Emotional Text Classification Using TF-IDF (Term Frequency-Inverse Document Frequency) And LSTM (Long Short-Term Memory)

Alfarizi, Muhammad Ibnu; Syafaah, Lailis; Lestandy, Merinda

doi:10.30595/juita.v10i2.13262

Cited by 11 publications

(8 citation statements)

References 13 publications

(23 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The more datasets used, the better the accuracy of the data obtained. In addition, TF-IDF (Term Frequency-Inverse Document Frequency) must also be considered in word weighting (Alfarizi et al, 2022). In developing technology in the digital era, like today, chatbot services are needed to maximize the provision of information to students.…”

Section: Discussionmentioning

confidence: 99%

Increasing the Effectiveness of Higher Education Academic Services Through the Implementation of the Chatbot Platform Using the SVM Machine Learning Algorithm

Setiyani

2023

JP2

View full text Add to dashboard Cite

The education sector has adopted technology and digitization. To create appropriate technology that can increase the effectiveness and efficiency of existing processes in education, especially higher education, innovations are needed that can provide value to tertiary institutions. The right support is needed to achieve value in academic services as the heart of higher education. However, many universities still have not been able to provide maximum service. This research aims to create a chatbot model to support effective academic services for tertiary institutions. This research belongs to the type of research design science research (DSR). The research procedures were carried out by collecting data, categorizing data, creating chatbot models, model evaluation, and model implementation. Data was collected by inviting resource persons through focus group discussions (FGD), with the criteria being prospective students, university students, and the public interested in academic services in tertiary institutions. The resource persons were asked questions about the academic services needed. Based on the data obtained, there were 257 questions related to academic services. Service categorization is the process of classifying questions based on the functions of divisions or departments in tertiary institutions. Based on the data collection and service categorization results, a chatbot model is created, followed by model evaluation and implementation. The research analysis results show that the academic management chatbot model that uses the SVM algorithm can classify questions asked through chatbots with an accuracy of 57%, performing to support higher education academic services.

show abstract

Section: Discussionmentioning

confidence: 99%

Increasing the Effectiveness of Higher Education Academic Services Through the Implementation of the Chatbot Platform Using the SVM Machine Learning Algorithm

Setiyani

2023

JP2

View full text Add to dashboard Cite

show abstract

“…To determine the relationship between words or phrases and documents, can use the Term Frequency-Inverse Document Frequency (TF-IDF) method [32]. TF-IDF is a method that measures the value or weight for each word (token) contained in a document in a particular document collection.…”

Section: Term Frequency -Inverse Document Frequency (Tf-idf)mentioning

confidence: 99%

Film Recommendation System Using Content-Based Filtering and the Convolutional Neural Network (CNN) Classification Methods

Nilla,

Setiawan

2024

J. Ilm. Tek. Elektro Komput. Dan Inform

View full text Add to dashboard Cite

Managing large amounts of data is a challenge faced by users, so a recommendation system is needed as an information filter to provide relevant item suggestions. Twitter is often used to find information about movie reviews that can be used a basis for developing recommendation systems. This research contributes to applying content-based filtering in the context of Convolutional Neural Network (CNN). To the best of the researcher's knowledge, there has been no research addressing this combination of method and classification. The main focus is to evaluate the development of a recommendation system by integrating and comparing similarity identification methods using the RoBERTa and TF-IDF approaches. In this research, Roberta and TF-IDF as vectorizer and classification methods are applied to form a model that can recognize patterns in data and produce accurate predictions based on its features. The total data used is 854 movies and 34086 film reviews from 44 Twitter accounts. The SMOTE method was applied as a technique to overcome data imbalance. The research was conducted three times with increasing accuracy results. The first experiment TF-IDF as baseline, SMOTE on CNN classification. The second experiment, applying baseline, SMOTE, embedding on CNN classification. The third experiment applied baseline, SMOTE, embedding, and optimizer to CNN classification. The experimental results show that TF-IDF as baseline, SMOTE, embedding and SGD optimizer with the best learning rate on CNN classification can provide optimal results with an accuracy rate of 86.41%. Thus, the system can provide relevant movie recommendations with good prediction accuracy and performance.

show abstract

“…This indicates a decline in the level of public trust in e-commerce services. This decline can be caused by a variety of factors, including changes in service policies, a decline in service levels, and a lack of effective service promotion [35] [36].…”

Section: A Dataset Tweet Collectionmentioning

confidence: 99%

Improving Sentiment Analysis in Digital Marketplaces through SVM Kernel Fine-Tuning

Abdul Fadlil,

Riadi,

Andrianto

2024

IJCDS

View full text Add to dashboard Cite

The rapid growth of the online market, particularly in the digital realm, has spurred the need for in-depth studies regarding marketing strategies through public opinion, especially on platforms like Twitter. The sentiments expressed in customer tweets hold significant insights into their satisfaction or dissatisfaction levels with a service. Therefore, the use of ML algorithms in sentiment analysis is imperative to detect whether such comments lean towards positivity or negativity regarding a service. This research focuses on sentiment analysis towards three major e-commerce platforms in Indonesia: Tokopedia, Shopee, and Lazada, through the utilization of Twitter. The classification process involves various stages, including preprocessing, feature extraction and selection, data splitting for classification, and evaluation. The selection of both linear and non-linear SVM models as the focus of this research is based on their ability to handle large and complex datasets. The linear kernel is chosen for its proficiency in cases with a linear relationship between features and class labels, while the non-linear SVM provides flexibility in dealing with complex and non-linear relationships. Based on the evaluation results of the SVM model on the dataset, it is found that the polynomial kernel provides the highest accuracy value of 93%, with a training data share of 85%. This model features strong prediction capabilities with a precision of 93% for negative and 93% for positive labels. Although the linear kernel and other kernels showed solid performance, the polynomial kernel provided the most optimal results in the context of online marketplace sentiment analysis using data from Twitter

show abstract

Emotional Text Classification Using TF-IDF (Term Frequency-Inverse Document Frequency) And LSTM (Long Short-Term Memory)

Cited by 11 publications

References 13 publications

Increasing the Effectiveness of Higher Education Academic Services Through the Implementation of the Chatbot Platform Using the SVM Machine Learning Algorithm

Increasing the Effectiveness of Higher Education Academic Services Through the Implementation of the Chatbot Platform Using the SVM Machine Learning Algorithm

Film Recommendation System Using Content-Based Filtering and the Convolutional Neural Network (CNN) Classification Methods

Improving Sentiment Analysis in Digital Marketplaces through SVM Kernel Fine-Tuning

Contact Info

Product

Resources

About