publications

2025

1. 

   Retrieval-Augmented Generation for Indonesian Criminal Law Information Using the LLaMA Model

   A Z Pratama, A Marinta, B Triyudanto, and 2 more authors

   Innovative Informatics and Artificial Intelligence Research, 2025

   Abs

   Law is a set of rules that regulate boundaries of behavior within society, is coercive in nature, and imposes sanctions on violators. One of its branches is criminal law, which focuses on violations of norms with the threat of sanctions. However, the legal system often provides excessive discretion to the judiciary, making legal outcomes difficult to predict and potentially disadvantageous, especially for individuals who lack legal understanding or cannot afford legal representation. This disparity poses a significant challenge to legal systems in many countries. In this study, we demonstrate that artificial intelligence (AI) based on Retrieval-Augmented Generation (RAG) can serve as an innovative solution to support a more equitable enforcement of the law. This technology integrates information retrieval and data-driven text generation to help the public understand their rights, access valid legal information, and obtain relevant legal guidance. Based on the implementation and testing of a criminal law chatbot using the LLaMA language model, questions, generated answers, and the relevance of the chatbot’s responses were evaluated. Out of ten tested questions, eight received relevant responses, while two were deemed irrelevant. Additionally, the chatbot’s capability to retrieve legal documents based on user-provided prompts was assessed. Of ten input prompts, the chatbot successfully identified eight relevant documents, achieving a hit rate of 80%. Thus, the application of RAG in this legal chatbot can provide an innovative solution to support law enforcement. In the future, the use of AI in legal systems has the potential to reduce information disparities, enhance transparency in legal processes, and create a more efficient and accessible legal system for all.

2. 

   Crowdsource, Crawl, or Generate? Creating SEA-VL, a Multicultural Vision-Language Dataset for Southeast Asia

   Samuel Cahyawijaya, Holy Lovenia, Joel Ruben Antony Moniz, and 89 more authors

   2025

3. 

   Feature Selection for Hypertension Risk Prediction Using XGBoost on Single Nucleotide Polymorphism Data

   Healthcare Informatics Research, 2025

4. 

   Anak Baik: A Low-Cost Approach to Curate Indonesian Ethical and Unethical Instructions

   Sulthan Abiyyu Hakim, Rizal Setya Perdana, and Tirana Noor Fatyanosa

   In Proceedings of the Second Workshop in South East Asian Language Processing, Jan 2025

2024

1. 

   Detection of AI-Generated Images From Various Generators Using Gated Expert Convolutional Neural Network

   R. Ahmad Fattah Saskoro, Novanto Yudistira, and Tirana Noor Fatyanosa

   IEEE Access, Jan 2024

2. 

   A Hybrid Approach with BRKGA and Data Mining for the Early/Tardy Scheduling Problem

   Israel Mendonça, Tirana Noor Fatyanosa, Masayoshi Aritsugi, and 1 more author

   In 2024 IEEE Congress on Evolutionary Computation (CEC), Jan 2024

3. 

   Meta-learning for vessel time series data imputation method recommendation

   Tirana Noor Fatyanosa, Neni Alya Firdausanti, Putu Hangga Nan Prayoga, and 3 more authors

   Expert Systems with Applications, Jan 2024

2023

1. 

   NusaCrowd: Open Source Initiative for Indonesian NLP Resources

   Samuel Cahyawijaya, Holy Lovenia, Alham Fikri Aji, and 45 more authors

   In Findings of the Association for Computational Linguistics: ACL 2023, Jul 2023

   Abs

   We present NusaCrowd, a collaborative initiative to collect and unify existing resources for Indonesian languages, including opening access to previously non-public resources. Through this initiative, we have brought together 137 datasets and 118 standardized data loaders. The quality of the datasets has been assessed manually and automatically, and their value is demonstrated through multiple experiments.NusaCrowd’s data collection enables the creation of the first zero-shot benchmarks for natural language understanding and generation in Indonesian and the local languages of Indonesia. Furthermore, NusaCrowd brings the creation of the first multilingual automatic speech recognition benchmark in Indonesian and the local languages of Indonesia. Our work strives to advance natural language processing (NLP) research for languages that are under-represented despite being widely spoken.

2. 

   Classification of unexposed potsherd cavities by using deep learning

   Israel Mendonça, Mai Miyaura, Tirana Noor Fatyanosa, and 4 more authors

   Journal of Archaeological Science: Reports, Jul 2023

   Abs

   The aim of the Deep Research for Akkon project is to use machine-learning techniques, specifically convolutional neural networks (CNNs), to analyze the large corpus of archaeological pottery fragments belonging to the Late Jomon period (ca. 4000 to 3200 aBP) in a non-invasive and non-destructive manner. In Japan, recent research conducted by using the ”impression method” revealed that ceramic vessels belonging to that era contain impressions of plant seeds. Studying these impressions allows archaeologists to better understand the cultural practices of the Jomon people and the path of rice diffusion in Japan. Most of the analysis was conducted by visually inspecting X-ray images. However, results based on using only X-rays are often inconclusive, leading researchers to use other invasive techniques that damage the potsherds. In this paper, we present a method that classifies X-ray images of potsherds by using deep learning. A dataset composed of 1036 images with seven classes was used to evaluate our approach. We generated different models using different CNN architectures and parameters and analyzed them separately. Then, we applied techniques such as ensemble and test-time augmentation to evaluate how such techniques impact our model. The best result was obtained by combining EfficientNetB7 trained with different parameters, achieving a classification rate of 90%.

