Social media analysis for product safety using text mining and sentiment analysis

The rapid growth of financial technology (fintech) services in Indonesia has significantly transformed digital transaction behavior, with digital wallets such as DANA becoming widely adopted. Despite high usage rates, public discourse on social media frequently highlights both positive experiences and recurring concerns related to system reliability and data security. Twitter, as a real-time public communication platform, provides a rich source of user-generated content that reflects public perception toward fintech services. This study investigates public sentiment toward DANA fintech services in Indonesia through sentiment analysis of Twitter data using machine learning approaches. Purpose:This study aims to analyze public perception of DANA fintech services based on sentiment expressed on Twitter and to compare the performance of several machine learning classification algorithms in identifying positive, neutral, and negative sentiments in Indonesian-language social media texts. Methods/Study design/approach: The study employs a quantitative descriptive approach using text mining and sentiment analysis techniques. Twitter data were collected via the Twitter API over a specified period and processed through preprocessing stages including text cleaning, tokenization, stopword removal, and stemming. Feature extraction was performed using the Term Frequency–Inverse Document Frequency (TF-IDF) method. Sentiment classification was conducted using Support Vector Machine (SVM), K-Nearest Neighbor (KNN), and Random Forest algorithms. Model performance was evaluated using accuracy, precision, recall, and F1-score metrics. Result/Findings:The results indicate that public sentiment toward DANA is predominantly positive, driven by perceptions of transaction convenience and speed, followed by neutral sentiment, while negative sentiment is mainly associated with system disruptions and data security issues. Among the evaluated algorithms, SVM achieved the highest classification accuracy (91.4%), outperforming Random Forest and KNN. These findings demonstrate that SVM is more effective for sentiment classification of short Indonesian-language texts on social media platforms. Novelty/Originality/Value:This study contributes to the growing body of sentiment analysis research in Indonesia by providing empirical evidence on public perception of fintech services using machine learning techniques. By systematically comparing multiple classification algorithms in the context of Indonesian Twitter data, the study offers methodological insights and practical implications for fintech service providers in improving service quality, communication strategies, and user trust.

This study aims to analyze the sentiment of the Hindu community toward religious issues on social media, with a specific focus on Kapuas Regency, Central Kalimantan. The development of social media as a digital public sphere has positioned platforms such as Instagram and Twitter as primary spaces for the spontaneous and open expression of opinions, perceptions, and religious attitudes. This study is important because religious issues in digital spaces often influence social harmony and shape religious discourse within society. The research employs a quantitative approach using text mining and sentiment analysis based on Natural Language Processing (NLP). Primary data were collected through web scraping techniques, utilizing Apify for Instagram and asynchronous Python scripts for Twitter, with relevant keywords, hashtags, and geographic indicators. The analysis process includes text preprocessing (cleaning, tokenization, stopword removal, and stemming), followed by sentiment classification using a lexicon-based approach with the InSet dictionary into three categories: positive, negative, and neutral. The analysis results were evaluated using a confusion matrix, along with precision, recall, and F1-score metrics to assess model reliability. The findings indicate that positive sentiment predominates on both Instagram and Twitter, followed by neutral sentiment, while negative sentiment appears in only a small proportion. Positive sentiment is generally associated with expressions of prayer, gratitude, tolerance, and calls for togetherness, whereas negative sentiment tends to emerge in discussions related to ritual differences or responses to socio-religious controversies. The sentiment analysis model achieved an accuracy of 100% on Instagram data (self-evaluation) and 74.4% on Twitter data (manual evaluation), with relatively high precision and recall values, indicating that the results are statistically reliable.

This study conducted text mining and sentiment analysis on 800 Emirates Airline customer reviews from Skytrax to evaluate airline service quality.Through frequency analysis, TF-IDF analysis, and 3-gram analysis, key service elements were identified, while sentiment analysis and correlation analysis were used to analyze factors affecting service satisfaction.The analysis revealed strong correlations between customer satisfaction and ground service (0.55), cabin staff service (0.49), food service (0.48), and seat comfort (0.44).Notably, in the TF-IDF analysis, the 'business-class' keyword rose from 10th to 6th in frequency ranking, highlighting the importance of premium service.The 3-gram analysis identified transit service and special meal requirements as key areas of customer concern.Time series analysis indicated a decline in service satisfaction following the pandemic, suggesting the need for improvement.The theoretical significance of this study is that it presents a comprehensive analytical framework that combines text mining and sentiment analysis to analyze the service quality of FSC Airlines.It also made methodological contributions by capturing dynamic changes in service quality perception through time series-based sentiment analysis.From a practical perspective, the study proposed specific improvements in ground services, standardization of cabin crew service, enhancement of special meal services, and innovation in business class seating.While this research is expected to advance airline service quality management, it has limitations in analyzing only English reviews from a single platform.Future research should include various platforms and language groups.

