Accuracy Of Sentiment Classification Research Articles

Enhancing Social Media Insights: Leveraging Artificial Intelligence for Sentiment Analysis

Social media platforms have become indispensable channels for communication and interaction, offering a wealth of data that can provide valuable insights into public sentiments. However, analyzing sentiment on these platforms poses significant challenges due to the diverse and unstructured nature of user-generated content. Traditional Natural Language Processing (NLP) techniques struggle to accurately classify sentiments expressed through text, images, emoticons, and multimedia elements. Moreover, the informal and nuanced language used in Electronic Word of Mouth (eWOM) further complicates sentiment analysis. In response, this paper explores the role of Artificial Intelligence (AI) in improving sentiment analysis on social media. By leveraging Machine Learning (ML) algorithms trained on large datasets, AI can enhance the accuracy and efficiency of sentiment classification, providing decision-makers with actionable insights into the sentiment landscape of social media. Keywords: Social media, Sentiment analysis, Artificial Intelligence, Machine Learning, Natural Language Processing, Electronic Word of Mouth.

Joint modeling of causal phrases-sentiments-aspects using Hierarchical Pitman Yor Process

Traditional sentiment-aware topic models assume that topic or sentiment transition occurs from either a sentence to the next sentence or from a word to the next word. Such models cannot capture a topic or sentiment transition at phrase boundaries. Further, most of the models adopt a sentiment lexicon to initialize sentiment priors and this approach induces coverage problems. To overcome the above-cited limitations, we have proposed a topic model that extracts aspects, sentiments, and causal phrases simultaneously by leveraging Hierarchical Pitman Yor Process (HPYP) that is modified using a sentiment component, a word-tagger to guide the causal phrase generation and a sentiment prior initialized through a sequential model to address coverage problems. We have evaluated our model on six datasets and found that the proposed model outperforms the baselines in terms of perplexity by 14%, topical coherence by 20%, topic diversity by 5%, sentiment classification task’s accuracy by 4% and, precision, recall and F1 score by 2%. Ablation studies assert that sequence model based sentiment prior initialization results in increasing the accuracy of sentiment classification by 2%.

SENTIMENT ANALYSIS OF WEIBO COMMENTS BASED ON CONVOLUTIONAL NEURAL NETWORK

Sentiment analysis of social media comments plays a crucial role in understanding public opinion and sentiment trends. This study focuses on sentiment analysis of Weibo comments, a popular microblogging platform in China, using a (CNN) approach. We propose a novel methodology that leverages the inherent structural dependencies among comments and users in the Weibo network to capture nuanced sentiment patterns. By representing Weibo comments as a graph, with users and comments as nodes and their interactions as edges, we exploit the relational information to enhance sentiment classification accuracy. Furthermore, we employ attention mechanisms to prioritize influential users and comments in the sentiment analysis process. Through extensive experiments on real Weibo datasets, our proposed CNN-based sentiment analysis framework demonstrates superior performance compared to traditional methods, achieving high accuracy in sentiment classification tasks. This research contributes to advancing sentiment analysis techniques in social media platforms and provides valuable insights for understanding public sentiment dynamics in online communities like Weibo.

Cross-Domain Aspect-Based Sentiment Classification with a Pre-Training and Fine-Tuning Strategy for Low-Resource Domains

Aspect-based sentiment classification (ABSC) is a crucial sub-task of fine-grained sentiment analysis, which aims to predict the sentiment polarity of the given aspects in a sentence as positive, negative, or neutral. Most existing ABSC methods are based on supervised learning. However, these methods rely heavily on fine-grained labeled training data, which can be scarce in low-resource domains, limiting their effectiveness. To overcome this challenge, we propose a low-resource cross-domain aspect-based sentiment classification (CDABSC) approach based on a pre-training and fine-tuning strategy. This approach applies the pre-training and fine-tuning strategy to an advanced deep learning method designed for ABSC, namely the attention-based encoding graph convolutional network (AEGCN) model. Specifically, a high-resource domain is selected as the source domain, and the AEGCN model is pre-trained using a large amount of fine-grained annotated data from the source domain. The optimal parameters of the model are preserved. Subsequently, a low-resource domain is used as the target domain, and the pre-trained model parameters are used as the initial parameters of the target domain model. The target domain is fine-tuned using a small amount of annotated data to adapt the parameters to the target domain model, improving the accuracy of sentiment classification in the low-resource domain. Finally, experimental validation on two domain benchmark datasets, restaurant and laptop, demonstrates significant outperformance of our approach over the baselines in CDABSC Micro-F1.

