Sentiment Analysis of Meme Images Using Deep Neural Network Based on Keypoint Representation

This study explores the role of sentiment analysis as a predictive tool for understanding and forecasting product launch success in the digital market. Sentiment analysis involves the classification of consumer sentiment expressed on social media platforms such as Twitter and Instagram, and it can significantly impact businesses by predicting consumer behavior and product performance. The research highlights the relationship between social media sentiment and product success, demonstrating that positive sentiment is strongly correlated with higher sales and consumer engagement, while negative sentiment can lead to declines. Machine learning models, including Support Vector Machines (SVM) and Random Forest, were employed to classify sentiment from large volumes of social media data and correlate it with product performance indicators such as sales volume and consumer interaction. The study found that sentiment analysis models were highly effective in predicting product success, with positive sentiment generally driving product profitability and negative sentiment posing a potential threat to brand reputation. Moreover, the analysis showed that social media sentiment provides real-time insights into consumer perceptions, enabling businesses to quickly adjust marketing strategies and product development plans. These findings underscore the importance of integrating sentiment analysis into product launch evaluations and strategic decision-making. Future research should explore the integration of sentiment analysis with other predictive market models and investigate the effects of fake reviews and post-purchase consumer behaviors on product success.

Sentiment analysis has emerged as a powerful tool in financial markets, offering the ability to harness market participants' collective wisdom and emotions to make more informed investment decisions. This research paper presents a comprehensive literature review on sentiment analysis models in predicting stock prices, focusing on the Bank Nifty Index, a critical benchmark in the Indian financial sector. The review begins by elucidating the significance of sentiment analysis in stock price prediction, highlighting its pivotal role in capturing market sentiment, investor emotions, and their impact on asset valuations. A detailed overview of sentiment sources, encompassing financial news, social media, reports, and other relevant data streams, provides insight into the rich tapestry of information that sentiment analysis can leverage. The paper delves into various sentiment analysis approaches, from traditional lexicon-based and rule-based methods to cutting-edge machine learning and deep learning techniques. It elucidates the intricate process of integrating sentiment analysis with stock price prediction models, exploring feature concatenation, time series integration, and weighted data to incorporate sentiment insights effectively. Evaluation metrics crucial for assessing the performance of both sentiment analysis models and stock price prediction models are thoroughly discussed. It concludes by highlighting the substantial potential of sentiment analysis in augmenting stock price prediction, offering investors and traders valuable tools to navigate the complexities of financial markets.

Sentiment analysis, a critical subfield of natural language processing, is widely employed to decipher the emotions andopinions expressed in textual data. With the growing prevalence of emojis in digital communication, understanding theircontribution alongside textual information has become paramount for comprehensive sentiment classification. This paperproposes a novel hybrid deep learning model for sentiment analysis that effectively integrates advanced feature fusion andattention mechanisms to address the challenges of analyzing multimodal data. By combining BERT-based textual embeddingsand pre-trained emoji embeddings, the model captures nuanced semantic and emotional information. A self-attentionmechanism further enhances the representation by identifying long-range dependencies and contextual relationships betweentext and emojis. The model was evaluated on the Sentiment140 dataset, achieving state-of-the-art performance with anaccuracy of 91.7%, an F1-score of 93.6%, and an AUC of 96.4%, outperforming existing models such as BERT-LSTM andRoBERTa-GRU. This superior performance demonstrates the effectiveness of multimodal fusion in sentiment classification,particularly for social media data where emojis play a significant role in emotional expression. The proposed architecture alsoshows strong generalizability, offering robust performance across diverse datasets. While computational complexity is a notedchallenge, future research could explore optimization techniques to improve efficiency without compromising accuracy. Thiswork highlights the potential of hybrid models to advance sentiment analysis by bridging the gap between textual and visual-emotional communication, setting a foundation for more comprehensive multimodal understanding in natural languageprocessing tasks.

