Text Sentiment Research Articles

Aspect-Level Sentiment Analysis of Customer Reviews Based on Neural Multi-task Learning

In the era of big data, major e-commerce platforms are facing the challenge of an exponential growth in the number of user comments. Effectively utilizing these comments has become an urgent issue. Traditional manual statistical methods are no longer able to meet the demands for accuracy and real-time analysis. The rise of artificial intelligence-based text mining techniques provides a new approach to address this problem. By building deep learning analysis models, it is possible to uncover user preferences and product characteristics, helping businesses proactively adjust sales strategies, improve product and service quality, and achieve precise marketing. Text sentiment analysis is one of the important research directions in the field of text mining. Its basic idea is to transform subjective texts with emotional color into structured data, and then use machine learning, deep learning, and other artificial intelligence technologies to extract emotional features and discover knowledge patterns. This article proposes a deep learning-based sentiment analysis model, which includes a shared sentiment prediction layer used to transfer emotional knowledge between different aspect categories and alleviate the problem of insufficient data. The model consists of two parts: the Aspect Category Detection (ACD) classifier based on attention mechanism and the Aspect Category Sentiment Analysis (ACSA) classifier. The ACD part generates word weights using attention mechanism, while the ACSA part predicts the sentiment of words and combines weights to determine the sentiment of aspect categories within sentences.

A novel two-stage wrapper feature selection approach based on greedy search for text sentiment classification

Sentiment analysis is a crucial step in obtaining subjective data from online text sources. Nevertheless, the substantial challenge of high dimensionality prevails within text classification. Addressing this, dimension reduction emerges as a valuable approach to enhance the efficacy of classification in the domain of machine learning. The discerning removal of redundant features not only expedites training processes but also bolsters the achievement of accurate classifications. It is worth noting that the effectiveness of distinct feature selection methodologies can be contingent upon the unique attributes inherent in diverse datasets. Within the purview of this investigation, a novel two-stage approach is introduced, characterized by a greedy search-based wrapper feature selection algorithm. The underpinning of this algorithm involves leveraging the outcomes yielded by filter-based feature selection techniques to establish a prioritized sequence for the scrutiny of features within the proposed framework. This strategic sequencing harnesses the cumulative insights from a series of filter-based methodologies, thereby facilitating the curation of feature subsets that underscore pivotal attributes. Nonetheless, it is acknowledged that the greedy selection approach primarily favors features with high-ranking scores, and thus, it may not adequately evaluate the potential of feature combinations that involve low-scoring elements. An extensive experimental inquiry was conducted across widely recognized sentiment analysis datasets to assess the performance of the introduced methodology. The computational findings eloquently demonstrate that the proposed algorithm attains an average accuracy of 96.88% for nine public sentiment datasets and 94.43% for the Humir datasets when coupled with the multinomial Naive Bayes classifier. Furthermore, the experimental outcomes conspicuously establish the superiority of the proposed technique in state-of-the-art studies across the same set of nine sentiment datasets and the Humir datasets.

Enhancing Stock Market Prediction: A Hybrid RNN-LSTM Framework with Sentiment Analysis

Objectives: Predicting stock prices with accuracy is a difficult but crucial endeavor for market participants. To increase the precision of stock market forecasts, this study suggests a novel method that blends sophisticated neural network algorithms with sentiment assessment. Methods: News data is pre-processed. Each text document received a sentiment score reflecting overall sentiment. These scores were integrated into the feature set, combining textual sentiment information with historical stock price data from BSE Sensex. The proposed model of hybrid RNN-LSTM is applied and compared with the Random Forest Regressor (RFR), and Support Vector Regressor (SVR). The LSTM model is also applied and tested on data without sentiment analysis scores. Findings: The proposed model yields promising results in stock market prediction accuracy. It significantly gives a low value for mean absolute error (0.036), mean squared error (0.021), and root mean square error (0.046) when compared with the SVR and RFR models. The R2 value is also compared with literature methods, and it shows a 0.40% to 5.5% enhancement in the scores. The results prove that the incorporation of sentiment analysis enriches the predictive capabilities of the model. Novelty: Sentiment analysis combined with the hybrid RNN-LSTM framework provides a new technique to increase the accuracy of stock market forecasts. Using sophisticated knowledge of market dynamics and sentiments, the proposed approach gives important results to market participants, investors, and analysts of financial markets. Keywords: Stock Market, Sentiment Analysis, Evaluation Metric, Prediction

