Sentiment Analysis Approach Research Articles

A Topic Modeling and Sentiment Analysis Approach for Benchmarking of Hotels Based on Online Reviews

This paper proposes a methodology that utilizes online reviews for performing competitive and customer-driven benchmarking. The utilization of online reviews may overcome the difficulty of obtaining information about competitors through conventional approaches, such as surveys. A topic modeling method, Latent Dirichlet Allocation (LDA), identifies the essential aspects discussed in the reviews. Combining the identified aspects with sentiment analysis, the best practices may be discovered with respect to each important aspect. In the case study, ten important aspects are identified, i.e., Room Amenities, Parking Area, Food & Beverage, Check-in/out, Fitness & Recreational Facilities, Hospitality, Room Quality, Staying Experience, Design, and Location. Furthermore, adjective words are used to obtain more specific customer needs regarding an aspect. Those customer needs may suggest improvements that hotels may implement, such as reducing the air-conditioner’s noise and improving the rooms’ soundproofing.

Sentiment Analysis using Latent Dirichlet Allocation for Aspect Term Extraction

This work proposes a sentiment analysis approach for decision-making in product design, analysis, and market share. The approach incorporates user-generated text data in the form of consumer reviews to extract product features using topic-based modeling methods. Latent Dirichlet Allocation (LDA) is employed to extract aspect categories from the data and identify the sentiment of each review using the VADER sentiment analyzer. The performance of the proposed method is evaluated in terms of accuracy, with an achieved result of 80%. The extracted topics are also summarized to provide leads for product design and quality assurance. The approach can be used by manufacturers, retailers, and suppliers to understand customers' opinions about their products better and make better decisions. LDA is a powerful unsupervised method that can extract latent topics from a collection of documents; this method has been widely used in text mining, information retrieval, and natural language processing. The accuracy can be improved by using more sophisticated models or more data.

An Ontology-based Sentiment Analysis Approach To Discovering Hidden Affected Objects

Nowadays, the topic of sentiment analysis has attracted a wide range of artificial intelligence technologies suchas Natural Language Processing (NLP), neural network. Long Short-Term Memory (LSTM), a deep neural network, has proven the effiency in the task of sentiment classification. However, there is a shortage of semantic knowledge from such classification results. This paper introduces a sentiment analysis approach based on a knowledge model to discover hidden affected objects, which combines ONtology and SentiMent technology, named ONSEM. In order to confirm the efficiency of the ONSEM framework, an experiment on a sentiment analysis dataset was developed and then yielded promising experimental results.

Analisis Performa Algoritma NBC, DT, SVM dalam Klasifikasi Data Ulasan Pengunjung Candi Borobudur Berbasis CRISP-DM

The approach of visitor sentiment analysis to Borobudur Temple tourist destinations in Indonesia can be classified using various algorithms to get optimal results. Good algorithm performance can be seen from the confusion matrix (accuracy, precision, recall) value, Area Under Curve (AUC) value, and Receiver Operating Characteristic (ROC). This study used the Naïve Bayes Classifier (NBC), Decision Tree (DT), and Support Vector Machine (SVM) algorithms against 3850 text data obtained from the Tripadvisor website, especially reviews of Borobudur Temple visitors. The method refers to the Cross-Industry Standard Process for Data Mining (CRISP-DM) for optimizing tourist destination products and services by paying attention to six stages: business understanding, data understanding, data preparation, modeling, evaluation, and deployment. The results of this study show that the results of NBC's algorithm performance evaluation can be seen to have a change in the confusion matrix value at the accuracy value from 98.73% to 95.6%, the precision value changed from 98.72% to 98.97%, the recall value also changed from 100% to 96.54%. In addition, the Area Under Curve (AUC) of NBC also changed from 0.500 (50%) to 0.693 (69.35%). In addition, the results of the DT algorithm performance evaluation showed a change in the confusion matrix value at the accuracy value from 97.55% to 94.40%, the precision value increased from 97.63% to 91.86%, the recall value also changed from 99.90% to 99.47%. The Area Under Curve (AUC) of DT value also changed from 0.591 (59.1%) to 0.932 (93.2%). The results of the SVM algorithm performance evaluation showed a change in the confusion matrix value at the accuracy value from 98.73% to 99.41%; the precision value changed from 98.72% to 100%, and the recall value also changed from 100% to 99.01%. The Area Under Curve (AUC) of the SVM value also changed from 0.961 (96.1%) to 1.00 (100%). In addition, the T-test results show that the SVM algorithm is more dominant compared to other algorithms, where the SVM algorithm T-test value is 0.994 compared to the DT algorithm T-test value of 0.944 and the NBC algorithm T-test value of 0.98. Based on the Receiver Operating Characteristic (ROC) value, it can be seen that the DT algorithm also shows good performance in addition to SVM. It indicates that in analyzing the sentiment of visitors to Borobudur Temple, the best-recommended algorithm is the Support Vector Machine