Social media has been a real-world sensor to observe the pulse of society. Although it provides unique communication opportunities, it also brings along vital challenges. One of them is hate speech, which attacks a single individual or targeted groups. Previous researchers claim that among social networks, the Twitter platform is mostly used to spread hate speech. However, data on a larger scale makes it hard to capture and understand the nature of hate speech on Twitter within specific food brands. In this study, sentiment analysis techniques were used to filter hate speech on Twitter and on three popular food brands in Malaysia. This study was conducted in five phases, namely raw data collection, pre-processing, sentiment analysis, visualization, and performance evaluation. This corpus consists of 28,898 data samples based on user tweet searches. A Twitter API was created and SQLite was used to store all the sample data. VADER sentiment analysis is used to classify tweet sentiment into positive, negative, and neutral. In the visualization phase, these three food brands are visualized using a histogram to gain sentiment analysis insights. Then, three machine learning methods were implemented to predict the best model for sentiment analysis. The Decision Tree classifier outperforms the average accuracy in Support Vector Machine and Logistic Regression with 99.99% for the training data set. This study provides insights to assist humans in making decisions. With the growth of opinions expressed in multimedia on social media, such as spoken feedback on Twitter, sentiment analysis has the potential to become a more news aggregation and low-cost endeavour.

The purpose of this study is to learn how people who speak different languages interpret the same issues, and to compare the results obtained and show the difference between their perspectives. To learn this point of view, we must first turn to open source intelligence. In this execution, a sentiment analysis application was designed using the Python programming language and the Natural Language Processing algorithms in the texts, which were taken as a data set of comments in Azerbaijani, Turkish, Russian and English languages from social media. As the data set, the comments made on 4 subjects: the declaration of Hagia Sophia as a mosque, the objection events that started with the natural gas hike in Kazakhstan, the natural disasters in Turkey, the Ukraine crisis. After loading the texts in four languages from the network environment, after preprocessing, the text was divided into 8 different categories (neutral, fear, joy, anger, sadness, surprise, disgust, shame) by means of the application written in Python programming language based on Data Mining and Machine Learning topics. In the study, precision, sensitivity, accuracy and F1 score were obtained by using Random Decision Forests, K - Near Neighbor Algorithm, Decision Trees, Support Vector Machine, Naive Bayes Algorithm, Logistic Regression, which are machine learning methods. By comparing the results, it was determined that the Logistic Regression method obtained the highest result. A sentiment analysis model was created using the Logistic Regression method, and sentiment analysis was performed for each subject at separation and the results were compared.

Purpose–This comprehensive review aims to analyze sentiment analysis models, focusing on their effectiveness in managing polarity shifts (positive, neutral, and negative sentiments). The investigation delves into specific aspects, including feature selection/extraction techniques, data augmentation, input datasets, and evaluation metrics. The goal is to provide valuable insights for both theoretical understanding and practical applications, guiding advanced research. Method–Following Kitchenham’s framework, the review conducts an exhaustive literature search, emphasizing entities embedded within sentiment analysis models. This methodological approach explores intricate challenges associated with polarity shifts, revealing opportunities for future research. Results–Key findings highlight the efficacy of sentiment analysis models in handling polarity shifts. Recursive Feature Elimination emerges as the most effective wrapper-based feature selection technique, with Word2Vec standing out for word embeddings. Notably, negation is identified as a significant polarity shifter. The study also identifies common datasets, machine learning, and deep-learning models, emphasizing the effectiveness of hybrid models and ensembles. Conclusion–This systematic review offers a comprehensive analysis of sentiment analysis models, providing insights into their status and key trends. The findings contribute to advancing the field, offering valuable guidance for researchers, practitioners, and developers working on polarity shift challenges, particularly in addressing implicit negation. Recommendations–Future research should prioritize refining models addressing implicit negation, requiring empirical investigations to assess their effectiveness. Additionally, efforts should focus on developing standardized evaluation metrics to accurately capture the intricacies of polarity shifts, ensuring a comprehensive assessment of model performance. Implementing these recommendations will advance the state-of-the-art. Research Implications–The review suggests practical and theoretical implications by highlighting effective models and their embedded entities. Recognition of explicit negation underscores theneed for models capable of discerning and addressing negated sentiments. Practitioners are encouraged to adopt diverse strategies, and research opportunities lie in exploring implicit negation for ongoing innovation in sentiment analysis research. Keywords–sentiment analysis, positive polarity, negative polarity, polarity shift, sentiment classification