This study proposes a new model for service quality measurement using sentiment analysis and text mining techniques. This model aims to overcome traditional methods' time, cost and implementation difficulties and provide a more dynamic and efficient approach to service quality measurement. In addition, in this model, instead of the dimensions used in service quality measurements, such as SERVQUAL or SERVPERF, it is shown how to determine new categories and keywords specific to the service sector in which the model is used by text mining. Thus, it is aimed at something other than reaching more accurate results in service quality measurement. To achieve the model’s purpose, it aims to develop a service quality measurement model using social media data processed by text mining and sentiment analysis. To find an answer to this question, the keywords "flood", "meter", "rain", "irrigation", "infrastructure", "sewerage", "sewage", "maintenance hole ", "aski", "waterless", "water" were extracted from 109.844 tweets sent to the Twitter account of a municipality between 2016 and 2022 by text mining method. Service quality was measured by subjecting 5766 tweets containing the keywords extracted to sentiment analysis. As a result of the service quality measurement, 1922 negative, 973 positive and 2871 neutral tweets were identified. The average negative score was 0.51, the average positive score was 0.11, and the average neutral score was 0.38.

PurposeOnline customer reviews could shed light into their experience, opinions, feelings, and concerns. To gain valuable knowledge about customers, it becomes increasingly important for businesses to collect, monitor, analyze, summarize, and visualize online customer reviews posted on social media platforms such as online forums. However, analyzing social media data is challenging due to the vast increase of social media data. The purpose of this paper is to present an approach of using natural language preprocessing, text mining and sentiment analysis techniques to analyze online customer reviews related to various hotels through a case study.Design/methodology/approachThis paper presents a tested approach of using natural language preprocessing, text mining, and sentiment analysis techniques to analyze online textual content. The value of the proposed approach was demonstrated through a case study using online hotel reviews.FindingsThe study found that the overall review star rating correlates pretty well with the sentiment scores for both the title and the full content of the online customer review. The case study also revealed that both extremely satisfied and extremely dissatisfied hotel customers share a common interest in the five categories: food, location, rooms, service, and staff.Originality/valueThis study analyzed the online reviews from English-speaking hotel customers in China to understand their preferred hotel attributes, main concerns or demands. This study also provides a feasible approach and a case study as an example to help enterprises more effectively apply social media analytics in practice.

Purpose This study examines the role of online customer reviews through text mining and sentiment analysis to improve customer satisfaction across various services within the UK banking sector. Additionally, the study analyses sentiment trends over a five-year period. Design/methodology/approach Using DistilBERT and Support Vector Machine algorithms, customer sentiments were assessed through an analysis of 20,137 Trustpilot reviews of HSBC, Santander, and Tesco Bank from 2018 to 2023. Data pre-processing steps were implemented to ensure data integrity and minimize noise. Findings Both positive and negative sentiments provide valuable insights. The results indicate a high prevalence of negative sentiments related to customer service and communication, with HSBC and Santander receiving 90.8% and 89.7% negative feedback, respectively, compared to Tesco Bank’s 66.8%. Key areas for improvement include HSBC’s credit card services and call center efficiency, which experienced increased negative feedback during the COVID-19 pandemic. The findings also demonstrate that DistilBERT excelled in categorizing reviews, while the SVM model, when combined with customer ratings, achieved 96% accuracy in sentiment analysis. Research limitations/implications This study focuses on UK bank consumers of HSBC, Santander, and Tesco Bank. A multi-country or cross-cultural study may further enhance our understanding of the approaches and findings. Practical implications Online customer reviews become more informative when categorised by service sector. To enhance customer satisfaction, bank managers should pay attention to both positive and negative reviews, and track trends over time. Originality/value The uniqueness of this study lies in its exploration of the importance of categorisation in text-mining-based sentiment analysis, its focus on the influence of both positive and negative sentiments, and its emphasis on tracking sentiment trends over time.

