Trends in Online Patient Perspectives of Neurosurgeons: A Sentiment Analysis

As people increasingly make hotel booking decisions relying on online reviews, how to effectively improve customer ratings has become a major point for hotel managers. Online reviews serve as a promising data source to enhance service attributes in order to improve online bookings. This paper employs online customer ratings and textual reviews to explore the bidirectional performance (good performance in positive reviews and poor performance in negative reviews) of hotel attributes in terms of four hotel star ratings. Sentiment analysis and a combination of the Kano model and importance-performance analysis (IPA) are applied. Feature extraction and sentiment analysis techniques are used to analyze the bidirectional performance of hotel attributes in terms of four hotel star ratings from 1,090,341 online reviews of hotels in London collected from TripAdvisor.com (accessed on 4 January 2022). In particular, a new sentiment lexicon for hospitality domain is built from numerous online reviews using the PolarityRank algorithm to convert textual reviews into sentiment scores. The Kano-IPA model is applied to explain customers’ rating behaviors and prioritize attributes for improvement. The results provide determinants of high/low customer ratings to different star hotels and suggest that hotel attributes contributing to high/low customer ratings vary across hotel star ratings. In addition, this paper analyzed the Kano categories and priority rankings of six hotel attributes for each star rating of hotels to formulate improvement strategies. Theoretical and practical implications of these results are discussed in the end.

BackgroundClinical data in social media are an underused source of information with great potential to allow for a deeper understanding of patient values, attitudes, and preferences.ObjectiveThis tutorial aims to describe a novel, robust, and modular method for the sentiment analysis and emotion detection of free text from web-based forums and the factors to consider during its application.MethodsWe mined the discussion and user information of all posts containing search terms related to a medical subspecialty (oculoplastics) from MedHelp, the largest web-based platform for patient health forums. We used data cleaning and processing tools to define the relevant subset of results and prepare them for sentiment analysis. We executed sentiment and emotion analyses by using IBM Watson Natural Language Understanding to generate sentiment and emotion scores for the posts and their associated keywords. The keywords were aggregated using natural language processing tools.ResultsOverall, 39 oculoplastic-related search terms resulted in 46,381 eligible posts within 14,329 threads. Posts were written by 18,319 users (117 doctors; 18,202 patients) and included 201,611 associated keywords. Keywords that occurred ≥500 times in the corpus were used to identify the most prominent topics, including specific symptoms, medication, and complications. The sentiment and emotion scores of these keywords and eligible posts were analyzed to provide concrete examples of the potential of this methodology to allow for a better understanding of patients’ attitudes. The overall sentiment score reflects a positive, neutral, or negative sentiment, whereas the emotion scores (anger, disgust, fear, joy, and sadness) represent the likelihood of the presence of the emotion. In keyword grouping analyses, medical signs, symptoms, and diseases had the lowest overall sentiment scores (−0.598). Complications were highly associated with sadness (0.485). Forum posts mentioning body parts were related to sadness (0.416) and fear (0.321). Administration was the category with the highest anger score (0.146). The top 6 forum subgroups had an overall negative sentiment score; the most negative one was the Neurology forum, with a score of −0.438. The Undiagnosed Symptoms forum had the highest sadness score (0.448). The least likely fearful posts were those from the Eye Care forum, with a score of 0.260. The overall sentiment score was much more negative before the doctor replied. The anger, disgust, fear, and sadness emotion scores decreased in likelihood, whereas joy was slightly more likely to be expressed after doctors replied.ConclusionsThis report allows physicians and researchers to efficiently mine and perform sentiment analysis on social media to better understand patients’ perspectives and promote patient-centric care. Important factors to be considered during its application include evaluating the scope of the search; selecting search terms and understanding their linguistic usages; and establishing selection, filtering, and processing criteria for posts and keywords tailored to the desired results.

