Sentiment Analysis of the Kampus Merdeka Program on Twitter Using Support Vector Machine and a Feature Extraction Comparison: TF-IDF vs. FastText

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.

Sentiment Analysis is a widely used text classification technique. It breaks down any given text or comments and classify the text either as positive or negative based on the views conveyed in it. Previous works done on sentiment classification used either lexicon based approach or machine learning techniques. Likewise, the major drawback of the existing systems was the focus on only binary classification of review such as positive or negative. Ignorance of the neutral review will result in misinterpretation of a customer's opinion about a product or movie, which will degrade the business or trend. In case of using only lexicon based approach, the system highly depends on the selection of lexicon resource and dictionary. In case of system built only using machine learning approach, the performance of the system depends on the algorithms chosen. This work presents a hybrid model to resolve the neutral class too. The proposed work combines a lexical approach (SentiWordNet) with the machine learning algorithms such as Support Vector Machine, Decision Tree, Logistic Regression and Naive Bayes for sentiment analysis to resolve the neutral opinions beyond the binary categorization of the customer's review. We have also compared the performance of these four machine learning algorithms along with the lexicon approach. The results proved that Support Vector Machine and Logistic Regression algorithms outperform the other two algorithms with an accuracy of about 80% which is on average differs by 6% to 10% when compared to other algorithms.

Sentiment analysis is the field of study that focuses on finding effectively the conduct of subjective text by analyzing people’s opinions, sentiments, evaluations, attitudes and emotions towards entities. The analysis of data and extracting the opinion word from the data is a challenging task especially when it involves reviews from completely different domains. We perform cross domain sentiment analysis on Amazon product reviews (books, dvd, kitchen appliances, electronics) and TripAdvisor hotel reviews, effectively classify the reviews to positive and negative polarities by applying various preprocessing techniques like Tokenization, POS Tagging, Lemmatization and Stemming which can enhance the performance of sentiment analysis in terms of accuracy and time to train the classifier. Various methods proposed for document-level sentiment classification like Naive Bayes, k-Nearest Neighbor, Support Vector Machines and Decision Tree are analysed in this work. Cross domain sentiment classification is useful because many times we might not have training corpus of specific domains for which we need to classify the data and also cross domain is favoured by lower computation cost and time. Despite poor performance in accuracy, the time consumed for sentiment classification when multiple testing datasets of different domains are present is far less in case of cross domain as compared to single domain. This work aims to define methods to overcome the problem of lower accuracy in cross-domain sentiment classification using different techniques and taking the benefit of being a faster method.

The principal aim of e-commerce systems is to improve the consumer experience, with customer input being essential to this objective. This study introduces an innovative method for analysing tweet data to enhance customer happiness. Key tweet features were retrieved to capture the sentiment expressed in customer tweets by utilising a combination of n-gram models and word embeddings. The features were subsequently employed to construct classification models utilising Support Vector Machines (SVM) and Artificial Neural Networks (ANN), both of which are frequently applied in sentiment analysis. A model based on Convolutional Neural Networks (CNN) was proposed to enhance the accuracy of classifying sentiment by categorizing tweets as either positive or negative. The study's results indicated that the CNN model surpassed both the SVM and ANN models, exhibiting more accuracy in sentiment classification of tweets. In addition to its precision, the CNN model offered significant insights into the interrelations among different tweet categories, enhancing comprehension of client mood and views. This extensive study underscores the need of understanding the complexities of customer feedback to improve e-commerce strategies and consumer delight as well as the efficiency of the CNN model for sentiment categorisation. By means of exact classification and analysis of sentiment in real-time twitter data, e-commerce systems may more successfully evaluate customer impressions and react dynamically to enhance their whole experience.