BackgroundLoneliness is a global public health issue affecting a considerable number of people as well as burdening the public health system and increasing the risk of other life-threatening and life-damaging conditions. In USA an estimated 17% adults aged 18–70 report loneliness. The monetary loss as result of loneliness is estimated to be between USD 8074.80 and USD 12,0777.70 per person per year in the United Kingdom. But the dynamics of loneliness are not understood. Social media platforms have become a valuable source of data to study this phenomenon. ObjectivesThis paper aims to visualize the frequency of loneliness-related themes and topics in Twitter data. By using natural language (NLP) processing, sentiment analysis, and topic modeling, we seek to understand prevalent sentiments and concerns. Through interactive tree maps and radar plots, we present an engaging view of loneliness dimensions, allowing users to explore and gain insights into this issue on social media. We focus on comparative analysis of USA and India through analyzing tweets from both countries on loneliness. These two countries are the biggest countries population-wise where access to Twitter is legally allowed. MethodsThis study consists of two parts. In the first part, we employ NLP techniques and machine learning algorithms to extract and analyze tweets containing keywords related to loneliness. Through sentiment analysis and topic modeling, we discern linguistic patterns and contextual information to categorize the recurring themes and topics. Advanced text analytics is used to gain nuanced insights into the experiences, emotions, and challenges connected with loneliness. In the second part, interactive visualizations are developed to present the findings in an engaging and intuitive manner. Techniques such as tree maps and radar plots are utilized to transform the analyzed data into visually appealing representations. ResultsThe analysis of Twitter data yields valuable knowledge about the prevalence and nature of themes and topics associated with loneliness. The interactive visualizations present a comprehensive view of the sentiments and concerns expressed by Twitter users. These interactive plots provide a holistic view of the distribution of themes and topics associated with loneliness, allowing experts to explore and interact with the data, gaining deeper insights into the complexities surrounding this issue. ConclusionThis paper successfully explores themes and topics related to loneliness on Twitter by employing NLP, sentiment analysis, and topic modeling. The interactive visualizations enhance the accessibility and usability of the findings, providing valuable insights for various stakeholders. The study contributes to a deeper comprehension of loneliness in the context of social media.

This study aims to construct an efficient consumer sentiment analysis model and apply it to optimize personalized marketing strategies for enhancing user conversion rates and satisfaction. The yf_amazon dataset and Chinese sentiment classification corpus are selected as foundational data by integrating natural language processing and deep learning technologies. Preprocessing procedures including tokenization, stop word removal, noise filtering, and sentiment annotation, are employed to clean and structure consumer reviews. A sentiment classification model based on Bidirectional Encoder Representations from Transformers (BERT) and Transformer architectures is developed and fine-tuned with parameters including a learning rate of 2e-5 and 4 training epochs. Evaluation metrics encompass accuracy, precision, recall, and F1 score. Experimental results demonstrate that the proposed model achieves 92.5% accuracy and a 0.90 F1 score in sentiment classification tasks. This model outperforms conventional Convolutional Neural Networks (83.6% accuracy, 0.83 F1 score) and Long Short-Term Memory models (86.2% accuracy, 0.85 F1 score). Further segmentation of user sentiment clusters based on analysis results enables the implementation of differentiated recommendation strategies and timelimited marketing campaigns. Ultimately, it increases user conversion rates from 25.0% to 32.7% (a 7.7 percentage point improvement) while significantly enhancing consumer satisfaction with personalized recommendations. These findings indicate that Transformer-based sentiment analysis models hold substantial practical value for advancing personalized marketing applications.

