Sentiment and Topic Modelling Analysis of Museum Reviews in the Context of Traveller Types: The Case of Kazakhstan

To analyze public perceptions of over-the-counter (OTC) hearing aids through sentiment and thematic analysis of online consumer reviews and their changes over time. Sentiment and thematic analysis. Online reviews from third-party and product websites. All English online consumer reviews posted between 2016 and 2024 for OTC hearing aids (83 models) were recorded (n = 21,727). Sentiment analysis was performed using Valence Aware Dictionary and Sentiment Reasoner (VADER), a rule-based sentiment analysis tool incorporating natural language processing. VADER provides scores for each review ranging from -1 (most negative), 0 (neutral), to 1 (most positive). Additional thematic analysis was performed for the top 100 most positive, neutral, and negative reviews (n = 300). Overall, mean (SD) VADER sentiment score of online reviews was generally positive at 0.587 (0.411). Multivariable regression analysis showed that higher VADER scores were associated with higher-priced and behind-the-ear (BTE) type hearing aids. Although there was a significant increase in a number of reviews after the Food and Drug Administration's new establishment of the OTC hearing aid category in 2022, the mean sentiment scores slightly decreased (β =-.10, [95%CI: -0.12 to -0.09]). Thematic analysis revealed that positive sentiments highlighted the affordability and time-saving benefits of OTC hearing aids as alternatives to prescription models. Negative sentiments centered on sound quality, challenges with customer service, and inadequate amplification for those with severe hearing loss. Customers generally viewed OTC hearing aids positively, while mixed experiences were present. When used as indicated for adults with mild to moderate hearing loss, OTC hearing aids may offer a viable alternative to prescription devices, improving accessibility and affordability.

Sentiment Analysis is a method to identify, extract, and quantify people's feelings, opinions, or attitudes. The wealth of online data motivates organizations to keep tabs on customers' opinions and feelings by turning to sentiment analysis tasks. Along with the sentiment analysis, the emotion analysis of written reviews is also essential to improve customer satisfaction with restaurant service. Due to the availability of massive online data, various computerized methods are proposed in the literature to decipher text sentiments. The majority of current methods rely on machine learning, which necessitates the pre-training of large datasets and incurs substantial space and time complexity. To address this issue, we propose a novel unsupervised sentiment classification model. This study presents an unsupervised mathematical optimization framework to perform sentiment and emotion analysis of reviews. The proposed model performs two tasks. First, it identifies a review's positive and negative sentiment polarities, and second, it determines customer satisfaction as either satisfactory or unsatisfactory based on a review. The framework consists of two stages. In the first stage, each review's context, rating, and emotion scores are combined to generate performance scores. In the second stage, we apply a non-cooperative game on performance scores and achieve Nash Equilibrium. The output from this step is the deduced sentiment of the review and the customer's satisfaction feedback. The experiments were performed on two restaurant review datasets and achieved state-of-the-art results. We validated and established the significance of the results through statistical analysis. The proposed model is domain and language-independent. The proposed model ensures rational and consistent results.

Bayesian game model based unsupervised sentiment analysis of product reviews

It is crucial to understand customers' sentiments and opinions about restaurants for both restaurant owners and academics in order to increase customer satisfaction and profitability. The aim of this study was to investigate customer sentiments of restaurants in Cappadocia, Türkiye. 38380 customer reviews of 386 restaurants in Cappadocia were obtained from Tripadvisor. Two different methods were used: rule-based sentiment analysis (RBSA) and machine learning (ML). The topics extracted from the reviews by RBSA were food, place, service, price, view, and staff and the percentages of these topics in the reviews were 41.45%, 23.94%, 11.36%, 9.23%, 8.18%, and 5.84%, respectively. For each topic, sentiment analysis was performed with ML to determine the proportion of positive, negative, and neutral sentiments. The highest positive sentiment content was found in food (40.15%), followed by staff (35.07%) and view (33.78%). Price (4.11%) and service (3.90%) were found to have the highest negative sentiment rates. The percentage of positive sentiment in reviews in Western languages was usually higher than in Far Eastern languages. Combining RBSA and ML techniques can enable both grammatical rules and artificial intelligence techniques while producing appropriate results. By understanding these sentiment patterns, restaurant owners can identify areas for improvement, while researchers can gain valuable insights into consumer behavior and sentiment analysis techniques.