Background Physician review websites have a significant influence on patients with regard to provider selection. The goal of this study was to utilize machine learning to understand what influences patient ratings for shoulder and elbow surgeons online. Methods All reviews and ratings were obtained from Healthgrades.com. The “Valence Aware Dictionary and sEntiment Reasoner” (VADER) software was used to score the ratings on a scale between -1 and +1. Word frequency analysis was also performed to provide context to the words used to describe surgeons; both positive and negative word frequencies were analyzed. A multiple logistic regression identified the odds these words/phrases were associated with a sentiment score >0.50. Both star ratings (out of 5 points) and sentiment scores were collected. Results 493 surgeons (6,381 reviews) were included. There was no difference in ratings based on gender or geography. Age analysis indicated that star ratings and sentiment scores were lower for older providers (p<0.01). “Pain” was the most commonly used phrase in both the best and worst reviewed surgeon reports. When a comment included the phrases “knowledgeable”, “confident”, “listen[s]”, “recommend”, or “comfortable”, surgeons’ online ratings were 1.6x, 2.7x, 3.2x, 2.6x, and 3.8x more likely to be positive (p<0.01). Approximately 1 out of 5 reviews included mention of ancillary characteristics (e.g., “wait”, “front desk”, “office”). Conclusion This unique code allows surgeons to analyze their field, from the individual to health system level, to see how they are being reviewed online. For shoulder and elbow surgeons, this study showed that more positive online reviews were seen for surgeons who are younger, have reduced office wait times, and have overall positive patient perceptions. Pain and pain management were the primary determinants of overall scores. As online ratings can influence a patient’s choice of provider, surgeons should consider implementing these findings to optimize their practice.

A typical trade-off in decision-making is between the cost of acquiring information and the decline in decision quality caused by insufficient information. Consumers regularly face this trade-off in purchase decisions. Online product/service reviews serve as sources of product/service related information. Meanwhile, modern technology has led to an abundance of such content, which makes it prohibitively costly (if possible at all) to exhaust all available information. Consumers need to decide what subset of available information to use. Star ratings are excellent cues for this decision as they provide a quick indication of the tone of a review. However there are cases where such ratings are not available or detailed enough. Sentiment analysis - text analytic techniques that automatically detect the polarity of text - can help in these situations with more refined analysis. This study was performed in two interrelated phases. In the first phase the potential impact of Sentiment Scores (sentiment analysis outcomes) was investigated through a comparison between these scores with an already established numerical rating denoted as star ratings in three different domains. The results show that sentiment scores tend to fall into neutral areas and are not able to detect extremes that were reported to be more beneficial for information acquisition purposes. As a result, to use the current sentiment analysis results as a substitute for star ratings, a partial linear filter was applied to sentiment analysis results in a way to highlight the subtle differences away from the "neutral zone". In the second phase, the impact of the extended version of sentiment scores on decision outcomes was examined through a controlled experiment. The examined decision was a purchase decision and the information provided was pages of reviews annotated with extended sentiment scores on each paragraph. Human subjects were used in the experiment and controlled data gathering sessions was designed. Results suggest that female consumers may use sentiment scores on review documents without other comparison aids to increase their confidence level in their purchase decisions.

A typical trade-off in decision-making is between the cost of acquiring information and the decline in decision quality caused by insufficient information. Consumers regularly face this trade-off in purchase decisions. Online product/service reviews serve as sources of product/service related information. Meanwhile, modern technology has led to an abundance of such content, which makes it prohibitively costly (if possible at all) to exhaust all available information. Consumers need to decide what subset of available information to use. Star ratings are excellent cues for this decision as they provide a quick indication of the tone of a review. However there are cases where such ratings are not available or detailed enough. Sentiment analysis - text analytic techniques that automatically detect the polarity of text - can help in these situations with more refined analysis. This study was performed in two interrelated phases. In the first phase the potential impact of Sentiment Scores (sentiment analysis outcomes) was investigated through a comparison between these scores with an already established numerical rating denoted as star ratings in three different domains. The results show that sentiment scores tend to fall into neutral areas and are not able to detect extremes that were reported to be more beneficial for information acquisition purposes. As a result, to use the current sentiment analysis results as a substitute for star ratings, a partial linear filter was applied to sentiment analysis results in a way to highlight the subtle differences away from the "neutral zone". In the second phase, the impact of the extended version of sentiment scores on decision outcomes was examined through a controlled experiment. The examined decision was a purchase decision and the information provided was pages of reviews annotated with extended sentiment scores on each paragraph. Human subjects were used in the experiment and controlled data gathering sessions was designed. Results suggest that female consumers may use sentiment scores on review documents without other comparison aids to increase their confidence level in their purchase decisions.