Sentiment analysis has become increasingly important in understanding user perceptions of digital platforms. This study focuses on analyzing TikTok application reviews from the Google Play Store in Indonesia using machine learning techniques. The research aims to investigate sentiment distribution and compare the performance of three popular machine learning models: Random Forest, Support Vector Machine (SVM), and Naive Bayes. The study employed a comprehensive methodology involving data collection, preprocessing, feature extraction, and model evaluation. A dataset of 10,000 TikTok reviews was collected and preprocessed using techniques such as case folding, tokenization, and stopword removal. The sentiment labeling process categorizes reviews into positive, negative, and neutral sentiments based on user ratings. The TF-IDF algorithm was used for feature extraction, and the SMOTE technique addressed class imbalance. Results revealed a predominance of negative sentiment (53.5%), followed by neutral (32.1%) and positive (14.4%) sentiment. Model performance comparisons at different data sharing ratios (80/20 and 70/30) demonstrated that Random Forest and SVM consistently outperformed Naive Bayes. At the 80/20 ratio, Random Forest achieved the highest accuracy of 83.73%, highlighting its effectiveness in sentiment classification. The research contributes to the field of sentiment analysis and natural language processing by providing insights into user experiences with the TikTok application in Indonesia. The findings can guide application developers in understanding user perceptions and improving user experience.

Text classification which is an integral part of text mining has caught much attention in various industries and fields recently. The ability is in assigning text documents to one or more pre-defined categories based on content similarity. While most of application of text classification focuses on document level, question classification works at much granular level such as sentence and phrase. There have been numerous studies on question classification in accordance to Bloom taxonomy in assessments to measure cognitive level of learners in higher learning institutions. But it has not been effective yet to resolve overlapping issue of Bloom taxonomy verb keywords being assigned to more than one category of Bloom taxonomy. The presence of this poses a problem in respect of classifying a particular question into a right category of Bloom taxonomy. And feature extraction plays an important role in improving the accuracy of classifier such as Support Vector Machine in question classification. Much of the past related research work focus on feature extraction methods such as bag of word (BOW) and syntactic analysis to classify questions and to address the issue, an improvement in feature extraction is needed. In view of this, this study proposes an integrated approach in feature extraction involving semantic aspect in classifying questions in accordance to Bloom taxonomy. Support Vector Machine classifier is used as it is well known for its high accuracy in text classification. With all this in place, an improved accuracy in classifying questions in accordance to Bloom taxonomy can be expected.

With the rapid increase of the number of Internet users and the amount of online comment data, a large number of referable information samples are provided for data mining technology. As a technical application of data mining, text sentiment classification can be widely used in public opinion management, marketing, and other fields. In this study, a combination approach to SVM (support vector machine) and IPSO (improved particle swarm optimization) is proposed to classify sentiment by using text data. First, the text data of 30,000 goods reviews and corresponding ratings are collected through the web crawler. Then, TFIDF (term frequency-inverse document frequency) and Word2vec are used to vectorize the goods review text data. Next, the proposed classification model is trained by the SVM, and the initial parameters of the SVM are optimized by the IPSO. Finally, we applied the trained SVM-IPSO model to the test set and evaluated the performance by several measures. Our experiment results indicate that the proposed model performed the best for text data sentiment classification. Additionally, the traditional machine learning model SVM becomes very effective after parameter optimization, which demonstrates that the parameters’ optimization by IPSO has successfully improved the classification accuracy. Furthermore, our proposed model SVM-IPSO significantly outperforms other benchmark models, indicating that it could be applied to improve the accuracy and efficiency for text data sentiment classification.

The rapid development of information technology has changed the way people interact and express their opinions on public policies, including the People's Housing Savings (Tapera) policy in Indonesia. People now primarily express their views openly on social media platforms like Twitter, generating a substantial amount of text data for analysis to understand public sentiment. However, the main challenge in this sentiment analysis is determining the most effective machine learning algorithm for classifying public opinion with high accuracy. This study aims to compare the performance of three machine learning algorithms, namely Naïve Bayes, Support Vector Machine, and Random Forest, in analyzing public sentiment towards the Tapera policy. This study analyzes public comment data obtained from Twitter. We measure the accuracy of each algorithm to determine its optimal performance in sentiment classification. The research method consists of several stages, starting with data collection, text preprocessing to clean and prepare data, and then applying the three algorithms to analyze sentiment. The results showed that Naïve Bayes had the highest accuracy of 69.17%, followed by Support Vector Machine with an accuracy of 68.42%, and Random Forest with an accuracy of 66.17%. This shows that Naïve Bayes is the most effective algorithm to use in sentiment analysis of public comments related to the Tapera policy, especially in the context of complex text data from social media. The conclusion of this study is that Naïve Bayes is superior in classifying public sentiment towards the Tapera policy compared to Support Vector Machine and Random Forest. As a result, this study makes a significant contribution to selecting the most appropriate machine learning algorithm for public sentiment analysis towards public policy, which in turn can help the government understand and respond to public perceptions more effectively.