Delineation of The Early 2024 Election Map: Sentiment Analysis Approach to Twitter Data

As a democratic country, the people hold an important role in determining power in Indonesia. The closest political agenda in Indonesia is the 2024 Election. A survey has been conducted by several private survey agencies regarding the 2024 political map which has revealed the top five names, namely Prabowo Subianto, Ganjar Pranowo, Anies Baswedan, Sandiaga Uno, and Ridwan Kamil. This study aims to describe the initial map of the 2024 Election through a sentiment analysis approach to Twitter data. This study uses tweet data that mentions five political figures during 2021. In general, the demographic condition of Twitter users that pros or cons to five political figures, among them: located on the Java, in the age group 19–29 years old, and male. The sentiment analysis method used is supervised learning with different methods for each figure. The difference in methods adjusts the best evaluation value given in each figure. The results showed that the highest positive sentimental tweets and the highest number of pro accounts was about Ganjar Pranowo. On the other hand, the highest negative sentiment and the highest number of contra accounts was about Prabowo Subianto. Many words that often appear on a figure's positive sentiment are expressions of hope, prayer, and support. On negative tweets, the word that comes up a lot relating to the work field or work region of the figures.Â

Defining an Integrated and Computed Methodology Approach for Sentiment and Psychographic Analysis in Tourism Research

High-performance computational resources and artificial intelligence-based tools can enhance tourism research and marketing. However, a formal methodological approach using digital technologies in this field is still missing. This research work presents the preliminary results of defining an integrated computational methodology in tourism research and marketing. In addition, the paper aims to provide guidelines for a methodological approach leveraging technological resources and Big Data. The proposed research method is based on online User-Generated Content (UGC) analysis through a psychographic approach based on the Big Five Model, Sentiment Analysis, and Machine Learning techniques. The study is supported by high-performance computing resources, artificial intelligence-based tools, and open-source Python-based software for data collection, text analysis, and psychographic attribution. Results show a remarkable performance of the BFF prediction model and confirm the role of personality in the tourists’ decision-making and appreciation of a site. Future developments of this project involve using the acquired structured dataset labeled with sentiment and psychographic attribution to create a further prediction model on tourist segments and appreciation as part of a marketing strategy in tourism management. Future research should push forward the development of further integrated and performing computer-based methodology in tourism research and marketing, leveraging the massive amount of data and the potential of high-performance computing techniques. The main contribution of this research effort is twofold: the definition of a general-purpose BFF/Sentiment Analysis methodology and the development of a prediction model from online UGC based on the Big Five personality traits in the tourism research scenario.

A sentiment analysis approach to the prediction of market volatility.

Prediction and quantification of future volatility and returns play an important role in financial modeling, both in portfolio optimisation and risk management. Natural language processing today allows one to process news and social media comments to detect signals of investors' confidence. We have explored the relationship between sentiment extracted from financial news and tweets and FTSE100 movements. We investigated the strength of the correlation between sentiment measures on a given day and market volatility and returns observed the next day. We found that there is evidence of correlation between sentiment and stock market movements. Moreover, the sentiment captured from news headlines could be used as a signal to predict market returns; we also found that the same does not apply for volatility. However, for the sentiment found in Twitter comments we obtained, in a surprising finding, a correlation coefficient of -0.7 (p < 0.05), which indicates a strong negative correlation between negative sentiment captured from the tweets on a given day and the volatility observed the next day. It is important to keep in mind that stock volatility rises greatly when the market collapses but not symmetrically so when it goes up (the so-called leverage effect). We developed an accurate classifier for the prediction of market volatility in response to the arrival of new information by deploying topic modeling, based on Latent Dirichlet Allocation, in order to extract feature vectors from a collection of tweets and financial news. The obtained features were used as additional input to the classifier. Thanks to the combination of sentiment and topic modeling even on modest (essentially personal) architecture our classifier achieved a directional prediction accuracy for volatility of 63%.