Sentiment Analysis on Placement Aspect

Sentiment analysis has emerged as a pivotal tool within the realm of natural language processing, finding widespread application across diverse domains such as marketing, social media analysis, and customer feedback assessment. This research delves into the specialized domain of sentiment analysis focused on the placement aspect, which revolves around the intricate process of aligning individuals with suitable job opportunities. The primary objective of this study is to dissect sentiments embedded within textual data associated with placements, including resumes, job listings, and interview appraisals, with the aim of extracting nuanced insights to inform strategic decision-making within recruitment and talent acquisition processes. Leveraging cutting-edge techniques in machine learning and natural language processing, our research employs a multifaceted approach encompassing text preprocessing, feature extraction, and sentiment classification methodologies. A spectrum of sentiment analysis algorithms is explored, ranging from lexicon-based methods to sophisticated machine learning models (such as Support Vector Machines and Naive Bayes) and deep learning architectures (including Recurrent Neural Networks and Transformers). Additionally, our investigation extends to assessing the influence of various factors—such as dataset dimensions, domain specificity, and feature selection techniques—on the accuracy and robustness of sentiment analysis outcomes.

Evaluating the Impact of Internet Users' Emotional Tendencies on Opinion Dynamics through Computer-Assisted Sentiment Analysis of Texts

- Internet communication is characterised by "non-linear flow" and "unorganised aggregation," and computer-assisted text sentiment analysis is an important means of evaluating the impact of netizens' emotions on the dynamics of public opinion. The uncertainty, anonymity, disorder, and blindness inherent in fast-gathering and fast-dispersing Internet communication drive the continuous generation of texts, which has an important impact on public opinion. In order to study this issue in depth, this paper chooses two text types of "popular Weibo" and "popular topics" in Sina Weibo as the object of study and analyses the issue in detail from the two dimensions of agenda construction and agenda setting. In the process of the study, the required text data were collected through Python, and the theoretical assumptions were verified with text analysis methods to explore the influence of text sentiment bias on opinion diffusion, with a view to providing new insights and references for the application of the theory of opinion diffusion and the field of computational diffusion.

Constructing a Fraud Risk Early Warning Model for Chinese Listed Companies Driven by Heterogeneous Data

In this paper, we propose a model for early warning of fraud risk among Chinese listed companies, leveraging the capabilities of heterogeneous data analysis. Given the complexity and multidimensionality of corporate fraud, our model integrates multiple data sources including financial reports, market behavior, and the sentiment of annual report texts, representing diverse sets of heterogeneous data. This approach employs advanced data processing techniques to handle and amalgamate heterogeneous data, ensuring robustness and accuracy. Utilizing machine learning algorithms, the model not only detects potential fraud signals but also quantifies the level of risk, providing stakeholders with a dynamic predictive tool. This research offers a comprehensive data-driven approach to fraud detection in the corporate sector, underscoring the importance of meticulous risk assessment using various data streams. It marks a critical step in proactive fraud management in an increasingly complex financial environment.

