Performance comparison of TF-IDF and Word2Vec models for emotion text classification

Cahyani, Denis Eka; Patasik, Irene

doi:10.11591/eei.v10i5.3157

Cited by 61 publications

(24 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The more often a word appears in many documents, the smaller the IDF value will be. The following is the formula for TF-IDF [19]:…”

Section: Feature Extraction Using Tf-idf Vectorizermentioning

confidence: 99%

Identify User Behavior based on Tweet Type on Twitter Platform using Agglomerative Hierarchical Clustering

Weninggalih

Sibaroni²

2022

mib

View full text Add to dashboard Cite

Information dissemination can occur through any media, including social media. One of the social media that has become a forum for disseminating information is Twitter. Through user-uploaded tweets, not a few comments are positive (praise/support) or negative (blasphemy), depending on the tweet. This study chooses politics as a discussion. Data crawling was carried out to obtain a dataset and raise the topic of Joko Widodo as a President of Indonesia, whose work is considered poor by the public, so they want him to resign immediately. This makes it interesting because we can identify user behavior from tweets about the topic. The choice of this topic was based on a lot of users who discussed it, so it was trending on Twitter. Preprocessing stage aims to eliminate missing values. After that, it then goes through the feature extraction process. The agglomerative Hierarchical Clustering Algorithm of the clustering method is applied in this research. This algorithm can directly set how many clusters to facilitate the clustering process. The result obtained 3 clusters with different user behavior. Negative user behavior is found in cluster 1, while positive user behavior is found in cluster 2.

show abstract

“…The more often a word appears in many documents, the smaller the IDF value will be. The following is the formula for TF-IDF [19]:…”

Section: Feature Extraction Using Tf-idf Vectorizermentioning

confidence: 99%

Identify User Behavior based on Tweet Type on Twitter Platform using Agglomerative Hierarchical Clustering

Weninggalih

Sibaroni²

2022

mib

View full text Add to dashboard Cite

show abstract

“…The greater the term in the document, the greater the weight value. The term frequency (TF) formula can be defined by [21]:…”

Section: Tf-idf Vectorizermentioning

confidence: 99%

Identify User Behavior based on Tweet Type on twitter Platform using Mean Shift Clustering

Fikriyah

Sibaroni

2022

mib

View full text Add to dashboard Cite

Twitter is a social media where users often get information from various fields. There are many problems with Twitter. For example, in Indonesia's political field, discussing the performance of the President of Indonesia and his staff who are not good, students and the public hold demonstrations in DKI Jakarta. They want the President of Indonesia to step down from office. When the problem is trending, some users have positive (praise) and negative (blasphemous) behavior, which is interesting to discuss in this study. Before the method stage, data preprocessing is carried out so that the data to be used becomes more efficient. Word weighting is also done using the TF-IDF Vectorizer. Then, the clustering method with the Mean Shift algorithm is applied to identify user behavior based on the type of tweet. This method can find information from a vast data set in a short time. Based on this algorithm, the results obtained are 67 clusters from the Mean Shift algorithm. From a total of 67 clusters obtained, 5 clusters were taken to identify user behavior. User behavior in clusters 0, 2, 3, and 4 is negative because it discusses the people who want the President of the Republic of Indonesia to resign from his position immediately. Meanwhile, user behavior in cluster 1 is positive because the topics discussed only information that the people of Lampung are already in Jakarta.

show abstract

“…After preprocessing, the data is ready to be processed at the next stage, weighting with TF-IDF. Term Frequency-Inverse Document Frequency, commonly known as TF-IDF, is a method of determining the weight of a word by giving different weights to each word in a document based on the frequency of words per document and the frequency of words in all documents [13]. The first step in this process is to calculate the frequency of appearance of a word in a document (TF) with the equation (1).…”

Section: Term Frequency-inverse Document Frequency (Tf-idf) Word Weig...mentioning

confidence: 99%

“…Next, the calculation of the number of documents containing a certain word is carried out, and then calculated its inverse (IDF) [13] with equations (2).…”

Section: Term Frequency-inverse Document Frequency (Tf-idf) Word Weig...mentioning

confidence: 99%

“…Note: idf t : inverse document frequency D : the number of all existing documents df t : number of documents containing the word t The last step in this process is to calculate the TF-IDF value by multiplying the TF result by the IDF calculation result [13] with equations (3). : number of occurrences of the word t idf t : inverse document frequency containing the word t…”

Section: Term Frequency-inverse Document Frequency (Tf-idf) Word Weig...mentioning

confidence: 99%

See 1 more Smart Citation

Sentiment and Discussion Topic Analysis on Social Media Group using Support Vector Machine

Adityani

Richasdy

Astuti

2022

mib

View full text Add to dashboard Cite

The growth of social media in this modern era is increasingly rapid, where people are very active digitally interacting with each other. People who have a common interest or simply like to be in a community often gather in an online group, especially on Facebook. Alumni of Telkom University are no exception, who are also actively discussing and sharing information in Telkom University Alumni Forum Facebook group (FAST). By using their status from that group, sentiment and topic discussion analysis can be performed to determine whether the polarity is positive, neutral, or negative. In Addition, topic modeling extracts what topics are often discussed in the group. In this research, sentiment analysis was performed using the Support Vector Machine (SVM) method. Also, the classification process involved TF-IDF for word weighting and confusion matrix as performance measurement. Several testing scenarios were carried out to get the best accuracy value. Based on the tests performed on the preprocessing technique and feature extraction n-gram addition, the highest accuracy value obtained is 80.56%. The result indicates that the best performance is obtained by combining preprocessing techniques without the stopword removal process and feature extraction unigram. Moreover, the topics discussed based on topic modeling results were related to telecommunication and Telkom, Indonesia, alumni, and FAST.

show abstract

Performance comparison of TF-IDF and Word2Vec models for emotion text classification

Cited by 61 publications

References 22 publications

Identify User Behavior based on Tweet Type on Twitter Platform using Agglomerative Hierarchical Clustering

Identify User Behavior based on Tweet Type on Twitter Platform using Agglomerative Hierarchical Clustering

Identify User Behavior based on Tweet Type on twitter Platform using Mean Shift Clustering

Sentiment and Discussion Topic Analysis on Social Media Group using Support Vector Machine

Contact Info

Product

Resources

About