The recent trend in the Indian menswear market has witnessed the infusion of western styles. Result of this is a promising market in India for men’s fashion accessories like caps, sunglasses, bracelets, rings etc. One among the accessories is earrings for man. This product category has got line extensions like studs, hoops, non-piercing magnetic type and piercing studs etc. Keeping in consideration the whooping growth of the men’s earrings market in the online purchase platforms, this study is done with an objective to explore the reviews for this product category to arrive at some insights. For this study descriptive research design has been adopted. Using the scraper tool in Python software, the data (user’s reviews for men’s earrings) was collected from the top two online vendors in India (Amazon and Flipkart). Sentiment analysis is done with R software using the text analytical package “sentiment”. Then the sentiment scores was deployed in ANOVA to test for any significant differences in star ratings and the product variant purchased by the customer taking the Comment Sentiment Scores and Title Sentiment Score. The results have shown that there is a significant difference between the star ratings and Comment Sentiment Scores, Title Sentiment Score. Bivariate correlation is applied to test the relationship between Comment Sentiment Score, Title Sentiment Score and star ratings. The result revealed that star ratings, Comment Sentiment Scores and Title Sentiment Score have a significant relationship with each other.

Sentiment Analysis of Online Patient-Written Reviews of Vascular Surgeons

Application of sentiment and word frequency analysis of physician review sites to evaluate refractive surgery care

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.

What are patients saying about you online? A sentiment analysis of online written reviews on Scoliosis Research Society surgeons.

Electronic word of mouth (eWOM) is a good source of information, and this includes customer reviews. Through this review, consumers make informed decisions. In this study, the researchers utilized Google Maps Reviews of customers of three known coffee shops. A google map review scraper was used to extract all customer's reviews and star ratings. In order to extract important information from reviews, opinion mining was done. MATLAB R2022a was used for sentiment analysis and opinion pre-processing. Each coffee shop's most popular words are represented using the Bigram model and the bag-of-words technique. This allows for the visual identification of the unique characteristics of these coffee businesses. According to the study's findings, coffee shop B had the most positive average percentage sentiment score (73%), while coffee shop C had the least negative average sentiment score. The Bigram model shows that customers enjoy the coffee these three coffee shops serve. However, when it comes to taste, location, bread, and pastries, coffee shop C has the most words. Lastly, the correlation values for star ratings vs sentiment scores for coffee shops A and B are r=0.4726 and r=0.4812. There is absolutely no association between sentiment score and star ratings for coffee shop C.

Text-driven sentiment analysis has been widely studied in the past decade, on both random and benchmark textual Twitter datasets. Few pertinent studies have also reported visual analysis of images to predict sentiment, but much of the work has analyzed a single modality data, that is either text or image or GIF video. More recently, as the images, memes and GIFs dominate the social feeds; typographic/infographic visual content has become a non-trivial element of social media. This multimodal text combines both text and image defining a novel visual language which needs to be analyzed as it has the potential to modify, confirm or grade the polarity of the sentiment. We propose a multimodal sentiment analysis model to determine the sentiment polarity and score for any incoming tweet, i.e., textual, image or info-graphic and typographic. Image sentiment scoring is done using SentiBank and SentiStrength scoring for Regions with convolution neural network (R-CNN). Text sentiment scoring is done using a novel context-aware hybrid (lexicon and machine learning) technique. Multimodal sentiment scoring is done by separating text from image using an optical character recognizer and then aggregating the independently processed image and text sentiment scores. High performance accuracy of 91.32% is observed for the random multimodal tweet dataset used to evaluate the proposed model. The research further demonstrates that combining both textual and image features outperforms separate models that rely exclusively on either images or text analysis.