Text Emotion Classification System Integrating Visual Communication and Deep Learning for Social Platform

With the development of modern information technology, social networks have become an im-portant platform for people to express, and at the same time, a large number of texts have been produced. However, the comment text has the characteristics of randomness and colloquialism in the way of expression, and also contains a lot of non-text data. Therefore, manually analyzing the emotional information in the text will consume a lot of time and the accuracy will be limited. To solve emotion classification, this study proposes a knowledge enhanced double loop emotion classification neural network model with attention mechanism. The study first preprocesses text data using a full sentence vector word vector model, then uses convolutional neural networks to recognize emotions in emoticons and emoticons in the text. Finally, the classification results are integrated using algorithms such as a dual loop sentiment classification neural network with pool-ing layers and attention mechanisms, K-nearest neighbors, and decision trees. The final compre-hensive expression recognition and text recognition results are used to obtain the text sentiment classification results. The experimental data shows that the model proposed in this study has an accuracy of 0.947 in the training set test, which is significantly better than other models. In da-tasets A and B, the accuracy of the research design model was 0.958 and 0.924, and the recall was 0.964 and 0.986, respectively. Compared to the baseline method or existing research models, the values of each indicator were significantly higher. The recall rate is the proportion of instances correctly identified as positive by the model to all actual positive instances, which can reflect the emotional classification performance of the research and design model. The higher the value, the better the performance of the model. In practical applications, the positive review rate of this model is above 0.9, which has obvious advantages compared to other models. This study utilizes deep learning techniques to classify sentiment in comment texts, providing reference for the field of text sentiment classification. In the e-commerce industry, it is possible to identify the emotions in user comments on products, further understand the product situation on the platform, and make targeted planning for product reserves, specifications, and so on.

Enhanced word vector space with ensemble deep learning model for COVID-19 Hindi text sentiment analysis

Enhanced word vector space with ensemble deep learning model for COVID-19 Hindi text sentiment analysis

Research on Sentiment Analysis of Micro-blog based on Attention-BiLSTM

The rapid development of the Internet gives birth to massive data. These data have the characteristics of multiple types, low density, and fast transmission speed. As one of the products of the era of big data, micro-blog provides users with a highly interactive platform. In order to explore the value and sentiment contained in micro-blog comment texts, and solve the problem of semantic loss and over-reliance on manual work in traditional text sentiment analysis, this paper proposes a deep learning model based on Attention-BiLSTM. In the embedding layer, the CBOW method of Word2vec is used to transform the text data into word vectors, and the attention mechanism is integrated to dynamically weight the input word vectors. Secondly, BiLSTM is used to extract text features, then fuse them in concat layer. After that the classification is realized by Softmax whose results of the output text are positive or negative. Finally, by comparing with TextCNN, LSTM and BiLSTM, the result shows that the Attention-BiLSTM model does have better classification effect, strong generalization and practicability.

Augmenting Sentiment Analysis Prediction in Binary Text Classification through Advanced Natural Language Processing Models and Classifiers

Sentiment analysis is a critical component in natural language processing applications, particularly for text classification. By employing state-of-the-art techniques such as ensemble methods, transfer learning and deep learning architectures, our methodology significantly enhances the robustness and precision of sentiment predictions. We systematically investigate the impact of various NLP models, including recurrent neural networks and transformer-based architectures, on sentiment classification tasks. Furthermore, we introduce a novel ensemble method that combines the strengths of multiple classifiers to improve the predictive ability of the system. The results demonstrate the potential of integrating state-of-the-art Natural Language Processing (NLP) models with ensemble classifiers to advance sentiment analysis. This lays the foundation for a more advanced comprehension of textual sentiments in diverse applications.

Unveiling the tapestry of mother–child interactions through text mining and sentiment analysis

The interaction between a mother and child stands as one of the most profound and intricate human connections, weaving a rich tapestry of behavioral and emotional bonds during the formative years. Although mother–child interactions have received substantial attention in the developmental science literature, few studies have tapped into the extensive corpus of speech data available to uncover the nuances of these interactions across developmental stages. This study applied text mining and sentiment analysis on narratives extracted from mother–child conversations to identify the developmental trend of mother–child interactions from early to middle childhood. The results, based on three key areas of dyadic interactions, demonstrated a shift toward more balanced turn-taking dynamics and linguistic congruence as children age. Also, there was a significant interdependence of mother and child expressed emotions across time. Further investigation of the dyadic emotionality revealed a nonlinear effect of mother-expressed emotion on child-expressed emotion: mother-expressed negative emotions followed a cubic-like pattern, while positive emotions followed a mild quadratic trend. Taken together, the findings of this study present a picture of progressive augmentation of mother–child synchrony over time.