Understanding and comprehending humans' views, beliefs, attitudes, or opinions toward a particular entity is sentiment analysis (SA). Advancements in e‐commerce platforms has led to an abundance of the real‐time and free forms of opinions floating on social media platforms. This real‐world data are imprecise and vague hence fuzzy logic is required to deal with such subjective data. Since opinions can be fuzzy in nature and definitions of opinion words can be elucidated differently; fuzzy logic has witnessed itself as an effective method to capture the expression of opinions. The study presents an elaborate review of the around 170 published research works for SA using fuzzy logic. The primary emphasis is focused on text‐based SA, audio‐based SA, and fusion of text‐audio features‐based SA. This article discusses the various novel ways of classifying fuzzy logic‐based SA research articles, which have not been accomplished by any other review article till date. The article puts forward the importance of SA tasks and identifies how fuzzy logic adds to this importance. Finally, the article outlines a taxonomy for sentiment classification based on the technique‐supervised and unsupervised in the SA models and comprehensively reviews the SA approaches specific to their task. Prominently, this study highlights the suitability of fuzzy‐based SA approaches into five different classes vis‐a‐vis (a) Sentiment Cognition from Words using fuzzy logic, (b) Sentiment Cognition from Phrases using fuzzy logic, (c) Fuzzy‐rule based SA, (d) Neuro‐fuzzy network‐based SA, and (e) Fuzzy Emotion Recognition.This article is categorized under:Algorithmic Development > Text MiningFundamental Concepts of Data and Knowledge > Motivation and Emergence of Data Mining

Sentiment analysis also referred to as opinion mining, plays a significant role in automating the identification of negative, positive, or neutral sentiments expressed in textual data. The proliferation of social networks, review sites, and blogs has rendered these platforms valuable resources for mining opinions. Sentiment analysis finds applications in various domains and languages, including English and Arabic. However, Arabic presents unique challenges due to its complex morphology characterized by inflectional and derivation patterns. To effectively analyze sentiment in Arabic text, sentiment analysis techniques must account for this intricacy. This paper proposes a model designed using the transformer model and deep learning (DL) techniques. The word embedding is represented by Transformer-based Model for Arabic Language Understanding (ArabBert), and then passed to the AraBERT model. The output of AraBERT is subsequently fed into a Long Short-Term Memory (LSTM) model, followed by feedforward neural networks and an output layer. AraBERT is used to capture rich contextual information and LSTM to enhance sequence modeling and retain long-term dependencies within the text data. We compared the proposed model with machine learning (ML) algorithms and DL algorithms, as well as different vectorization techniques: term frequency-inverse document frequency (TF-IDF), ArabBert, Continuous Bag-of-Words (CBOW), and skipGrams using four Arabic benchmark datasets. Through extensive experimentation and evaluation of Arabic sentiment analysis datasets, we showcase the effectiveness of our approach. The results underscore significant improvements in sentiment analysis accuracy, highlighting the potential of leveraging transformer models for Arabic Sentiment Analysis. The outcomes of this research contribute to advancing Arabic sentiment analysis, enabling more accurate and reliable sentiment analysis in Arabic text. The findings reveal that the proposed framework exhibits exceptional performance in sentiment classification, achieving an impressive accuracy rate of over 97%.

In recent years, more and more scholars have begun to focus on sentiment analysis on social media. Current sentiment analysis collects all relevant data, including public thoughts, opinions, and feelings, from a variety of open sources. In addition, it automatically predicts different aspects of outcomes or trends based on information collected globally in real time. This research area explores how to extract sentiment information from different modalities (e.g., text, images, and audio). However, the currently existing techniques face several challenges. It is difficult to achieve effective interaction with completely heterogeneous data, and these techniques cannot adequately guarantee data security during data interaction, which is particularly important when dealing with sensitive information. Therefore, this article introduces existing methods for protecting data privacy. Based on this foundation, we propose a novel algorithm called collaborative dynamic hierarchical federated learning with hypernetwork aggregation (CDHFL-HA), which is suitable for sentimental analysis. CDHFL-HA ensures that the data remain local to each participant while leveraging the data similarity between participants on the server and processing interference data on the participant to enhance the accuracy of the current sentimental analysis. In addition, an essential aspect considered in the proposed algorithm is explainability. Understanding the decisions and predictions made by sentiment analysis models is crucial for gaining trust and acceptance in real-world applications. CDHFL-HA incorporates explainability features, providing insights into the decision-making process, thus enhancing the interpretability of sentiment analysis results. Numerous experimental results show that the algorithm outperforms existing algorithms in complex scenarios, with a minimum accuracy of 0.6007 and a maximum of 0.9962. In addition, it can be seen from the experimental results in this article, that the communication parameters in the experiments are similar to those of other federated learning, while the number of training rounds is improved by up to 50% (i.e., 20 rounds faster) relative to other algorithms.