This research aims to conduct sentiment analysis of e-grocery application reviews using the Support Vector Machine (SVM) algorithm. Sentiment analysis is used to distinguish between positive and negative reviews by users who have provided reviews so that an evaluation of the services offered can be made. This research uses scraping techniques to obtain all the needed review data, focusing only on reviews of the Segari and Sayurbox applications. Datasets were collected from reviews using a library in Python, namely, google-play-scraper, obtained by the sayurbox application 4235 reviews and the segari application 5575. The dataset collected does not yet have a label, and the labeling process is impossible to perform manually by looking at the reviews one by one because it takes a long time and requires an expert in the field of language who can interpret the reviews and group them into positive and negative sentiments. Therefore, the sentiment-labeling process applies a lexicon-based method that works based on the inset lexicon dictionary by calculating each review's polarity value. The analysis process of this research uses the SVM algorithm because the SVM method has been proven to provide consistent and accurate results in various classification tasks, including sentiment analysis. The results show that the lexicon-based method and SVM produce good accuracy in determining the sentiment of e-grocery reviews, with a vegetable box application accuracy rate of 94%. In comparison, the segari application accuracy rate reached 97%.

Lazada app reviews on the Google Play Store become useful information if processed properly. Existing or new users can analyze app reviews to get information that can be used to evaluate the service. The activity of analyzing app reviews is not enough just to look at the number of stars, it is necessary to look at the entire content of the review comments to be able to know the purpose of the review. A sentiment analysis system is a system used to automatically analyze reviews to obtain information including sentiment information that is part of online reviews. This time the data will be classified using the Naive Bayes method. A total of 1,000 user reviews of the Lazada app were collected to form a dataset. The purpose of this study was to conduct sentiment analysis of Lazada app reviews on Google Play Store using Naive Bayes algorithm. This stage of research involves data collection, labeling, pre-processing, sentiment classification, and evaluation. In the pre-processing stage, there are 6 stages, namely Cleaning, Case Factoring, Word Normalization, Tokenization, Hyphen Removal, and Base Word Formation. The TF-IDF (Term Frequency - Inverse Document Frequency) method is used for word weighing. The data will be grouped into two categories, namely negative and positive. Next, the data will be evaluated using accuracy parameters. The test results showed an accuracy value of 84%, then for the grouping of negative and positive reviews, it was found that Lazada application reviews tended to be negative.

In the era of digitization, where everything is available online, and further, information about everything is online, we have transcended into a new phase—finding subjective information about everything in the form of reviews. Performing sentiment analysis on reviews left by customers can prove to be essential for service providers as it can help them engineer better products/services, understand sales dynamics, and target audiences appropriately. The more detailed the analysis, the deeper the insight gained. Due to this, sentiment analysis of online reviews has become a project taken up frequently, prompting data scientists to compile and share numerous data repositories. For this research, we have selected one such dataset consisting of reviews for several products and attempted to build an intelligent model using Bidirectional LSTM that can classify reviews as positive, negative, and neutral, rather than categorizing them only as positive and negative.

The rapid development of information and communication technology, particularly in social media platforms, has created an environment where users actively share their experiences and opinions related to various services and applications. One platform that has gained significant popularity is TikTok, a video-sharing application that has become a global phenomenon. With the increasing number of TikTok users, user reviews on distribution platforms such as the Google Play Store have become a crucial source of information. Sentiment analysis of these reviews can provide deep insights into how users respond to the application, while also offering valuable feedback for developers. The research aims to conduct sentiment analysis of TikTok user reviews on the Google Play Store using the Term Frequency-Inverse Document Frequency (TF-IDF) weighting method and the Support Vector Machine (SVM) algorithm as a classification method to achieve optimal results. There are three main stages: the initial stage involves data collection and data pre-processing, followed by the pattern recognition stage, which includes TF-IDF weighting and SVM classification. The final stage consists of evaluation and analysis. The opinion classification obtained includes three categories: positive, negative, and neutral. Based on the evaluation results, the proposed method successfully achieved high accuracy for the 70-30% training-testing split, reaching 84%. The conclusion drawn from these evaluation results indicates that the proposed method can be utilized in the sentiment analysis process of TikTok user reviews.