A Comprehensive Multimodal Approach to Assessing Sentimental Intensity and Subjectivity using unified MSE model

In the dynamic realm of multimodal learning, where Representation Learning serves as a pivotal key, our research introduces a groundbreaking approach to understanding sentiment and subjectivity in audio and text. illustration from self-supervised learning, we've innovatively combined multi-modal and Unified--modal tasks, emphasizing the crucial aspects of consistency and distinctiveness. Our training techniques, likened to the art of fine-tuning an instrument, harmonize the learning process, prioritizing samples with distinctive supervisions. Addressing the pressing need for robust datasets and methodologies in combinational text and audio sentiment analysis, we present the Multimodal Opinion-level Sentiment Intensity dataset (MOSI). This meticulously annotated corpus offers insights into subjectivity, sentiment intensity, text features, and audio nuances, setting a benchmark for future research. Our method not only excels in generating Unified-modal supervisions but also stands resilient against benchmarks like MOSI and MOSEI, even rivaling human-curated annotations on the challenging datasets. This pioneering work paves the way for deeper explorations and applications in the burgeoning field of sentiment analysis.

An enhancement methodology for predicting transaction amounts for freelancers on freelance platforms: Based on sentiment analysis

Freelancing is a type of labor arrangement in which independent freelancers use their discretionary time to perform ad hoc tasks, usually of a single nature, with the aim of getting paid. This study proposes an augmented method for predicting the amount of freelance transactions using textual sentiment analysis. First, a unique feature labeled Freelancer Sentiment was selected to summarize the positive or negative sentiment orientation of freelancers. Subsequently, Naive Bayes algorithm is applied to process the text data from the freelancers platform to finally develop a Model for Computing Word Sentiment Values. The model helps to accurately calculate the sentiment values associated with emotional words. Finally, the Word Frequency-Inverse Document Frequency (TF-IDF) algorithm is used to construct the Text Sentiment Value Calculation Model, so as to accurately calculate the sentiment values of freelancers. The results of the comparison experiments of the five commonly used prediction models show that the mean squared error (MSE) of the model that includes the freelancer sentiment feature is significantly reduced by 6%-11% compared with the model that does not include the freelancer sentiment feature. This study contains theoretical explorations and practical implications. First, the proposed approach of extracting features from textual data to build predictive models provides a valuable reference for future enhancement of predictive modeling on freelance platforms, especially those that rely on unstructured data. Second, incorporating textual sentiment value features relevant to freelancers can significantly improve the accuracy of predicting transaction amounts. Third, the calculation of word and text sentiment values employs a series of algorithms that target specific features of the freelancer platforms text data. This approach is important for improving the accuracy of feature value calculation.

Research on the Evolution Trend of Group Psychological Security Risks Under Public Health Emergencies: Mining and Analysis Based on Social Media Data.