More recently, as images, memes and graphics interchange formats have dominated social feeds, typographic/infographic visual content has emerged as an important social media component. This multimodal text combines text and image, defining a novel visual language that must be analysed because it has the potential to modify, confirm or grade the sentiment's polarity. The problem is how to effectively use information from the visual and textual content in image‐text posts. This article presents a new deep learning‐based multimodal sentiment analysis (MSA) model using multimodal data such as images, text and multimodal text (image with embedded text). The text analytic unit, the discretization control unit, the picture analytic component and the decision‐making component are all included in this system. The discretization unit separates the text from the picture using the variant and channel augmented maximally stable extremal regions (VCA‐MSERs) technique, which are then analysed as discrete elements and fed into the appropriate image and text analytics units. The text analytics system utilizes a stacked recurrent neural network with multilevel attention and feedback module (SRNN‐MAFM) to detect the sentiment of the text. A deep convolutional neural network (CNN) structure with parallel‐dilated convolution and self‐attention module (PDC‐SAM) is developed to forecast the emotional response to visual content. Finally, the decision component employs a Boolean framework including an OR function to evaluate and classify the output into three fine‐grained sentiment classes: positive, neutral and negative. The proposed work is simulated in the python platform using the STS‐Gold, Flickr 8k and B‐T4SA datasets for sentiment analysis of text and visual and multimodal text. Simulation outcomes proved that the suggested method achieved better accuracy of 97.8%, 97.7% and 90% for text, visual and MSA individually compared to other methods.

This research constructs a process flow of social media sentiment analysis and explains relationship between comments on social media and stock. One of the popular social media for Stock Exchange of Thailand investors is Pantip.com, a website providing service as a webboard with tagging feature. In this research, all posts tagged by ‘Stock’ on this website were crawled to files. Then comments were scraped and cleaned. For model training, some comments were labelled into three classes; positive, negative and neutral. Sentiment analysis model was constructed by Naive Bayes Classification technique. In evaluation, the model shown that it performed 74% accuracy. This model was utilised to classify comments into sentiments. When all comments were completely classed, sentiment types were counted by date. Finally, correlation matrices were constructed to find relationship between number of sentiments and stock. The research found that number of sentiments from social media relate to ADVANC and CPALL stock volumes. Moreover, the correlation always reaches to the peak on trading day then it gradually declines with the magnitude depending on the day length after trading day.