Knowledge Transfer Mechanisms Using Text Mining and Sentiment Analysis – Case in an Online Collaboration

This study aims to test, examine, and validate text-based human-machine knowledge transfer (KT) by comparing it with human-human KT. The online discussion experiment was carried out via WhatsApp group chats. Chat sentiment was determined using text mining and sentiment analysis and then compared with the respondent's understanding of the knowledge obtained from interviews. The results have shown that human-machine KT is close to human-human KT. By analyzing the correlation coefficient between the two, it is proven that sentiment indicates an understanding of knowledge. Positive sentiment shows similar or in-line understanding between the source and recipient of knowledge and indicates the achievement of KT objectives. Neutral sentiment indicates incomprehension due to the failure of KT. Meanwhile, negative sentiment is ambiguous; it may indicate an incomprehension or a misunderstanding of the knowledge received. This study contributes to the area of knowledge and sentiments, showing that the effectiveness of text-based KT activity can be identified using the sentiment analysis approach.

A hybrid optimization algorithm using BiLSTM structure for sentiment analysis

Sentiment analysis can assist consumers in providing clear and objective sentiment recommendations based on large amounts of data, and it is helpful in overcoming unclear human flaws in subjective assessments. Existing sentiment analysis methods, on the other hand, must be enhanced in terms of robustness and accuracy. To improve marketing strategies based on product reviews, a reliable mechanism for forecasting sentiment polarity should be implemented. This paper proposes a new approach for sentiment analysis called Taylor–Harris Hawks Optimization driven long short-term memory (THHO- BiLSTM). By incorporating Taylor series in HHO, Taylor–HHO is formed, which aids in improving the BiLSTM classifier's performance by picking optimal weights in the hidden layers. The proposed method was evaluated using Amazon product reviews and reviews from the Taboada corpus benchmark datasets, yielding findings with 96.93% and 93% accuracy, respectively. When compared to existing approaches, the suggested model exceeds them in terms of accuracy. The proposed approach helps manufacturers improve their products based on user feedback.

A Novel Approach for Emotion Detection and Sentiment Analysis for Low Resource Urdu Language Based on CNN-LSTM

Emotion detection (ED) and sentiment analysis (SA) play a vital role in identifying an individual’s level of interest in any given field. Humans use facial expressions, voice pitch, gestures, and words to convey their emotions. Emotion detection and sentiment analysis in English and Chinese have received much attention in the last decade. Still, poor-resource languages such as Urdu have been mostly disregarded, which is the primary focus of this research. Roman Urdu should also be investigated like other languages because social media platforms are frequently used for communication. Roman Urdu faces a significant challenge in the absence of corpus for emotion detection and sentiment analysis because linguistic resources are vital for natural language processing. In this study, we create a corpus of 1021 sentences for emotion detection and 20,251 sentences for sentiment analysis, both obtained from various areas, and annotate it with the aid of human annotators from six and three classes, respectively. In order to train large-scale unlabeled data, the bag-of-word, term frequency-inverse document frequency, and Skip-gram models are employed, and the learned word vector is then fed into the CNN-LSTM model. In addition to our proposed approach, we also use other fundamental algorithms, including a convolutional neural network, long short-term memory, artificial neural networks, and recurrent neural networks for comparison. The result indicates that the CNN-LSTM proposed method paired with Word2Vec is more effective than other approaches regarding emotion detection and evaluating sentiment analysis in Roman Urdu. Furthermore, we compare our based model with some previous work. Both emotion detection and sentiment analysis have seen significant improvements, jumping from an accuracy of 85% to 95% and from 89% to 93.3%, respectively.

