Scholarlink

Every day, the amount of words created increases at an exponential rate. Possible sources include: in-app messengers like Telegram and WhatsApp; social mediaSocial media platforms like Instagram and Facebook; online retailers like AmazonAmazon; Google searches; news publishing platforms; and many others. There is a constant flood of textual information from all these sources. The analysis of this kind of information can provide valuable insight for business owners curious about the public’s perception of their product, brand, or service. Many methods have been presented recently to get understanding from these records. Accurate polarity identification of consumer evaluations is, nevertheless, a continuous and intriguing subject due to the difficulties associated with dealing with texts of large size. This makes it difficult to precisely interpret the linguistic data contained in things like customer feedback and social mediaSocial media remarks. Extensive efforts have been made in the past to streamline the process of deriving precise interpretations from this data. The CNNCNN (ConvNet)-LSTMLong short-term memory network-Attention (CLATT) method is a novel approach to sentiment analysisSentiment analysis that utilises a convolutional neural network (CNNConvolutional Neural Network (CNN)) as an encoder and a long short-term memory network (LSTMLong short-term memory network) as a decoder. This method also encompasses data collection, preprocessingPreprocessing, feature encoding, and classification. When analysing such data, it is essential to consider how the data were collected, processed, and organised. Rat Optimisation Algorithm (ROA) is used to fine-tune the hyper-parameters of the suggested model, boosting its classification efficacy. The research evaluated the significance of the proposed models using various textual datasets. Better or at least equivalent results may be shown when using the suggested method of predicting attitudes, and the computational complexity is reduced. The results of this study highlight the critical role that sentiment analysisSentiment analysis plays in deducing useful information from social mediaSocial media posts and customer reviews.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Consumer Sentiment Analysis Using ROA-Based Hybrid Deep Learning Model from Twitter Data

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Sreekanth Rallapalli,
  • K. Jayasheel Kumar,
  • Tarunika Sharma

摘要

Every day, the amount of words created increases at an exponential rate. Possible sources include: in-app messengers like Telegram and WhatsApp; social mediaSocial media platforms like Instagram and Facebook; online retailers like AmazonAmazon; Google searches; news publishing platforms; and many others. There is a constant flood of textual information from all these sources. The analysis of this kind of information can provide valuable insight for business owners curious about the public’s perception of their product, brand, or service. Many methods have been presented recently to get understanding from these records. Accurate polarity identification of consumer evaluations is, nevertheless, a continuous and intriguing subject due to the difficulties associated with dealing with texts of large size. This makes it difficult to precisely interpret the linguistic data contained in things like customer feedback and social mediaSocial media remarks. Extensive efforts have been made in the past to streamline the process of deriving precise interpretations from this data. The CNNCNN (ConvNet)-LSTMLong short-term memory network-Attention (CLATT) method is a novel approach to sentiment analysisSentiment analysis that utilises a convolutional neural network (CNNConvolutional Neural Network (CNN)) as an encoder and a long short-term memory network (LSTMLong short-term memory network) as a decoder. This method also encompasses data collection, preprocessingPreprocessing, feature encoding, and classification. When analysing such data, it is essential to consider how the data were collected, processed, and organised. Rat Optimisation Algorithm (ROA) is used to fine-tune the hyper-parameters of the suggested model, boosting its classification efficacy. The research evaluated the significance of the proposed models using various textual datasets. Better or at least equivalent results may be shown when using the suggested method of predicting attitudes, and the computational complexity is reduced. The results of this study highlight the critical role that sentiment analysisSentiment analysis plays in deducing useful information from social mediaSocial media posts and customer reviews.