is a research hotspot [2] that involves obtaining people's attitudes from natural language fields, such as text and images. Text is the most concise and direct way to convey emotions, and we express emotions and thoughts mainly through words and language; therefore, the most direct and meaningful way to understand netizens' attitudes towards things is from the emotions expressed in text.…”

Section: Introductionmentioning

confidence: 99%

Chinese Multicategory Sentiment of E-Commerce Analysis Based on Deep Learning

Li,

Wang,

et al. 2023

Electronics

View full text Add to dashboard Cite

With the continuous rise of information technology and social networks, and the explosive growth of network text information, text sentiment analysis technology now plays a vital role in public opinion monitoring and product development analysis on networks. Text data are high-dimensional and complex, and traditional binary classification can only classify sentiment from positive or negative aspects. This does not fully cover the various emotions of users, and, therefore, natural language semantic sentiment analysis has limitations. To solve this deficiency, we propose a new model for analyzing text sentiment that combines deep learning and the bidirectional encoder representation from transformers (BERT) model. We first use an advanced BERT language model to convert the input text into dynamic word vectors; then, we adopt a convolutional neural network (CNN) to obtain the relatively significant partial emotional characteristics of the text. After extraction, we use the bidirectional recurrent neural network (BiGRU) to bidirectionally capture the contextual feature message of the text. Finally, with the MultiHeadAttention mechanism we obtain correlations among the data in different information spaces from different subspaces so that the key information related to emotion in the text can be selectively extracted. The final emotional feature representation obtained is classified using Softmax. Compared with other similar existing methods, our model in this research paper showed a good effect in comparative experiments on an e-commerce text dataset, and the accuracy and F1-score of the classification were significantly improved.

show abstract