Abstract—This paper presents a hybrid sentiment analysis system that integrates lexicon-based, traditional machine learning, and deep learning techniques to classify textual data into positive, negative, or neutral sentiments. The system leverages Python-based libraries such as Scikit-learn, NLTK, and Transformers to pre- process text, extract features, and apply models in- cluding Support Vector Machine (SVM), Bidirectional Encoder Representations from Transformers (BERT), Linear Regression, and Valence Aware Dictionary and sEntiment Reasoner (VADER). The framework pro- cesses customer reviews from e-commerce platforms (Amazon, Flipkart) and social media (Instagram) using web scraping and provides actionable insights through sentiment summarization and visualizations (bar and pie charts). Experimental results demonstrate BERT’s superior performance with 92.3% accuracy, followed by SVM (85.6%), Linear Regression (81.2%), and VADER (76.8%). The system addresses challenges like sarcasm, class imbalance, and scalability, offering a scalable, user-friendly solution for real-world applications in e- commerce, social media analytics, and brand reputation management. Index Terms—Sentiment Analysis, Machine Learning, Deep Learning, Natural Language Processing, BERT, SVM, VADER, Text Classification, Data Visualization.

Background and Objectives: Twitter Sentiment Classification is one of the most popular fields in information retrieval and text mining. Millions of people of the world intensity use social networks like Twitter. It supports users to publish tweets to tell what they are thinking about topics. There are numerous web sites built on the Internet presenting Twitter. The user can enter a sentiment target and seek for tweets containing positive, negative, or neutral opinions. This is remarkable for consumers to investigate the products before purchase automatically. Methods: This paper suggests a model for sentiment classification. The goal of this model is to investigate what is the role of n-grams and sampling techniques in Sentiment Classification application using an ensemble method on Twitter datasets. Also, it examines both binary and multiple classifications, which are classified datasets into positive, negative, or neutral classes. Results: Twitter Classification is an outstanding problem, which has very few free resources and not available due to modified authorization status. However, all Twitter datasets are not labeled and free, except for our applied dataset. We reveal that the combination of ensemble methods, sampling techniques, and n-grams can improve the accuracy of Twitter Sentiment Classification. Conclusion: The results confirmed the superiority of the proposed model over state-of-the-art systems. The highest results obtained in terms of accuracy, precision, recall, and f-measure.======================================================================================================Copyrights©2020 The author(s). This is an open access article distributed under the terms of the Creative Commons Attribution (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, as long as the original authors and source are cited. No permission is required from the authors or the publishers.======================================================================================================

The Indonesian government's implementation of a booster vaccination program as part of its COVID-19 response has generated diverse public reactions, particularly on social media platforms like Twitter. This study aims to analyze public sentiment regarding booster vaccines by examining Twitter data to understand the prevailing discourse and attitudes toward this policy. The research employs sentiment analysis, a text mining and processing technique, to classify tweets into positive, neutral, and negative categories. The study utilizes the Support Vector Machine (SVM) algorithm, evaluating its performance through a multi-class parameter assessment. Two multi-class strategies, One-against-one (OAO) and One-against-rest (OAR) are combined with various kernels (Sigmoid, Polynomial, and RBF) to identify the most accurate model for sentiment classification. The results show that the OAO method with the RBF kernel achieves the highest accuracy of 96%, outperforming other combinations like OAO with Polynomial (95.2%) and Sigmoid (93.7%) kernels. Similarly, the RBF kernel performs best with 95.5% accuracy in the OAR approach. Using the optimal model, sentiment analysis classifies 49 tweets as positive, 927 as neutral, and 24 as negative, revealing a predominantly neutral public sentiment with limited positive and negative opinions. In conclusion, this study demonstrates the effectiveness of SVM, particularly the OAO method with the RBF kernel, for sentiment analysis of social media data. The findings provide insights into public perceptions of the booster vaccine program, offering policymakers a data-driven basis for designing targeted communication strategies to address concerns and enhance public acceptance.