An Empirical Evaluation of Adapting Hybrid Parameters for CNN-based Sentiment Analysis

Sentiment analysis aims to understand human emotions and perceptions through various machine-learning pipelines. However, feature engineering and inherent semantic gap constraints often hinder conventional machine learning techniques and limit their accuracy. Newer neural network models have been proposed to automate the feature learning process and enrich learned features with word contextual embeddings to identify their semantic orientations to address these challenges. This article aims to analyze the influence of different factors on the accuracy of sentiment classification predictions by employing Feedforward and Convolutional Neural Networks. To assess the performance of these neural network models, we utilize four diverse real-world datasets, namely 50,000 movie reviews from IMDB, 10,662 sentences from LightSide Movie_Reviews, 300 public movie reviews, and 1,600,000 tweets extracted from Sentiment140. We experimentally investigate the impact of exploiting GloVe word embeddings on enriching feature vectors extracted from sentiment sentences. Findings indicate that using larger dimensions of GloVe word embeddings increases the sentiment classification accuracy. In particular, results demonstrate that the accuracy of the CNN with a larger feature map, a smaller filter window, and the ReLU activation function in the convolutional layer was 90.56% using the IMDB dataset. In comparison, it was 80.73% and 77.64% using the sentiment140 and the 300 sentiment sentences dataset, respectively. However, it is worth mentioning that, with large-size sentiment sentences (LightSide’s Movie Reviews) and using the same parameters, only a 64.44% level of accuracy was achieved.

Improved playstore review sentiment classification accuracy with stacking ensemble

In today's digital era, user reviews on the Playstore platform are an invaluable source of information for developers, offering insights that are critical for service improvement. Previous research has explored the application of stacking ensemble methods, such as in the context of predicting depression among university students, to enhance prediction accuracy. However, these studies often do not explicitly detail the data acquisition process, leaving a gap in understanding the applicability of these methods to different domains. This research aims to bridge this gap by applying the stacking ensemble approach to improve the accuracy of sentiment classification in Playstore reviews, with a clear exposition of the data collection method. Utilizing Logistic Regression as the meta classifier, this methodology is executed in several stages. Initially, data was collected from user reviews of online loan applications on Google Playstore, ensuring transparency in the data acquisition process. The data is then classified using three basic models: Random Forest, Naive Bayes, and SVM. The outputs of these models serve as inputs to the Logistic Regression meta model. A comparison of each base model output with the meta model was subsequently carried out. The test results on the Playstore review dataset demonstrated an increase in accuracy, precision, recall, and F1 score compared to using a single model, achieving an accuracy of 87.05%, which surpasses Random Forest (85.6%), Naive Bayes (85.55%), and SVM (86.5%). This indicates the effectiveness of the stacking ensemble method in providing deeper and more accurate insights into user sentiment, overcoming the limitations of single models and previous research by explicitly addressing data acquisition methods.

Word2Vec and SVM Fusion for Advanced Sentiment Analysis on Amazon Reviews

As a matter of fact, sentiment analysis plays a crucial role in different kinds of reviews for online shopping. With this in mind, this study explores the application of Word2Vec and Support Vector Machine (SVM) in sentiment analysis of Amazon reviews. To be specific, initially, this study conducted preprocessing and feature extraction on a large-scale review dataset, revealing deep semantic relationships between words through the Word2Vec model. Subsequently, this study utilized SVM for model training and optimization, achieving efficient and accurate sentiment classification. According to the analysis, preliminary experimental results indicate that this combined method can effectively capture complex patterns and relationships in the text, demonstrating significant advantages in enhancing the accuracy as well as efficiency of sentiment analysis compared to traditional methods. Overall, these results shed light on providing a powerful tool for e-commerce platforms to understand better and analyze consumer opinions and feelings as well as guiding further exploration.