This thesis examines the issue of ill-formed text (microtext) and shows its impact on sentiment analysis. The source of microtext is Twitter, WhatsApp, Facebook and other social media platforms. Words or phrases not in their standard language format like “lol” (laugh out loud), “c u 2nite” (see you tonight) and “plz” (please) are called out of vocabulary (OOV) terms. On the other hand, words or phrases in their standard forms, for example, “tonight”, “love”, or “talk to you later”, are called In Vocabulary (IV) terms. The motivation behind microtext normalization is its growing presence on both online and offline platforms. Microtext might appear to be an ill-formed text, which requires find-and-replace methods. Traditionally, in a Natural Language Processing (NLP) context, microtexts such as abbreviations and length shortening or extending are normalized, but phonemic and graphemic variation is often ignored. However, the features of microtext can be traced back to social, cognitive and geographic influences. To understand social media text, it is essential to develop a microtext normalization module for traditional and modern NLP models. The first section of this thesis focuses on combining unsupervised learning methods in 2 different chapters, 4 and 5. The first method includes lexicon creation to transform the most frequent microtexts and emoticons on social media and applies them to sentiment analysis. This method shows an accuracy increase of 4% when microtext normalization is applied before sentiment analysis. This method is extended with an application to a pre-trained chatbot to improve the chatbot’s understanding of microtext (social media slangs). This method shows a mean BLEU score of 0.8 for the SMS and Tweet dataset. The second method incorporates a phonetic based approach to transform any microtext into its phonetic equivalent before 1normalizing it. This method shows an accuracy improvement of 6% for sentiment analysis on SenticNet. The second part of the unsupervised learning is in Chapter 5, introducing the IPA-based method for microtext normalization and sentiment analysis. Epitran was used to transform a word to its International Phonetic Alphabet (IPA) equivalent. The International Phonetic Alphabet (IPA) is an alphabetic system of phonetic notation based primarily on the Latin script. There is a significant improvement in microtext normalization task and sentiment analysis over Soundex. The results also show that there is very little redundancy when words are transformed to phonetics in IPA. In addition, the accuracy of polarity detection using this method has an improvement over baseline by 5%. Chapter 6 discusses a simple sequence-to-sequence based Deep Learning method. As the language is dynamic, it is not easy to maintain the lexicon. Deep Learning alleviates the static lexicon with a probabilistic model to transform microtext into its standard form. The findings suggest that a semiotic class of microtext was easy to transform, whereas the phonetic class of microtext was not handled easily. The proposed model improves the pre-trained sentiment analysis model by 4%. We also introduce a corpus for English microtext normalization. The corpus contains microtext-containing sentences along with their correctly spelt sentences and their correct polarity. This dataset can tackle two NLP problems, i.e., microtext normalization and sentiment analysis. This corpus consists of 38% positive, 39.9% negative and 22.1% neutral. A pre-trained model shows an accuracy of 46.4 % on OOV text for sentiment analysis, showing considerable room for improvement. In summary, this thesis introduces several approaches to microtext normalization analysis and focuses on sentiment analysis application. Microtext cannot be thought of as a one-dimensional problem to NLP. It encapsulates many factors like regional (geographical), ethnic (national and racial), and social (class, age, gender, socioeconomic status and education), which cannot be ignored given the growing importance of social media in our daily life.

Sentiment analysis mainly studies the emotional tendencies of texts from grammar, semantic rules and other aspects. The texts from social network are characterized by less words, irregular grammar, data noise and so on, which have increased the difficulty of emotion analysis. In order to improve the performance of machine learning in sentiment analysis, this study proposed the Majority Decision Algorithm to classify the emotional tendentious of the text in WeChat, combined the characteristics of five classifiers and integrated the classification results of five classifiers, eventually the text can be classified in WeChat. Firstly, this study utilized the BlueStacks to crawl the cache of WeChat Moment developed by Tencent company. Secondly, the cache was processed by Python to get the WeChat dataset. After the Chinese word segmentation, data cleaning and segmentation, the sentiment classification experiment were carried out using different classifiers. Finally, a Majority Decision Algorithm composed of five classifiers was established. It included, Naive Bayes (sklearn), Naive Bayes (SnowNLP), SVM (linear), SVM (RBF) and SGD. Then, the comparison was carried out between the performance of the algorithm and the five classifiers. Results show that the precision rates of the five classifiers are 0.8598, 0.8154, 0.8511, 0.8739 and 0.8678; the recall rates are 0.8544, 0.8482, 0.9380, 0.9226 and 0.9349; F1 scores are 0.8571, 0.8315, 0.8924, 0.8975 and 0.9001, respectively. The algorithm of the Precision rate, Recall rate and F1 score were 0.8804, 0.9349 and 0.9069, respectively, indicating that algorithm in current study significantly improved the performance, which can be effectively applied into the new text form of WeChat Moment. The study can provide theoretical reference for sentiment classification of Chinese text based on machine learning.