Garuda - Garba Rujukan Digital

Wisnu Agastya

Universitas Dian Nuswantoro

Author-ID : 3040561

Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering Energy Engineering

Published : 3 Documents Claim Missing Document

Claim Missing Document

Articles

Sequential Model for Mapping Compound Emotions in Indonesian Sentences - Aripin; Wisnu Agastya; Hanny Haryanto
Journal of Applied Intelligent System Vol 5, No 1 (2020): Journal of Applied Intelligent System
Publisher : Universitas Dian Nuswantoro and IndoCEISS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33633/jais.v5i1.4264

This research proposes mapping Indonesian sentences with single and multiple structures into emotion classes based on a multi-label classification process. The result of this research can apply in various fields, including the development of facial expressions in virtual character animation. Applications in other fields are facial expression analysis, human-computer interaction systems, and other virtual facial character system applications. In previous research, the classification process used for emotion mapping was usually based only on the frequency of occurrence of adjectives. The resulting emotion classes are less representative of sentence semantics. In this research, the proposed sequential model can take into account the semantics of the sentence so that the results of the classification process are more natural and representative of the semantics of the sentence. The method used for the emotion mapping process is multi-label text classification with continuous values between 0-1. This research produces the tolerant-method that utilizes the error value to deliver accuracy in the model evaluation process. The tolerant-method converts the predicted-label, which has an error value less than or equal to the error-tolerant value, to the actual-label for better accuracy. The model used in the classification process is a sequential model, including one-dimensional Convolution Neural Networks (CNN) and bidirectional Long Short-Term Memory (LSTM). The CNN model generates feature maps of each input in a partial way. Meanwhile, bidirectional LSTM captures information from input data in two directions. Experiments were performed using test data on Indonesian sentences. Based on the experimental results, bidirectional LSTM can produce an accuracy of 91% in the 8: 2 data portion and error-tolerant of 0.09.Keywords : Sequential Model, Mapping Compound Emotions, Sentence Semantics, Indonesian Sentences

Pemetaan Emosi Dominan pada Kalimat Majemuk Bahasa Indonesia Menggunakan Multinomial Naïve Bayes Wisnu Agastya; Aripin
Jurnal Nasional Teknik Elektro dan Teknologi Informasi Vol 9 No 2: Mei 2020
Publisher : Departemen Teknik Elektro dan Teknologi Informasi, Fakultas Teknik, Universitas Gadjah Mada

This study aimed at mapping Indonesian sentences into emotion classes based on the classification process in those sentences. The results of emotion mapping can be applied in various fields, such as production of animated films and games, analysis of facial expressions, human-computer interactions, and development of other expressive virtual characters, specifically to produce facial expressions that match the spoken sentences. The method used for the emotion mapping process was the text classification using multinomial naïve Bayes model that was accompanied by dominant boundary equation. Multinomial naïve Bayes model in the text classification is used to determine the types and the emotional intensity of Indonesian sentences, whereas dominant boundary equation iss used to determine the threshold in order to identify the dominant classes. The emotion classes used as references are six basic emotion classes according to Paul Ekman, i.e., happiness, sadness, anger, fear, disgust, and surprise. The experiment on the process of mapping emotions used Indonesian single and compound sentences. The experimental results show that the text classification using multinomial naïve Bayes model accompanied by dominant boundary equation can map compound sentences into several classes of dominant emotions.

Ekstraksi Emosi Majemuk Kalimat Bahasa Indonesia Menggunakan Convolutional Neural Network Aripin; Wisnu Agastya; Hanny Haryanto
Jurnal Nasional Teknik Elektro dan Teknologi Informasi Vol 10 No 2: Mei 2021
Publisher : Departemen Teknik Elektro dan Teknologi Informasi, Fakultas Teknik, Universitas Gadjah Mada

Facial expressions can strengthen the information conveyed in interactive communication. In the field of developing virtual characters specifically for facial characters, facial expressions are needed to animate a facial virtual character to make it look natural like a human. One type of emotional expression is a compound emotional expression, which is a combination of two or more basic emotions. For example, the expression of disappointed emotions is a combination of anger and sadness. Facial expressions can appear due to emotional stimulation, one of which is the meaning of the sentence. This research aims to extract emotional data from Indonesian sentences using the multi-label classification process of the CNN model so as to produce compound facial expressions that are applied in virtual character animation. The basic emotion classes used in the classification process are anger, disgust, fear, happiness, sadness, and surprise. Based on the experimental results, the CNN model can produce an accuracy of 94.5% with the composition of training data and test data is 8: 2. The classification process result shows that each sentence can produce more than one basic emotion class that forms compound expressions. The results of the visualization of compound expressions for each sentence can represent compound expressions.

Title

Found 3 Documents
Search

Abstract

Abstract

Abstract

Title Search

Found 3 Documents Search

Abstract

Abstract

Abstract

Title

Found 3 Documents
Search