Sentiment Analysis in the Age of Generative AI

In the rapidly advancing age of Generative AI, Large Language Models (LLMs) such as ChatGPT stand at the forefront of disrupting marketing practice and research. This paper presents a comprehensive exploration of LLMs’ proficiency in sentiment analysis, a core task in marketing research for understanding consumer emotions, opinions, and perceptions. We benchmark the performance of three state-of-the-art LLMs, i.e., GPT-3.5, GPT-4, and Llama 2, against established, high-performing transfer learning models. Despite their zero-shot nature, our research reveals that LLMs can not only compete with but in some cases also surpass traditional transfer learning methods in terms of sentiment classification accuracy. We investigate the influence of textual data characteristics and analytical procedures on classification accuracy, shedding light on how data origin, text complexity, and prompting techniques impact LLM performance. We find that linguistic features such as the presence of lengthy, content-laden words improve classification performance, while other features such as single-sentence reviews and less structured social media text documents reduce performance. Further, we explore the explainability of sentiment classifications generated by LLMs. The findings indicate that LLMs, especially Llama 2, offer remarkable classification explanations, highlighting their advanced human-like reasoning capabilities. Collectively, this paper enriches the current understanding of sentiment analysis, providing valuable insights and guidance for the selection of suitable methods by marketing researchers and practitioners in the age of Generative AI.

Social Media Monitoring Of Airbnb Reviews Using AI: A Sentiment Analysis Approach For Immigrant Perspectives In The UK

This paper presents a novel approach for monitoring social media content related to Airbnb reviews, explicitly focusing on the sentiments expressed by immigrants in the United Kingdom. The proposed system, a Quick Search System, leverages machine learning techniques to perform sentiment analysis on many Airbnb reviews. The system aims to provide timely and insightful information about the experiences and sentiments of immigrants in the UK, as reflected in their Airbnb reviews. By employing state-of-the-art machine learning algorithms, the system enables efficient and accurate sentiment classification, allowing for the identification of key themes and sentiments expressed by immigrant users. The study demonstrates the potential of this approach in gaining a deeper understanding of immigrant perspectives within the context of peer-to-peer accommodation, and its implications for social media monitoring and customer satisfaction management.This present study has conducted a critical analysis utilizing efficient feature extraction techniques, including N-grams and TF-IDF, to optimize identifying positive, neutral, and negative feedback. Furthermore, five different models were utilized, and the training and testing processes were accompanied by parameter tuning. Ultimately, the study concluded that the Random Forest (RF) classifier performed exceptionally well, achieving a 95% accuracy rate.

Autoregressive Feature Extraction with Topic Modeling for Aspect-based Sentiment Analysis of Arabic as a Low-resource Language

This paper proposes an approach for aspect-based sentiment analysis of Arabic social data, especially the considerable text corpus generated through communications on X (formerly known as Twitter) for expressing opinions in Arabic-language tweets during the COVID-19 pandemic. The proposed approach examines the performance of several pre-trained predictive and autoregressive language models; namely, Bidirectional Encoder Representations from Transformers (BERT) and XLNet, along with topic modeling algorithms; namely, Latent Dirichlet Allocation (LDA) and Non-negative Matrix Factorization (NMF), for aspect-based sentiment analysis of online Arabic text. In addition, Bidirectional Long Short Term Memory (Bi-LSTM) deep learning model is used to classify the extracted aspects from online reviews. Obtained experimental results indicate that the combined XLNet-NMF model outperforms other implemented state-of-the-art methods through improving the feature extraction of unstructured social media text with achieving values of 0.946 and 0.938, for average sentiment classification accuracy and F-measure, respectively.