Background Understanding the attitudes and perceptions of the general population is necessary for organizing health promotion initiatives. During outbreaks, social media has a significant impact on creating social perceptions. This study aims to identify and examine the emotions expressed and topics of discussion among Indian citizens related to COVID-19 third wave, from the messages posted on Twitter using text mining techniques. Methods Twitter messages (tweets) were downloaded using Twitter API from June 1, 2021 to July 10, 2021. After pre-processing the downloaded messages, 8933 unique tweets from various individuals were taken into account for the analysis. To identify and extract emotions expressed by the people in the Twitter texts, the text mining and sentiment analysis package "syuzhet" in R software was used. In order to identify the concerns and themes of discussion by the public during the pandemic, topic analysis was done using the Latent Dirichlet Allocation (LDA) technique. To understand and measure the tweets' reachability in relation to the themes and emotions conveyed, an engagement metrics analysis was also conducted. Results According to the emotional analysis performed on Twitter messages about the COVID-19 third wave, anticipation was exhibited maximum in 4180 tweets, followed by fear in 4070, and trust in 4001 tweets. Results of topic modeling revealed that there were widespread discussions and concerns about preventative measures to deal with the COVID-19 third wave. Engagement metrics verified that the greatest number of individuals liked and retweeted tweets expressing disgust. Maximum people favorited tweets with information on preventive measures for COVID-19, and a large number of individuals re-shared messages comparing various aspects of different COVID-19 waves. Conclusion Data from social media platforms can be used to comprehend public opinions and emotions during pandemics and emergencies. This can assist public health stakeholders in managing pandemic circumstances by planning effective health communication strategies that quickly reach a larger audience. The study's findings will assist stakeholders and public health experts in effectively using social media to communicate information about COVID-19 that combat people's negative emotions and concerns. Future research can use a similar approach to comprehend people's perspectives and concerns during outbreaks and emergency circumstances.

In the internet age, social media and mobile devices are the most important tools of communication and marketing in the exhibition and event industry. Only a limited research has explored exhibition visitors’ engagement and preference perception through social networking media. Our study explores relationship between Facebook fan pages and visitor engagements of the exhibitions. The study found that the number of visitors, Facebook fans like counts, comment counts, and emotional factors (sentiment polarity) have a significant positive correlation. On the other hand, numbers of visitor rapidly grow every year, but comment sentiment polarity of a positive emotion drops every year. This reduction is the warning sign that the exhibition management and marketing strategy need to be improved in order to continue to have a positive visitor engagement emotion. Our linear Discriminate Analysis (LDA) result found that visitors were highly concerned with gamification activities for an incentive prize both to create involvement and to enhance exhibition experience. Our finding also shows the relevance of both FB like counts and comments generated by social media in relation to the visitor engagement performance. This pioneer study utilized both text mining and sentimental analysis into the exhibition study, especially the area of Asian research.

Text analytics are well-known in the modern era for extracting information and patterns from text. However, no study has attempted to illustrate the pattern and priorities of newspaper headlines in Bangladesh using a combination of text analytics techniques. The purpose of this paper is to examine the pattern of words that appeared on the front page of a well-known daily English newspaper in Bangladesh, The Daily Star, in 2018 and 2019. The elucidation of that era’s possible social and political context was also attempted using word patterns. The study employs three widely used and contemporary text mining techniques: word clouds, sentiment analysis, and cluster analysis. The word cloud reveals that election, kill, cricket, and Rohingya-related terms appeared more than 60 times in 2018, whereas BNP, poll, kill, AL, and Khaleda appeared more than 80 times in 2019. These indicated the country’s passion for cricket, political turmoil, and Rohingya-related issues. Furthermore, sentiment analysis reveals that words of fear and negative emotions appeared more than 600 times, whereas anger, anticipation, sadness, trust, and positive-type emotions came up more than 400 times in both years. Finally, the clustering method demonstrates that election, politics, deaths, digital security act, Rohingya, and cricket-related words exhibit similarity and belong to a similar group in 2019, whereas rape, deaths, road, and fire-related words clustered in 2018 alongside a similar-appearing group. In general, this analysis demonstrates how vividly the text mining approach depicts Bangladesh’s social, political, and law-and-order situation, particularly during election season and the country’s cricket craze, and also validates the significance of the text mining approach to understanding the overall view of a country during a particular time in an efficient manner.

Social media has become a major aspect of online activity, and thus an essential part of cybercrime and cyber terrorism-related operations. As LEA’s (law enforcement agencies) focus upon cybercrime and cyber terrorism threats increase, so does the requirement to consider the potential application of social media as a vital aspect of any cyber defense strategy. This chapter considers social media's role in society and the characteristics that influence the way people use it. A number of case study scenarios are presented to demonstrate the potential application of social media in combatting a variety of criminal threats including human trafficking, hostage situations, and other organized crime. Each scenario considers the potential impact of a number of example technologies such as text mining, NLP (natural language processing) and sentiment analysis supported by an understanding of the relevance of people's usage characteristics and behaviors within social media.

