Emotion Classification on Indonesian Twitter Dataset

Saputri, Mei Silviana; Mahendra, Rahmad; Adriani, Mirna

doi:10.1109/ialp.2018.8629262

Cited by 58 publications

(73 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Besides doing hierarchical multi-label classification and dataset balancing, another thing that needs to be tried to improve the accuracy of this research is to add a semantic feature, namely word embedding (Mikolov et al, 2013) in the feature extraction process. In some text classification experiments in the Indonesian language (Saputri et al, 2018;Jannati et al, 2018), adding word embedding features to basic features such as word ngrams is shown to improve classification performance because the word embedding feature can recognize word meaning that cannot be captured by features such as frequency term, orthography and lexicon features.…”

Section: Discussionmentioning

confidence: 99%

Multi-label Hate Speech and Abusive Language Detection in Indonesian Twitter

Ibrohim¹,

Budi²

2019

Proceedings of the Third Workshop on Abusive Language Online

163

122

View full text Add to dashboard Cite

Hate speech and abusive language spreading on social media need to be detected automatically to avoid conflicts between citizens. Moreover, hate speech has a target, category, and level that also need to be detected to help the authority in prioritizing which hate speech must be addressed immediately. This research discusses multi-label text classification for abusive language and hate speech detection including detecting the target, category, and level of hate speech in Indonesian Twitter using machine learning approaches with Support Vector Machine (SVM), Naive Bayes (NB), and Random Forest Decision Tree (RFDT) classifier and Binary Relevance (BR), Label Power-set (LP), and Classifier Chains (CC) as the data transformation method. We used several kinds of feature extractions which are term frequency, orthography, and lexicon features. Our experiment results show that in general the RFDT classifier using LP as the transformation method gives the best accuracy with fast computational time.

show abstract

Section: Discussionmentioning

confidence: 99%

Multi-label Hate Speech and Abusive Language Detection in Indonesian Twitter

Ibrohim¹,

Budi²

2019

Proceedings of the Third Workshop on Abusive Language Online

163

122

View full text Add to dashboard Cite

show abstract

“…Ekman's model divides emotions into six emotional labels, namely happiness, anger, fear, disgust, sadness, and surprise [31]; these labels are universal in different cultures.Currently, social media makes users tend to express emotions through text posts. One of the social media that has the highest user growth rate in Indonesia is Twitter; active Twitter users in Indonesia occupy the third position in the Asia Pacific from 2012 to 2018 [28]. Research result [8] using the Naïve Bayes method with a combination of N-gram features, but this study only produced the highest accuracy of 55.54%.…”

mentioning

confidence: 87%

Emotion Detection in Twitter Social Media Using Long Short-Term Memory (LSTM) and Fast Text

Riza¹,

Charibaldi²

2021

Int.J.Artif.Intell.Robot.

View full text Add to dashboard Cite

Emotion detection is important in various fields such as education, business, employee recruitment. In this study, emotions will be detected with text that comes from Twitter because social media makes users tend to express emotions through text posts. One of the social media that has the highest user growth rate in Indonesia is Twitter. This study will use the LSTM method because this method is proven to be better than previous studies. Word embedding fast text will also be used in this study to improve Word2Vec and GloVe that cannot handle the problem of out of vocabulary (OOV). This research produces the best accuracy for each word embedding as follows, Word2Vec produces an accuracy of 73,15%, GloVe produces an accuracy of 60,10%, fast text produces an accuracy of 73,15%. The conclusion in this study is the best accuracy was obtained by Word2Vec and fast text. The fast text has the advantage of handling the problem of out of vocabulary (OOV), but in this study, it cannot improve the accuracy of word 2vec. This study has not been able to produce very good accuracy. This is because of the data used. In future works, to get even better results, it is expected to apply other deep learning methods, such as CNN, BiLSTM, etc. It is hoped that more data will be used in future studies.

show abstract

“…Adapun pengujian pada penelitian ini menggunakan k-fold validation dengan nilai k=10, yang berarti data akan dibagi menjadi 10 bagian dan setiap 9 bagian akan dipakai sebagai data learning atau pelatihan, dan 1 bagian dipakai sebagai data testing atau pengujian (Saputri, Mahendra, & Adriani, 2019).…”

Section: Pembobotan Term Tf-idfunclassified

Klasifikasi Rating Otomatis pada Dokumen Teks Ulasan Produk Elektronik Menggunakan Metode N-gram dan Naïve Bayes

Trianto

Triyono

Arum

2020

JIUP

View full text Add to dashboard Cite

Online product ratings usually provide descriptive reviews and also reviews in the form of ratings. Likewise, what was done at the Lazada online store. Descriptive review can provide a clear view compared to a rating review to other potential buyers. However, in reality there is a mismatch between the description review and the rating given. This creates a lack of information for sellers as well as potential buyers. Automatic classification of buyer descriptive reviews is proposed in this study so that there is a match between descriptive reviews and rating reviews. This automatic classification descriptive review uses the Naive Bayes algorithm with n-gram feature extraction and TF-IDF word weighting. The results of this study obtained the best accuracy of 94.06%, a recall of 91.73% and precision of 90.71% in Bigram feature extraction. With this accuracy value it can be used as a reference or model for classifying product description reviews, so that the feedback process between sellers and buyers can run well.

show abstract

Emotion Classification on Indonesian Twitter Dataset

Cited by 58 publications

References 9 publications

Multi-label Hate Speech and Abusive Language Detection in Indonesian Twitter

Multi-label Hate Speech and Abusive Language Detection in Indonesian Twitter

Emotion Detection in Twitter Social Media Using Long Short-Term Memory (LSTM) and Fast Text

Klasifikasi Rating Otomatis pada Dokumen Teks Ulasan Produk Elektronik Menggunakan Metode N-gram dan Naïve Bayes

Contact Info

Product

Resources

About