Exploring the Subjectivity of English Academic Discourse in the Context of Big Data

Abstract This study develops a sentiment analysis model for English academic discourse based on word information to effectively understand and analyze the sentiment tendencies in English literary texts. The structure of the model includes word embedding layer, character-level feature extraction, word-level feature extraction and feature fusion and classification layer. The word embedding layer realizes the mapping between word vectors and word vectors by microblogging pre-trained word vectors. The character-level feature extraction session uses a multi-window convolutional layer to capture N-Gram information. In contrast, the word-level feature extraction obtains deeper semantic information through a Bi-LSTM layer and fuses it with character-level information to enhance robustness. The feature fusion and classification layer further combines these features and determines the fusion weights through a linear layer to achieve sentiment classification. In performance tests, the model achieves 92.5% sentiment classification accuracy on the standard dataset, an improvement of about 6% compared to traditional methods. In particular, the accuracy is improved by 5% when dealing with text with sentiment polarity transition, showing good adaptability. In addition, using 657 positive and 679 negative sentiment words as seed words effectively expands the sentiment lexicon and enhances the comprehensiveness and accuracy of sentiment analysis.

Unraveling Emotions: Contemporary Approaches in Sentiment Analysis

Sentiment analysis, an essential component of natural language processing, plays a pivotal role in deciphering public opinion and emotional cues within the vast sea of user-generated content on social media platforms. This paper presents a focused analysis on sentiment analysis leveraging two different techniques, namely VADER (Bag of Words approach) and the RoBERTa model, an extension of the BERT architecture, known for its outstanding performance in a wide range of Natural Language Processing (NLP) tasks. Model A achieved a 47% accuracy in sentiment classification, while Model B demonstrated a higher accuracy of 73%. The findings not only highlight the performance disparities between the two models but also offer insights into the factors contributing to their varying degrees of accuracy. This research underscores the significance of model selection in sentiment analysis tasks and contributes to a better understanding of their applicability in real-world scenarios

ASPECT ORIENTED SENTIMENT ANALYSIS OF USER TEXT MESSAGES

Aspect-oriented sentiment analysis plays a crucial role in understanding users' opinions and sentiments towards specific product or service features. This study investigates intelligent algorithms for aspect-oriented tone analysis of user text messages, focusing on smartphone reviews as a case study. The study includes data collection and preprocessing, studying the methods and performance of models for aspect extraction and tone analysis. The performance of these models was compared on test datasets containing customer reviews. For aspect extraction, we combine cross-lingual syntactic analysis with topic models to improve accuracy over the combination of Russian-language syntactic analysis and topic models. In particular, the BERT transformer-based model, BER Topic, shows high performance in aspect detection due to its ability to understand the context in sentences. In the sentiment analysis task, the RuBERT-tiny model based on the BERT transformer outperforms the others, showing higher accuracy in sentiment classification in smartphone reviews. This study provides valuable insights into aspect-oriented tonality analysis, emphasizing the importance of selecting appropriate methods and approaches. The results provide researchers and practitioners with recommendations for effective aspect-oriented tonality analysis.

User Sentiment Analysis Methods for Elderly Social Media Networks

For the analysis of user sentiment in social media networks for the elderly population, emotional sentences are first extracted to classify movie reviews. Afterwards, social network data of the elderly population based on user search behavior is analyzed. The movie reviews of elderly social media users are analyzed for rating prediction. The research results indicate that the accuracy of sentiment classification results is in descending order of Dirichlet, maximum entropy, and support vector machine. The highest classification accuracy of the three algorithms is 87.1%, 86.9%, and 86.5%, respectively. The classification accuracy of the first level classifiers of Dirichlet, maximum entropy, and support vector machine are 90.7%, 88.7%, and 87.4%, respectively. The classification accuracy of the second level classifier is 86.7%, 83.7%, and 80.4%, respectively. The predictive analysis results of the research method are superior to those generated by using Slope One. The method proposed in the study can promote emotional analysis of film review texts, improving the analysis accuracy.