Background Online surgeon reviews can significantly influence a patient’s selection of a provider, and are important in the movement towards quality-based physician compensation models. Written reviews, however, are subjective and are thus difficult to quantitatively analyze. Sentiment analysis using artificial intelligence (AI) provides the ability to quantitatively assess surgeon reviews to provide actionable feedback. Purpose The objective of this study is to quantitatively analyze the online written reviews of AOSSM surgeons utilizing sentiment analysis and report trends in the most frequently used words in the best and worst reviews. Study Design Cross-sectional study using publicly-available online reviews Methods Online reviews and star-ratings of AOSSM surgeons were obtained from healthgrades.com and zocdoc.com. A sentiment analysis algorithm was used to compute sentiment analysis scores of each written review. Sentiment scores were validated against star-ratings. Positive and negative word and word-pair frequency analysis was performed to identify common items associated with high and low scores. A multiple logistic regression was run on clinically relevant phrases. Results Following the inclusion and exclusion criteria, 18,386 AOSSM surgeon reviews were analyzed for 2071 surgeons. There was no significant difference in sentiment scores by provider gender. Surgeons who are younger than 50 years old had more positive reviews (mean sentiment = +0.536 versus +0.458, p < 0.01). The most frequently used and meaningful bi-grams used to describe top-rated surgeons are words correlating with kindness, caring personalities, and efficiency in pain management; whereas, those with the worst reviews are often characterized as unable to relieve the pain of their patients. The multiple logistic regression was significant for several clinically relevant words that confer greater or less odds of an improved score. Pain is significantly correlated with a decreased odds of receiving a positive review and positive behavioral factors confer a greater odds of receiving a positive review. Conclusion Sentiment analysis provides a means of quantifying written reviews of surgeons, and analysis of the reviews. This study provides insight into factors contributing to positive reviews, especially surgeon confidence, staff friendliness, warm disposition, and pain relief. Clinical Relevance This study delineates factors that impact the public reviews on AOSSM providers.

Besides technical and fundamental analysis, machine learning and sentiment analysis obtained from non-structural news and comments have been studied extensively in financial market prediction in recent years. It is still uncertain how to combine predictions from news, sentiment scores or financial data. In this study, we provide a methodology to achieve this issue. Besides the methodology, this study differs from previous studies in terms of data coverage and used models in both sentiment analysis and prediction. Our study consists of weekly predictions by ensemble learning and feature selection methods using 683 variables for stocks traded in the Borsa Istanbul 30 index. In addition, we predicted sentiment scores from news of 18 different sectors and combined both predictions with weighted normalized returns. We used Random Forests, Extreme Gradient Boosting and Light Gradient Boosting Machines of ensemble learning methods for predictions. From the parameters such as training set length, estimation methods, variable selection methods, number of variables, and the number of models in the prediction method, we took the combination that gives the best result. For sentiment scores, tests were performed using BERT, Word2Vec, XLNet and Flair methods. Then, we extracted final sentiment scores from the news. With the proposed trade system, we combined the results obtained from these financial variables and the news sentiment scores. Final results show that we achieved a better performance than both predictions made by using sentiment scores and financial data in terms of weekly return and accuracy.KeywordsSentiment analysisEnsemble learningStock Market PredictionFeature selection

This paper aims to analyze the relationship between the students' numerical rating and the qualitative measure of the students' written comments in the faculty evaluation using sentiment analysis. The dataset which consists of the numerical ratings and students' feedback obtained from the faculty evaluation system was used in the experiment. An ensemble model which consists of five machine learning algorithms was used to analyze and identify the polarity of the written comments of the students. The overall sentiment score was computed for each faculty and was compared to the numerical score using the statistical technique, Pearson's correlation coefficient. The result indicates that there is significance but very small relationship between the numerical rating and the overall sentiment scores. Based on the result, universities and colleges should exploit written comments since it is rich with observations and insights about the performance and effectiveness of a teacher. Moreover, sentiment analysis technique can be used to identify students' feeling towards teaching.