In the digital age, people's attitudes and psychological security towards public health emergencies will be shared. Similar or identical psychological security states are prone to clustering and differentiation, while differentiated group psychological security is more prone to polarization, leading to group psychological security risks and then posing a threat to social stability and national security. However, existing studies mostly use qualitative analysis methods to study group emotional risks. There are still limitations in the study of dynamics of group psychological security risks through mining real data of social media. The study aims to use intelligent analysis methods to understand how group psychological security risks dynamically change. The study draws on text sentiment analysis, Markov chains and time series analysis to construct a framework for the evolution of group psychological security risks. Based on this framework, text data was crawled on Sina Weibo platform, mainly consisting of posts during public health emergencies (March 1st to June 30th, 2022) in Shanghai, and a psychological security lexicon in the field of public health emergencies was constructed. This laid the foundation for identifying the tendencies, intensity, and transitions of individual text psychological security, and then exploring the evolution trend of group psychological security risks. Compared with the generation and reduction periods, group psychological security risks are more likely to occur during the outbreak and recovery periods, and the intensity level is also higher. The overall intensity of group psychological security risks shows an evolution trend of first increasing, then decreasing, and then increasing again. The paper provides an opportunity to explore the dynamics of psychological security in the digital space. Meanwhile, we call on the government and relevant management departments to pay more attention to the group psychological security risks formed during the outbreak and recovery periods of public health emergencies, and take corresponding measures in a timely manner to guide the public to transform the extreme psychological security state into the normal psychological security state, in order to prevent and resolve group psychological security risks, promote social stability and national security.

Integrating TextBlob and VADER for Dynamic Sentiment Analysis: A GUI-Based Approach with Emotion Visualization and Confidence Assessment

Abstract: This research presents a Graphical User Interface (GUI) application developed in Python using the Tkinter library for sentiment analysis. The GUI integrates two distinct sentiment analysis libraries, TextBlob and VADER, to provide a comprehensive analysis of text sentiment. Each library's sentiment analysis results are displayed along with confidence scores, allowing users to gauge the reliability of the analysis. Additionally, the GUI features a visual indicator in the form of a lamp that dynamically changes color based on the detected emotion (positive, negative, or neutral) inferred from the sentiment analysis results. Moreover, the GUI includes a combined sentiment score, calculated by averaging the sentiment scores obtained from both TextBlob and VADER, enhancing the overall accuracy and reliability of the sentiment analysis. This GUI provides users with an intuitive and interactive tool for analyzing text sentiment, with the ability to visualize emotions and assess sentiment confidence.

Learning deep news sentiment representations for macro-finance

This paper introduces custom neural network techniques to the problem of latent economic factor extraction for voluminous news analytics data. In the context of macro-financial news, we derive low-dimensional representations of time series that arise in textual sentiment analyses spanning various topics. We explore three applications for compressed news sentiment data: nowcasting GDP growth, explaining asset class returns in a panel data analysis, and time series momentum investment. Our empirical study shows that nonlinear data representations based on supervised autoencoder architectures compare favorably to alternatives across all applications. In specific, we demonstrate that augmenting autoencoders with supervision tasks based on common asset class returns and market characteristics disciplines the dimension reduction and naturally supports the transparency of resulting representations. Taken together, our findings position supervised autoencoders as attractive competitor models alongside PCA and PLS approaches.

Can real-time investor sentiment help predict the high-frequency stock returns? Evidence from a mixed-frequency-rolling decomposition forecasting method

This research examines the predictive effect of real-time investor sentiment on high-frequency stock returns. Utilizing text sentiment analysis, we extract investor sentiment with a half-hour frequency from the stock message board. The RR-MIDAS method is used to model half-hourly sentiment and three-minute stock returns, and economic analysis reveals that investor sentiment significantly affects the stock returns during seven high-frequency periods, and the influence gradually weakens. Subsequently, we propose the “MF-EEMD-ML” prediction system, which introduces a rolling decomposition algorithm into the RR-MIDAS framework for predicting high-frequency trend items combined with real-time forum sentiment. The results, using rolling EMD decomposition for comparison, show that the “MF-EEMD-ML” system achieves a maximum reduction of 19.18 % in MAE, 19.08 % in RMSE, 11.71 % in SMAPE, and a maximum improvement of 16.66 % in DS. Additionally, the outcomes of the Diebold-Mariano (DM) tests also demonstrate that the “MF-EEMD-ML” prediction system significantly outperforms both the “MF-EMD-ML” system and the LR model.

