Article Per Year (5 Year)
p-Index From 2021 - 2026
1.235
P-Index
This Author published in this journals
All Journal IJCCS (Indonesian Journal of Computing and Cybernetics Systems) Prosiding KOMMIT Proceedings of KNASTIK Publikasi Eksternal VISIKES Jurnal technoscientia Jurnal Teknologi Jurnal Linguistik Komputasional Contemporary Mathematics and Applications (ConMathA) JURNAL TEKNOLOGI TECHNOSCIENTIA The Indonesian Journal of Computer Science
Edi Winarko
Jurusan Ilmu Komputer dan Elektronika, FMIPA UGM
Chemical Engineering, Chemistry & Bioengineering Computer Science & IT Control & Systems Engineering Earth & Planetary Sciences Education Electrical & Electronics Engineering Engineering Industrial & Manufacturing Engineering Materials Science & Nanotechnology Mathematics Mechanical Engineering

Published : 14 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 14 Documents
Search

 1   2 
Voice-Based Emotion Identification Based on Mel Frequency Cepstral Coefficient Feature Extraction Using Self-Organized Maps and Radial Basis Function Nikmah, Asrivatun; Damayanti, Auli; Winarko, Edi
Contemporary Mathematics and Applications (ConMathA) Vol. 7 No. 1 (2025)
Publisher : Universitas Airlangga

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20473/conmatha.v7i1.68246

Abstract

Speech recognition is one of the most popular research fields, one of which is about emotion identification. Voice-based emotion identification is carried out to determine the pattern of emotions using the depth analysis mechanism of voice signal development and feature extraction that carries the emotional characteristic parameters of the speaker's voice. Furthermore, the emotional characteristics of the speaker's voice are classified using an artificial neural network method to recognize patterns. In this study, emotion identification from voice signal data is classified into angry, sad, happy, and neutral emotions. The stages of voice-based emotion identification, including the feature extraction stage using the mel frequency cepstral coefficient, produce coefficient values, which will be used in the identification stage using the Self Organized Maps method on the Radial Basis Function.
Multi-Domain Sentiment Analysis on Ibu Kota Nusantara (IKN) Tweets Using CNN-LSTM Prasastio, Fahmi Reza; Winarko, Edi
IJCCS (Indonesian Journal of Computing and Cybernetics Systems) Vol 19, No 2 (2025): April
Publisher : IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia.

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22146/ijccs.104925

Abstract

The construction of Ibu Kota Nusantara (IKN) is a national project aimed at relocating Indonesia’s capital from Jakarta to East Kalimantan. This project has sparked various public opinions, which are widely expressed through social media platforms such as Twitter. Sentiment analysis of these opinions is crucial for understanding public perception of the IKN project. However, previous sentiment analysis studies have often overlooked domain variations in the analyzed data, such as economy, environment, and politics, each of which has distinct linguistic characteristics. This study aims to develop a multi-domain sentiment analysis model by comparing three main methods: CNN-LSTM, CNN, and LSTM. The multi-domain model is designed to address the differences in characteristics across domains and enhance the model’s ability to capture more complex sentiment patterns. The results indicate that multi-domain models outperform single-domain models, as they improve classification performance by leveraging information from multiple domains. CNN-LSTM proved to be the best model, achieving the most balanced Accuracy and F1-Score across various scenarios. The use of Keyword Embedding also significantly enhances model performance, particularly benefiting LSTM, which initially had the lowest performance.
DEVELOPMENT OF CHATBOT FOR PRE-DIAGNOSIS AND RECOMMENDATION OF ANXIETY DISORDER USING DIET AND SENTENCE TRANSFORMER MODELS Winarko, Edi; Suryanti, Angel Berta Desi
IJCCS (Indonesian Journal of Computing and Cybernetics Systems) Vol 19, No 3 (2025): July
Publisher : IndoCEISS in colaboration with Universitas Gadjah Mada, Indonesia.

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22146/ijccs.95900

Abstract

 Previous research on chatbots for pre-diagnosis and recommendation of anxiety disorders has been limited to therapy aids.  Comparing NLU DIET and LogisticRegressionClassifier models, this chatbot system calculates anxiety levels using GAD-7, DASS, and STAIT/STAIS-5 methods along with Sentence Transformer (SBERT) for semantic similarity.Intent classification testing yielded 95% accuracy for NLU DIETClassifier and 99% for LogisticRegressionClassifier. The Dialog Model achieved 68% accuracy with TEDPolicy. Testing involved 35 randomly selected respondents, including students and workers. From their interactions, the SBERT recommendation model scored 30% MAP, 26% for the Indobert base and paraphrase-multilingual-MiniLM-L12-v2 models.The average satisfaction and performance rating for the chatbot system was 3.7 out of 5. This research addresses the need for a prototype chatbot for pre-diagnosis and recommendation of anxiety disorders, with the best NLU model being LogisticRegressionClassifier at 99% accuracy and the dialog model at 68%. However, the recommendation system still has a low MAP due to the use of non-valid clinical data as references, suggesting room for improvement in future research.
Long-Context Transformer Models for Meeting Summarization: A Comparative Study of Full Fine-Tuning and Parameter-Efficient Tuning Winarko, Edi; Katarina Keishanti Joanne Kartakusuma
The Indonesian Journal of Computer Science Vol. 14 No. 6 (2025): The Indonesian Journal of Computer Science
Publisher : AI Society & STMIK Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33022/ijcs.v14i6.5054

Abstract

The growing volume of virtual meetings has increased the need for effective long-document summarization systems that capture essential discussion points from lengthy transcripts. However, existing transformer-based models often struggle to handle long-context inputs and require substantial computational resources for fine-tuning. Moreover, prior work provides limited comparative analysis of full fine-tuning and parameter-efficient fine-tuning (PEFT) specifically for meeting summarization tasks. This study systematically evaluates three long-sequence Transformer architectures—LongT5, BigBird, and LED—on the MeetingBank dataset using both full fine-tuning and PEFT strategies. Models are assessed through ROUGE scores, BERTScore, parameter efficiency, and qualitative error analysis. Experimental results show that LongT5 with full fine-tuning achieves the best performance (ROUGE-1 = 0.675, BERTScore F1 = 0.921), outperforming BigBird as the next-best model by 31.6% in ROUGE-1. PEFT reduces trainable parameters by over 90% and remains competitive only for LongT5 (ROUGE-1 = 0.543, BERTScore F1 = 0.872), while BigBird and LED experience severe degradation, producing semantically weak and incoherent summaries despite low validation loss. These findings highlight that PEFT effectiveness is highly model-dependent and that validation loss alone is an unreliable indicator of generative quality. The study contributes a comprehensive benchmarking analysis and practical insights into optimizing long-document meeting summarization under computational constraints.
Co-Authors Aaqilah Salsabila, Zaneta Adi Nugroho Adi Nugroho Ahmad Subhan Yazid Armadyah Amborowati Ashari Sutrisno Auli Damayanti Edhy Sutanta (Jurusan Teknik Informatika IST AKPRIND Yogyakarta) Herny Februariyanti Katarina Keishanti Joanne Kartakusuma Khabib Mustofa Kusumawati, Mifaidah Lilis Sulistyorini Muhammad Davi Naura, Nuha Nikmah, Asrivatun Prasastio, Fahmi Reza R Azizah Retantyo Wardoyo Santi Martini Suryanti, Angel Berta Desi Suwanto Raharjo Wijang Widhiarso