Real-time Sentiment Analysis on Social Networks using Meta-model and Machine Learning Techniques

Sentiment analysis is a critical task in social media analysis, enabling the understanding of user attitudes and opinions towards various topics. This paper proposes a real- time sentiment analysis system for social networks that utilizes a meta-model and machine learning techniques to accurately classify user sentiment. The proposed system integrates textual and visual data from social media posts to improve sentiment classification accuracy. The methodology includes data collection and preprocessing, feature extraction and selection, and the proposed meta-model for sentiment analysis. The system utilizes several machine learning techniques, including SVM, CNN, and LSTM networks. We evaluated the proposed system on a large-scale dataset and compared its performance with several state- of-the-art methods. The evaluation metrics, including accuracy, precision, recall, and F1-score, showed that our proposed system outperforms existing methods. The proposed system’s ability to handle multimodal data and achieve high accuracy in real- time makes it suitable for various applications, including social media monitoring and marketing analysis. The proposed system’s limitations provide opportunities for further research, such as developing more efficient algorithms and models that require less training data, and improving techniques for handling noisy and ambiguous data, such as sarcasm and irony. In conclusion, the proposed real-time sentiment analysis system using a meta-model and machine learning techniques provides a robust and efficient solution for sentiment analysis on social networks. The proposed system's performance and potential applications demonstrate its importance in the field of social media analysis.

Exploring Real-Time Sentiment Analysis Prototype for Retail Industry

The rise of social media platforms has revolutionized the way consumers interact with retailers and express their opinions on products and services. Online retailers particularly need to keep a close eye on customer sentiment in real-time to make informed decisions about their offerings and improve customer satisfaction. However, efficiently analysing large volumes of unstructured text data from social media in real-time poses a significant challenge. This research aimed to develop a scalable, real-time sentiment analysis system tailored for online retailers using Reddit as the data source. The system comprises three main components: a data extraction and streaming pipeline, a sentiment analysis model, and a web application with real-time analytics. To address the data extraction challenge, a job queue-based system was implemented using Node.js, ‘BullMQ’, and Redis to create and manage campaigns for data streaming from Reddit. The data was streamed using Kafka, a distributed streaming platform, to enable efficient real-time processing. The sentiment analysis model was developed using a Naive Bayes classifier after experimenting with other machine learning and deep learning techniques. In the conducted study, the sentiment analysis model's performance was evaluated using standard metrics tailored to the context of online retail sentiment analysis. An accuracy of 0.6737 was achieved, reflecting the model's ability to correctly classify approximately 67.37 per cent of the sentiments in the test data. Concurrently, an F1 score of 0.7894 was recorded and the Area Under the Curve (AUC) value on the test data was measured at 0.5468, a metric that, while acceptable, suggests room for further refinement in the model's discriminatory ability between classes. The integration of the Data Version Control (DVC) system provided a mechanism for fine-tuning the model according to specific data requirements of various tenants. These results, taken together, not only validate the feasibility of employing a Naive Bayes classifier for real-time sentiment analysis in the retail context, but also provide a baseline for future research aimed at enhancing both the accuracy and efficiency of sentiment classification. The project’s evaluation focused on the performance of the sentiment analysis model, the efficiency of the Kafka streaming and real-time Spark pre-processing pipeline, and the backend infrastructure, including the job queuing system and WebSocket implementation. Various evaluation techniques, such as graphs and literature comparisons, were used to assess the system’s performance. In conclusion, this project successfully demonstrated the feasibility of a scalable, real-time sentiment analysis system for online retailers using Reddit data. The system has the potential to help retailers better understand customer opinions and make data-driven decisions for their businesses. Future work could include exploring alternative data sources, experimenting with more advanced sentiment analysis techniques, and enhancing the web application’s user interface and analytics capabilities.