Online English writing teaching method that enhances teacher–student interaction

Abstract A significant component of the online learning platform is the online exercise assessment system, which has access to a wealth of past student exercise data that may be used for data mining research. However, the data from the present online exercise system is not efficiently used, making each exercise less relevant for students and decreasing their interest and interaction with the teacher as she explains the activities. In light of this, this research creates an exercise knowledge map based on the connections between workouts, knowledge points, and previous tournaments. The neural matrix was then improved using cross-feature sharing and feature augmentation units to deconstruct the workout recommendation model. The study also developed an interactive text sentiment analysis model based on the expansion of the self-associative word association network to assess how students interacted after the introduction of the personalized exercise advice teaching approach. The outcomes demonstrated that the suggested model’s mean diversity value at completion was 0.93, an increase of 0.14 and 0.23 over collaborative filtering algorithm and DeepFM (deep factor decompose modle), respectively, and that the proposed model’s final convergence value was 92.3%, an improvement of 2.3 and 4.1% over the latter two models. The extended model used in the study outperformed the support vector machine (SVM) and Random Forest models in terms of accuracy by 5.9 and 1.7%, respectively. In terms of F1 value indicator, the model proposed by the research has a value of 90.4%, which is 2.5 and 2.1% higher than the SVM model and Random Forest model; in terms of recall rate indicators, the model proposed by the research institute has a value of 94.3%, which is an increase of 6.2 and 9.8% compared to the latter two models. This suggests that the study’s methodology has some application potential and is advantageous in terms of customized recommendation and interactive sentiment recognition.

Data-Driven Insights from 10,000 Reviews: Fostering Sustainability through Rapid Adaptation to Guest Feedback

The present research aims to explore how customer satisfaction and discontent may influence the financial success of luxury hotels by analyzing more than 10,000 reviews from the Radisson Blu Hotel in Bucharest and financial data spanning a decade. Text mining and sentiment analysis were used in order to pinpoint crucial elements that could impact visitor experience and the way in which they might correlate with hotel revenues in the long run. Research has linked positive visitor evaluations to revenue growth, while negative feedback does not consistently lead to financial declines, suggesting that premium brands may be resilient to mild online criticism. Our research examines how customer feedback sentiment and business income are interconnected throughout time, emphasizing the significance of handling guest contentment in order to reduce resource spending and create a sustainable feedback loop between guests and businesses. Our analysis challenges the idea that negative evaluations always have a detrimental impact on financial performance and highlights the importance of long-term positive feedback in the hospitality sector. This study enhances comprehension of the influence of customer feedback on the luxury hotel business, providing valuable insights for service industry experts on utilizing reviews in the future in order to gain strategic benefits and also develop their businesses in a sustainable way.

Sentiment analysis of Twitter user text based on the BERT model

Deep Neural Networks (DNNs) utilizing Recurrent Neural Network (RNN) architectures have found extensive application in text sentiment analysis. A prevailing notion suggests that augmenting the model's capacity can significantly improve accuracy and overall model performance. Building upon this premise, this paper advocates the adoption of a larger BERT model for text sentiment analysis. Bidirectional Encoder Representations from Transformers (BERT) is a sophisticated pre-trained language comprehension model that leverages Transformers as feature extractors. However, as the amount of model data increases, exceeding the memory limitations of a single GPU, algorithm optimization becomes crucial. Therefore, this paper employs two methods, namely data parallelism and GPipe parallelism, to accelerate and optimize the BERT model. Compared to a single GPU, training speed almost linearly increases with the addition of more GPUs. In addition, this research investigates the accuracy of the most advanced language model, chatgpt, by reannotating the dataset. During training, it was observed that the accuracy of the chatgpt-annotated dataset significantly declined in both RNN and BERT models. This indicates that chatgpt still exhibits some errors in sentiment text analysis.