Analisis Sentimen Bahasa Indonesia Pada Tempat Wisata Di Kabupaten Sukabumi Dengan Naive Bayes Classifier

Sentiment analysis of comments from visitors to tourist attractions and the public on tourist attractions in Sukabumi Regency which is one of the areas with various categories of tourist objects and is a sector of economic income for the surrounding community or for related parties such as the government and managers, in sentiment analysis research This includes using the Nave Bayes classification algorithm to examine the sentiment of tourist visitors and the performance of the classification model used. The data used in this research was taken from the website from Tripadvisor and Google Maps using a crawling technique, which then processed the data by a pre-processing process and then applied a classification to the data and got a sentiment visualization by processing word frequency on tourist visitor sentiment data. The results of the accuracy of the model used were re-tested with the k-fold cross validation method and the results of sentiment visualization got the frequency of words that most often appear on negative sentiment labels are garbage, beaches, lacking, places, roads, parking, dirty, entering, caring, clean , expensive, pay, manage, good and water.

Research on Logistics Service Assessment for Smart City: A Users’ Review Sentiment Analysis Approach

The innovative development of logistics has become a powerful starting point and strong support for the construction of smart cities. An accurate evaluation of logistics service quality can promote intelligent transformation, and upgrading logistics enterprises can improve the urban public service infrastructure. In this study, we propose a logistics service quality evaluation model based on a combination of sentiment analysis technology and a traditional index evaluation system. With the help of sentiment analysis technology, the model focuses on extracting the sentiment characteristics of logistics service quality from user reviews and analyzing user attitudes from different aspects. We designed a new logistics service quality evaluation index system by improving the SERVQUAL model. The system uses sentiment analysis technology to explore evaluation content through feature extraction and builds relations between the evaluation content and indices. Additionally, we use sentiment orientation analysis with different indices to comprehensively evaluate service quality. The experimental analysis shows that the proposed model and algorithm have high accuracy.

Islamic Accounting Standard: An Intertemporal Sentiment Analysis

This study aims to look at related sentiments “Islamic Accounting Standard” around the world published by journals with that theme. This study uses a qualitative method with a sentiment analysis approach. The data used is secondary data with the theme “Islamic Accounting Standard” obtained from the Dimensions database of 100 scientific publications. Data is processed using software SentiStrength for extracting and classifying sentiments. The results of the study found that neutral sentiment was the highest with a percentage of 39%, followed by negative sentiment with 32% and positive sentiment with a percentage of 29%. Related research Islamic Accounting Standard experienced fluctuations in the number of sentiments throughout the period 1990 to 2023. The highest number of sentiments occurred in 2018 with a total of 13 scientific publications, which were dominated by neutral sentiment of 9 scientific publications. There are several positive and negative issues related to the topic of sharia accounting standards which can be further elaborated.

Factors affecting user experience of contact tracing app during COVID-19: an aspect-based sentiment analysis of user-generated review

ABSTRACT This study aims to identify the critical factors influencing the user experience of contact tracing apps and the sentiments around them. For this purpose, we used Google play reviews of Aarogya Setu, a contact tracing app developed in India. First, we establish the relationship between review sentiment and review rating using regression between sentiment polarity and review rating. Then, we used a hybrid aspect-based sentiment analysis approach that uses unsupervised linguistic techniques to determine statistically significant concepts present in the review texts and cluster them into representative aspects that were then tagged under human supervision. Finally, supervised deep learning methods were applied for exhaustive extraction of the aspects and associated sentiments from the reviews. The final exercise of determining the key influencing factors was done by grouping these aspects under factors identified by marketing experts. A total of nine factors were identified, with the usefulness of the app being the most important factor. The findings of this study are essential for the development team and government to improve the application and increase adoption.

Using Deep Learning Techniques in Forecasting Stock Markets by Hybrid Data with Multilingual Sentiment Analysis

Electronic word-of-mouth data on social media influences stock trading and the confidence of stock markets. Thus, sentiment analysis of comments related to stock markets becomes crucial in forecasting stock markets. However, current sentiment analysis is mainly in English. Therefore, this study performs multilingual sentiment analysis by translating non-native English-speaking countries’ texts into English. This study used unstructured data from social media and structured data, including trading data and technical indicators, to forecast stock markets. Deep learning techniques and machine learning models have emerged as powerful ways of coping with forecasting problems, and parameter determination greatly influences forecasting models’ performance. This study used Long Short-Term Memory (LSTM) models employing the genetic algorithm (GA) to select parameters for predicting stock market indices and prices of company stocks by hybrid data in non-native English-speaking regions. Numerical results revealed that the developed LSTMGA model with hybrid multilingual sentiment data generates more accurate forecasting than the other machine learning models with various data types. Thus, the proposed LSTMGA model with hybrid multilingual sentiment analysis is a feasible and promising way of forecasting the stock market.