Sentiment analysis of amazon user reviews using a hybrid approach

PurposeThis study has aimed to thoroughly assess user sentiments and perceptions regarding the National Library of India (NLI). It has attempted to provide significant insights into user satisfaction by examining its strengths and shortcomings across key categories including collection, environment, facilities, location, management and staff. The study has contributed to the understanding of the factors influencing the attributes of libraries, facilitating improvements in services and enhancing the overall user experience.Design/methodology/approachThis study employed a mixed-methods approach combining quantitative and qualitative analysis to assess user sentiments towards the NLI. Using Google Maps reviews, the study utilized web scraping, content analysis and sentiment analysis to categorize reviews as positive, negative or neutral, providing insights into user experiences and an in-depth analysis of the views and opinions of the NLI.FindingsThe study involved sentiment analysis and content analysis of 818 Google Maps reviews to assess user satisfaction with the NLI. The results demonstrate 624 reviews as positive sentiments, 70 instances pinpointed specific negative concerns, primarily related to staff behavior and certain facilities, and 124 neutral reviews suggested mixed viewpoints among users. This analysis highlights the critical role of attributes such as collection quality, environment and facilities in shaping user perceptions, emphasizing the need for focused improvements based on user feedback. The study revealed six attributes, namely collection, environment, facilities, location, management and staff that influence the user perception.Research limitations/implicationsThis study is focused solely on Google Maps reviews. Hence, the results cannot be generalized to all online platforms. Reliance on online reviews may not fully capture the views of all the library users. Additionally, the scope of the study is limited to English-language comments, potentially overlooking valuable insights from non-English-language reviews.Practical implicationsThis study provides valuable insights for the NLI to enhance user satisfaction by addressing the specific concerns raised in online reviews. The findings offer actionable guidance for library management to refine services and maintain favorable public perceptions.Originality/valueThis study contributes to the field by providing a comprehensive analysis of user sentiment through sentiment and content analysis of online reviews, offering unique insights into the NLI’s public perception. The identification of key strengths and weaknesses adds practical value to library management for refining services. The originality of this study lies in its unique approach to evaluating user experiences, which guides future research and improvement efforts in library services.

The global prevalence and popularity of video gaming continue to rise, with games rapidly advancing in graphics, controls, storylines, and device hardware optimization. Today, video games span a wide range of platforms, including desktops, smartphones, tablets, consoles, virtual reality (VR) headsets, and even smartwatches. Despite platform differences, all games are designed to engage players and foster a positive player experience (PX), driven by diverse motivations. In this digital age, players frequently share feedback through star ratings and descriptive reviews, offering valuable insights into playability and PX. Analyzing this user-generated content is essential for understanding current trends and improving player-centric game design across platforms. This study focuses on smartphone games, motivated by the widespread global use of smartphones. We analyzed 4,595 player reviews from 289 smartphone games across 33 genres on Android and iOS platforms. Our three-step process involved: (1) collecting player reviews, (2) conducting binary sentiment analysis using both machine learning (ML) and lexicon-based approaches, and (3) performing thematic analysis on reviews classified by the best-performing sentiment analysis method. We found that the lexicon-based approach outperformed the machine learning approach, achieving an F1 score of 0.91 compared to the best performing ML model’s F1 score of 0.85. While the thematic analysis approach identified 23 playability and player experience factors grouped into four main categories: in-game advertisements (6 factors), sales and customer service (5 factors), technical issues (9 factors), and game design and gameplay (3 factors). Within the game design and gameplay category, we further identified a total of 34 sub-factors distributed across its three factors i.e., game design (3 sub-factors), gameplay (23 sub-factors), and players’ subjective perspectives (8 sub-factors). This study contributes to the fields of games user research, smartphone games, and player experience through analysis of reasonably large number of player reviews from a significant sample of smartphone games. Further, this work lays the groundwork for more specialized and in-depth research in smartphone gaming domain, thereby deepening our understanding of playability and player experience of smartphone games as well as advancing the notion for a platform-centric game design.