Nowadays, with the increasing number and use of social media platforms, people now share their experiences about a product they have bought or a place they have been to on social media platforms more frequently. Considering the volume of data on social media platforms, it is considered that there is some meaningful information for institutions or companies in the reviews and experiences shared on social media platforms. As such, it is important to improve the methods of extracting meaningful information from the reviews and experiences shared on social media and to know which method is better. In this study, the classification successes of the bag of words and the fastText word representation methods, which are among the word representation methods in sentiment analysis methods mentioned above, were compared by using Turkish reviews performed for touristic places. Besides, while performing the comparison process, it was measured whether the process of separating the words into their roots and negation of the words, which is the preliminary stage of the sentiment analysis process, contributed to the classification success. In the study, both two-class (positive, negative) sentiment analysis and three-class (positive, negative, neutral) sentiment analysis were performed. Six data sets were created to carry out the mentioned comparison operations. The data sets were first classified using the Naive Bayes (NB), Multinomial Naive Bayes (MNB), k-Nearest Neighbor (k-NN) and Support Vector Machines (SVM) algorithms, which are frequently used in text mining, and based on bag of words word representation method, they were classified with WEKA program. After the test results of all data sets were obtained according to the bag of words word representation method, the tests of the fastText word representation method were carried out using the fastText library of the Python programming language. Classification procedures were carried out with 10-fold cross-validation methods, and f-score values of the classification processes were obtained. Finally, it was determined that bag of words word representation method performed a more successful classification than the fastText word representation method in two-class emotion analysis, while the fastText word representation method performed a more successful classification process than bag of words word representation method in three-class emotional analysis. It was observed that the process of separating the words into their roots and negating the words, which are the preliminary processes of sentiment analysis, did not contribute positively or negatively to the classification processes performed with the fastText word representation method. However, it was determined that it had a minor contribution to sentiment analysis processes performed by using bag of words word representation method. In the two-class sentiment analysis, the most successful classification result was achieved by using the machine learning model created with the SVM algorithm with the value of 0.91 f-score employing bag of words word representation method. In the three-class sentiment analysis, the most successful classification result was achieved with the machine learning model created using the fastText word representation method with the value of 0.78 f-score.

This study aims to analyse consumer experiences, purchase behaviours, and emotional responses through Booking and Expedia’s mobile applications. The 2000 user reviews collected from Google Play were subjected to a comprehensive sentiment analysis, text mining, and topic modelling process to identify the key elements that shape consumers’ emotional experiences and purchase decisions. According to the results of text mining and sentiment analysis performed with Python’s WordNet library, 81.9% of Booking.com reviews are positive, 8.4% are negative, and 11.3% are neutral, whereas 55.8% of Expedia reviews are positive, 37.8% are negative, and 8.0% are neutral. In the topic modelling analysis, Booking.com emphasised ease of booking, while Expedia emphasised difficulties in cancellation and refund processes. These findings provide valuable insights into how consumers’ emotional states and purchasing behaviours are reflected in their experiences with mobile applications. The study enables the development of strategic recommendations for marketing management to better analyse consumers’ expectations and experiences.

From measurement scale to sentiment scale: Examining the effect of sensory experiences on online review rating behavior

The COVID-19 pandemic has changed the lives of consumers in virtually every nation. Based upon the theory of psychological reactance and psychoevolutionary theory of emotion, we hypothesize how such lifestyle changes affect consumers perceiving and responding to companies' communications messages. The theories also suggest that consumers in different cultures may respond to COVID-19 differently. To test our hypotheses, we implemented a Python scraper to collect companies’ Instagram posts pre- and during the COVID-19 lockdown. A machine learning algorithm was applied on the collected post photos to automatically identify certain photo characteristics, such as indoor versus outdoor, and with a single person versus many people; a text mining and sentiment analysis was implemented on the collected post captions to identify the salient emotion each caption exhibited, such as joy and anticipation. After that, we conducted a regression discontinuity analysis of photo characteristics or caption emotion on number of likes or comments to identify consumers’ response change due to the COVID-19 pandemic. The estimation results supported our hypotheses and suggested tactics that could improve consumer communications effectiveness in this changed time. Viewing COVID-19 as an example of a current event in the ever-changing world, this paper suggests that such events could impact consumer response and behavior, and that companies’ marketing and advertising strategies should be responsive to such events.Supplementary InformationThe online version contains supplementary material available at 10.1057/s41270-021-00138-3.

The study analyses Turkish and English tweets about climate change on the social media platform Twitter and comparatively examines individuals” perceptions, concerns, and emotional reactions to this issue. A total of 2,046 Turkish and 18,000 English tweets were collected; 1,104 Turkish and 6,449 English tweets were analyzed after the cleaning process. Artificial intelligence-based methods such as text mining, sentiment analysis, and topic modelling are used. Topic modelling with Latent Dirichlet Allocation (LDA) identified prominent themes in tweets in both languages. Sentiment analysis is performed using deep learning techniques to categories tweets into positive, negative, and neutral categories. The findings show that English tweets contain stronger emotional reactions, while Turkish tweets contain a higher proportion of neutral expressions. Additionally, it was observed that the perception of climate change can differ in local and global contexts. Based on a multidimensional analysis of social media data, the study provides valuable insights into the development of environmental communication strategies. The comparison of Turkish and English tweets contributes to understanding the effects of cultural contexts on climate change perception. The findings have important implications for policymakers and environmental awareness campaigns, as they highlight the need for tailored communication strategies that consider cultural differences in climate change perception.