Semisupervised neural biomedical sense disambiguation approach for aspect-based sentiment analysis on social networks

Patient narratives on social networks contain large amounts of objective information, such as the descriptions of examinations and interventions. Sentiment analysis (SA) models are mostly used to evaluate the conveyed sentiments by patients in these narratives to assess positive or negative clinical outcomes or to judge the impact of a drug or a medical condition. To date, many state-of-the-art SA models often result in false assessment coverage due to the natural medical entities recognition deficiency and ambiguity problem. In this work, we propose a semisupervised-based neural sense disambiguation approach that helps to substantially define ambiguities, their levels, and the relational mappings between biomedical targets and dependencies for accurate aspect-based sentiment prediction. Three main modules are proposed: (1) generate a sentiment value based on extracted concepts and their synsets, (2) encode the representations of the contextual senses and sentiment inputs, and (3) estimate an aspect-based sentiment weight based on the context-dependency sentiment units vs. the biomedical sense. Both intrinsic and extrinsic evaluation proved how the proposed method have succeeded in pruning contextual sense feature generation and showed a strong agreement for biomedical data property parameterization and ambiguity type extraction. Thus, the model offers a significant rate of discrimination of biomedical natural concept senses by critically analysing constraints from conjunctions of positive or negative contextual semantics. A total of 21% of the vocabulary is drug names, 11% is a multiword drug reaction expressions, 7% is disease symptoms, and 5% is disease-related concepts such as symptoms and related therapy terms. Furthermore, the experiments on a multisource data from Twitter and health-related forums have overshadowed sentiment assessment and achieved an accuracy of 0.91 regarding concepts-based biomedical aspects. These results provide fresh insights into how to investigate biomedical knowledge, e.g., Medical Subject Headings (MeSH) and PubMed, to clarify the correspondence of various biomedical descriptive entities, definitions, and data properties from shared medication-related content.

A novel metaheuristic optimisation approach for text sentiment analysis

Automated sentiment analysis is considered an area in natural language processing research that seeks to understand a text author's mood, thoughts, and feelings. New opportunities and challenges have arisen in this field due to the popularity and accessibility of a variety of resources of ideas, such as online review websites, personal blogs, and social media. Feature selection, which can be conducted using metaheuristic algorithms, is one of the steps of sentiment analysis. It is crucial to use high-performing algorithms for feature selection. This paper applies the Horse herd Optimisation Algorithm (HOA) for feature selection in text sentiment analysis. HOA is a metaheuristic algorithm and uses six key behaviours to simulate the social performance of horses of various ages, to solve high-dimensional optimisation problems. In order to improve HOA, this paper adds another behaviour of horses to the basic algorithm; thus, the new algorithm uses seven key behaviours of horses of different ages to imitate their social performance. It is then discretised and converted to a multi-objective algorithm. The improved algorithm's performance is evaluated using 15 CEC benchmark functions, and the results are compared to the Binary Social Spider Algorithm, the Binary Grey Wolf Optimizer, and the Binary Butterfly Optimization Algorithm. The new algorithm, the Multi-objective Binary Horse herd Optimisation Algorithm (MBHOA), excels at solving high-dimensional complex problems. To evaluate the algorithm's performance in feature selection, as a practical example, it is employed in text sentiment analysis and examined on various data sets. The simulation results indicate that MBHOA has a better performance in analysing sentiment compared to similar approaches.