Advancement in Bangla Sentiment Analysis: A Comparative Study of Transformer-Based and Transfer Learning Models for E-commerce Sentiment Classification

Background: As a direct result of the Internet's expansion, the quantity of information shared by Internet users across its numerous platforms has increased. Sentiment analysis functions at a higher level when there are more available perspectives and opinions. However, the lack of labeled data significantly complicates sentiment analysis utilizing Bangla natural language processing (NLP). In recent years, nevertheless, due to the development of more effective deep learning models, Bangla sentiment analysis has improved significantly. Objective: This article presents a curated dataset for Bangla e-commerce sentiment analysis obtained solely from the "Daraz" platform. We aim to conduct sentiment analysis in Bangla for binary and understudied multiclass classification tasks. Methods: Transfer learning (LSTM, GRU) and Transformers (Bangla-BERT) approaches are compared for their effectiveness on our dataset. To enhance the overall performance of the models, we fine-tuned them. Results: The accuracy of Bangla-BERT was highest for both binary and multiclass sentiment classification tasks, with 94.5% accuracy for binary classification and 88.78% accuracy for multiclass sentiment classification. Conclusion: Our proposed method performs noticeably better classifying multiclass sentiments in Bangla than previous deep learning techniques. Keywords: Bangla-BERT, Deep Learning, E-commerce, NLP, Sentiment Analysis

Sentiment Analysis of Public Opinion on the 2024 Presidential Election in Indonesia Using Twitter Data with the K-NN Method

Twitter is often used by the public as a platform to speak and express their opinions, especially in the context of the 2024 Presidential Election. Tweets related to the '2024 Presidential Election' can be used as a source of data for social media analysis to determine whether the expressed opinions tend to be positive or negative. The research process involves data collection of tweets, preprocessing, tokenization, class attribute determination, directory filling, sentiment analysis, and classification steps, including testing the value of k and testing the confusion matrix. The research and testing results show that the K-NN method successfully achieves a sentiment classification accuracy rate of 86.48%.

Hotel Comment Emotion Classification Based on the MF-DFA and Partial Differential Equation Classifier

Due to the significant value that hotel reviews hold for both consumers and businesses, the development of an accurate sentiment classification method is crucial. By effectively distinguishing the authenticity of reviews, consumers can make informed decisions, and businesses can gain insights into customer feedback to improve their services and enhance overall competitiveness. In this paper, we propose a partial differential equation model based on phase-field for sentiment analysis in the field of hotel comment texts. The comment texts are converted into word vectors using the Word2Vec tool, and then we utilize the multifractal detrended fluctuation analysis (MF-DFA) model to extract the generalized Hurst exponent of the word vector time series to achieve dimensionality reduction of the word vector data. The dimensionality reduced data are represented in a two-dimensional computational domain, and the modified Allen–Cahn (AC) function is used to evolve the phase values of the data to obtain a stable nonlinear boundary, thereby achieving automatic classification of hotel comment texts. The experimental results show that the proposed method can effectively classify positive and negative samples and achieve excellent results in classification indicators. We compared our proposed classifier with traditional machine learning models and the results indicate that our method possesses a better performance.

Review on positional significance of LSTM and CNN in the multilayer deep neural architecture for efficient sentiment classification

The deep learning revolution in the current decade has transformed the artificial intelligence industry. Eventually, deep learning techniques have become essential for many computational modeling tasks. Nevertheless, deep neural models provide a high degree of automation for natural language processing (NLP) applications. Deep neural models are extensively used to decode public reviews subjective to specific products, services, and other social activities. Further, to improve sentiment classification accuracy, several neural architectures have been developed. Convolutional neural networks (CNN) and Long-short term memory (LSTM) are the popular deep models employed in ensemble architectures for sentiment classification tasks. This review article extensively compares the competence of CNN and LSTM-based ensemble models to improve the sentiment accuracy for online review datasets. Further, this article also provides an empirical study on various ensemble models concerning the position of LSTM and CNN for efficient sentiment classification. This empirical study provides deep learning researchers with insights into building effective multilayer LSTM and CNN models for many sentiment analysis tasks.