This study investigates sentiment analysis methodologies within the framework of CRISP-DM (Cross-Industry Standard Process for Data Mining), aiming to discern the efficacy of various algorithms in sentiment classification tasks. The research uses a structured approach to evaluate SVM, NBC, DT, and K-NN algorithms with the SMOTE oversampling technique, uncovering distinct performance metrics and limitations. Results indicate SVM achieving 59.88% accuracy, NBC at 59.25%, DT with 52.09%, and K-NN obtaining 54.80%, highlighting the differential precision, recall, and f-measure. Additionally, content analysis identifies pertinent themes such as Biometric security, Cloud storage, and Emotion Analysis, enriching sentiment dynamics comprehension. The toxicity scores of analyzed videos reveal nuanced sentiment nuances, with the first video exhibiting Toxicity: 0.13227 and the second scoring Toxicity: 0.12794. This study underscores the significance of informed algorithm selection and evaluation methodologies within CRISP-DM, fostering optimized sentiment analysis outcomes while acknowledging diverse topical nuances.

People share their comments and reviews on public platforms in advanced social media systems. The customer’s perception of the product is reviewed by analysing the sentiment of product reviews, thus assisting in business decision-making. In most of the prevailing works, the sentence type of product review was not recognised to analyse the sentiment; thus, the complexity of the sentiment analysis process increased. Thus, this study performs sentence-type assessment-based product review sentiment analysis using beta divergence divide and conquer (BeDi-DC) and Log-Squish Convolutional Neural Network (Log-Squish CNN). Initially, the input product review data were preprocessed, followed by word count extraction. Next, the data were clustered with the Permutation Distribution Hierarchical Clustering (PerDHC) algorithm and classified into real and fake reviews by the proposed Log-Squish CNN approach. Subsequently, the BeDi-DC technique was used to identify the sentence types of real reviews. Word sense disambiguation is performed on the multi-target review to identify the exact target. Next, to analyse the sentiwords and their score values, the Mean-Senticircle Method (MeSM) was utilised. Finally, using the Log-Squish CNN model, the sentiment of the review was classified as positive, neutral, or negative. The accuracy, f-measure, and degree correlation attained by the proposed model are 98.99%, 98.78%, and 0.845, respectively, thus outperforming the prevailing models.

Sentiment analysis, which deals with people’s sentiments as they appear in the growing amount of online social data, has been on the rise in the past few years. In its simplest form, sentiment analysis deals with the polarity of a given text, i.e., whether the opinion expressed in it is positive or negative. Sentiment analysis, or opinion mining applications on websites and the social media range from product reviews and brand reception to political issues and the stock market. The vast majority of the research in sentiment analysis has mostly dealt with English data, where there’s an abundance of readily available and annotated for sentiment corpora. With a few notable exceptions, the research in other minor languages such as Greek is lacking. This paper deals with sentiment analysis of electronic product reviews written in Greek. To this end, a small dataset of 480 positive and negative reviews is compiled and used, taken from the popular Greek e-commerce website, www.skroutz.gr. Different computational models for training and testing the dataset are evaluated, ranging from simple Naive Bayes with n-gram features to state-of-the-art BERT. The results look very promising for such a small corpus.

The popularity of the Internet has brought profound influence to electronic commerce. A kind of review-oriented consumption mode is gradually expanding in the market and consumers will refer to the reviews provided by consumers who bought the product in the past. How to accurately analyze users' sentiments from massive data of e-commerce reviews has become one of the key issues for e-commerce platforms. Current standard sentiment analysis classifies overall sentiment of e-commerce reviews without an extended description of the entity. We set up an optimized Aspect-based sentiment analysis (ABSA) that includes four elements: aspect, category, polarity, and opinion. Aiming at the above problems, this paper proposes a Chinese e-commerce reviews sentiment analysis algorithm based on BERT. By using pre-training model, we use the BIO(B-begin,I-inside,O-outside) data labeling pattern to label entities and study sentiment analysis by the annotation data. Experimental results on the Taobao cosmetics review datasets show that compared with the ordinary deep learning methods, our approach in the accuracy rate and the F1 score has significant improvement.