Twitter’da COVID-19 aşılarına karşı kamu duyarlılığının çoğunluk oylama sınıflandırıcısı temelli makine öğrenmesi ile duygu analizi

Dünyada milyarlarca kullanıcısı bulunan sosyal medya platformlarının yükselişiyle birlikte bilginin yayılması her zamankinden daha kolay hale gelmiştir. COVID-19 pandemisi aşılar da dâhil olmak üzere birçok konunun tartışılmasında sosyal medya kullanımını artırmıştır. Bu çalışmanın amacı, Türkiye’de, özellikle sosyal medya kullanıcılarının COVID-19 aşılarına ilişkin tutumunu ve endişelerini daha iyi anlamak adına Twitter üzerinde elde edilen aşıyla ilgili tweetlerin makine öğrenmesi ile kamu duyarlılığını analiz etmektir. Bu amaç doğrultusunda çalışma altı farklı sınıflandırma görevinde kullanılan makine öğrenmesi algoritması karşılaştırılarak en yüksek doğruluk oranına sahip Destek Vektör Makinesi, XGBoost ve Rastgele Orman ile bir kolektif öğrenme yöntemi olan çoğunluk oylama yöntemi geliştirilmiştir. Çoğunluk oylama yöntemlerinde birisi olan Yumuşak Oylama yöntemi hem Sert Oylama yaklaşımdan hem de bireysel diğer altı makine öğrenmesi yaklaşımlarından daha yüksek başarı oranı ile %90,5 başarı oranına ulaşmıştır. En yüksek doğruluk oranına sahip olan Yumuşak Oylama yöntemi ile Twitter’dan elde edilen 153 güne ait 412.588 adet günlük tweet analiz edilerek sonuçlar raporlanmıştır. Çalışmanın bulguları son derece çarpıcı olup, diğer ülkeler üzerine yapılan çalışmalardan da farklılık göstermektedir. Bu çalışma bildiğimiz kadarıyla Türkiye’de COVID-19 aşılarına yönelik duygu analizi gerçekleştiren ilk çalışma olmakla birlikte sosyal medya üzerinden duygu analizi yaklaşımıyla COVID-19 aşılarına ilişkin duyarlılığı izlemek için değerli ve kolayca uygulanan bir araç olduğunu göstermektedir.

Rule-Based Arabic Sentiment Analysis using Binary Equilibrium Optimization Algorithm.

With the development of websites and social networks, Internet users generate a massive amount of comments and information on the Web. Sentiment analysis, also called opinion mining, offers an opportunity to mine the people’s sentiments and emotions from the textual comments. In the last decade, sentiment analysis has been applied in research areas such as recommendation and support systems and has become an area of interest for many researchers. Therefore, many studies have been carried out on English, while other languages, such as Arabic, received less attention. Increasingly, sentiment analysis researchers use machine learning due to its excellent performance. However, the generated models are black boxes and non-interpretable by the users. The rule-based classification is a promising approach for generating interpretable models. This work proposes a classification rule-based Arabic sentiment analysis approach together with a new binary equilibrium optimization metaheuristic algorithm as an optimization method for classification rule generation from Arabic documents. The proposed approach has been experimented on the Opinion Corpus for Arabic (OCA) and generates a classification model of thirteen rules. The comparison results with state-of-the-art methods show that the proposed approach outperforms all other white-box models regarding classification accuracy.

Kurdish Language Sentiment Analysis: Problems and Challenges

The increasing usage of blogs, social networks, and forums for sharing opinions on a certain topic has created vast amounts of internet data. Therefore, Sentiment Analysis has gained great popularity among researchers and industry for analyzing the polarity of users' opinions. In recent years, Sentiment Analysis has been applied to various languages using machine learning-approach, corpus-based approach, and deep learning techniques since it is beneficial for creating an effective recommender system. The Kurdish Language is an Indo-European language, one of the official languages in Iraq, and it is also widely used in Turkey, Iran, and Syria. Although the importance of this Language is spoken by over 40 million people, to the best of our knowledge, no research has been done regarding the challenges and problems of Kurdish sentiment analysis. Our research aims to highlight the latest studies and examine the most critical challenges of applying sentiment analysis approaches to the Kurdish Language. The study includes determining each challenge in each step of sentiment analysis processing in the Kurdish Language. In addition, our proposed methodology that could help address most of these challenges is implementing a hybrid approach by combining machine learning and lexicon-based approaches to improve the proficiency of sentiment classification in the Kurdish Language.