Support Vector Machines (SVM), one of the new techniques for text classification, have been widely used in many application areas. SVM try to find an optimal hyperplane within the input space so as to correctly classify the binary classification problem. We present a novel heuristic text classification approach based on genetic algorithm (GA) and SVM. Simulation results demonstrate that GA and SVM are integrated effectively, and have good classification accuracy.

This project explores visitor experiences by conducting sentiment analysis and clustering analysis on 1,973 visitor reviews posted on Google Maps and TripAdvisor. Utilizing machine learning algorithms like Support Vector Machine (SVM) and the K-Means algorithm, the project achieved 87.66% accuracy in sentiment classification and identified three optimal clusters. Then, a detailed analysis of both results were conducted. The insights are visualized through an interactive dashboard, enabling museum management to make informed decisions. Despite data limitations, the project offers strategic recommendations for ongoing improvement, aiming to enhance visitor satisfaction and engagement at the Terengganu State Museum.

This study explores the dynamics of public sentiment toward SpaceX, focusing on how it fluctuates in response to key events, including successful missions and technical setbacks. Using sentiment analysis on a dataset of SpaceX-related tweets, this research captures the emotional reactions of the public, classifying them into positive, neutral, and negative categories. The analysis reveals distinct patterns: positive sentiment predominates during major achievements, such as rocket launches and new technological advancements, while negative sentiment spikes following failures or delays. The results demonstrate how public perception of SpaceX is intricately tied to the company’s performance, reflecting both excitement for its successes and frustration for its setbacks. By examining these sentiment trends, this research offers insights into how companies in the space exploration sector can manage their public relations efforts and strategically engage with audiences on social media platforms. The study employs the TextBlob sentiment analysis tool, which classifies tweet polarity and subjectivity, to categorize public sentiment in a straightforward yet effective manner. Through time-series visualizations, the study tracks how sentiment evolves over time, highlighting key fluctuations tied to SpaceX’s milestones. Additionally, the research integrates visualizations like word clouds and bar charts to identify frequent keywords associated with both positive and negative sentiments, providing a deeper understanding of public discourse surrounding the company. The study underscores the role of Twitter as a significant tool for shaping public perception, particularly in high-visibility industries like space exploration, where real-time feedback can influence both public opinion and corporate strategies. This research contributes to the broader field of sentiment analysis by focusing on the tech and space industries, where public sentiment plays a pivotal role in shaping business success and technological innovation. By examining SpaceX’s public image through sentiment trends, this study highlights the importance of real-time sentiment monitoring in shaping company strategies. Future studies could extend this analysis to include other companies in the space sector or incorporate more sophisticated machine learning models for deeper sentiment classification.

This study aims to demonstrate the methods for detecting negations in a sentence by uniquely evaluating the lexical structure of the text via word sense disambiguation. Additionally, the proposed method examined all the unique features of the related expressions within a text to resolve the contextual usage of the sentence and the effect of negation on sentiment analysis. The application of popular expression detectors skips this important step, thereby neglecting the root words caught in the web of negation, and making text classification difficult for machine learning and sentiment analysis. This study adopts the Natural Language Processing (NLP) approach to discover and antonimize words that were negated for better accuracy in text classification. This method acts as a lens that reads through a given word sequence using a knowledge base provided by an NLP library called WordHoard in order to detect negation signals. Early results show that our initial analysis improved traditional sentiment analysis that sometimes neglects word negations or assigns an inverse polarity score. The SentiWordNet analyzer was improved by 35%, the Vader analyzer by 20% and the TextBlob analyzer by 6%.