3. 

   ImputAnom: Anomaly Detection Framework Using Imputation Methods for Univariate Time Series

   Tirana Noor Fatyanosa, Mahendra Data, Neni Alya Firdausanti, and 3 more authors

   In Information Integration and Web Intelligence, Jul 2023

4. 

   Hybrid Feature Selection Framework for Building Resource Efficient Intrusion Detection Systems Model in the Internet of Things

   Tirana Noor Fatyanosa, and Mahendra Data

   In Proceedings of the 8th International Conference on Sustainable Information Engineering and Technology, Jul 2023

5. 

   Conducting Vessel Data Imputation Method Selection Based on Dataset Characteristics

   Tirana Noor Fatyanosa, Neni Alya Firdausanti, Luis Francisco Japa Soto, and 3 more authors

   IOP Conference Series: Earth and Environmental Science, Jun 2023

2022

1. 

   Two-Stage Sampling: A Framework for Imbalanced Classification With Overlapped Classes

   Neni Alya Firdausanti, Tirana Noor Fatyanosa, Mahendra Data, and 2 more authors

   In 2022 IEEE International Conference on Big Data (Big Data), Jun 2022

2021

1. 

   Phrase Table Combination Based on Symmetrization of Word Alignment for Low-Resource Languages

   Sari Dewi Budiwati, Al Hafiz Akbar Maulana Siagian, Tirana Noor Fatyanosa, and 1 more author

   Applied Sciences, Feb 2021

   Abs

   Phrase table combination in pivot approaches can be an effective method to deal with low-resource language pairs. The common practice to generate phrase tables in pivot approaches is to use standard symmetrization, i.e., grow-diag-final-and. Although some researchers found that the use of non-standard symmetrization could improve bilingual evaluation understudy (BLEU) scores, the use of non-standard symmetrization has not been commonly employed in pivot approaches. In this study, we propose a strategy that uses the non-standard symmetrization of word alignment in phrase table combination. The appropriate symmetrization is selected based on the highest BLEU scores in each direct translation of source–target, source–pivot, and pivot–target of Kazakh–English (Kk–En) and Japanese–Indonesian (Ja–Id). Our experiments show that our proposed strategy outperforms the direct translation in Kk–En with absolute improvements of 0.35 (a 11.3% relative improvement) and 0.22 (a 6.4% relative improvement) BLEU points for 3-gram and 5-gram, respectively. The proposed strategy shows an absolute gain of up to 0.11 (a 0.9% relative improvement) BLEU points compared to direct translation for 3-gram in Ja–Id. Our proposed strategy using a small phrase table obtains better BLEU scores than a strategy using a large phrase table. The size of the target monolingual and feature function weight of the language model (LM) could reduce perplexity scores.

2. 

   An Automatic Convolutional Neural Network Optimization Using a Diversity-Guided Genetic Algorithm

   Tirana Noor Fatyanosa, and Masayoshi Aritsugi

   IEEE Access, Feb 2021

3. 

   BERT Goes Brrr: A Venture Towards the Lesser Error in Classifying Medical Self-Reporters on Twitter

   Alham Fikri Aji, Haryo Akbarianto Wibowo, Made Nindyatama Nityasya, and 2 more authors

   In , Feb 2021

4. 

   To Optimize, or Not to Optimize, That Is the Question: TelU-KU Models for WMT21 Large-Scale Multilingual Machine Translation

   Sari Dewi Budiwati, Tirana Fatyanosa, Mahendra Data, and 5 more authors

   In Proceedings of the Sixth Conference on Machine Translation, Nov 2021

   Abs

   We describe TelU-KU models of large-scale multilingual machine translation for five Southeast Asian languages: Javanese, Indonesian, Malay, Tagalog, Tamil, and English. We explore a variation of hyperparameters of flores101_mm100_175M model using random search with 10% of datasets to improve BLEU scores of all thirty language pairs. We submitted two models, TelU-KU-175M and TelU-KU- 175M_HPO, with average BLEU scores of 12.46 and 13.19, respectively. Our models show improvement in most language pairs after optimizing the hyperparameters. We also identified three language pairs that obtained a BLEU score of more than 15 while using less than 70 sentences of the training dataset: Indonesian-Tagalog, Tagalog-Indonesian, and Malay-Tagalog.

5. 

   ParaCotta: Synthetic Multilingual Paraphrase Corpora from the Most Diverse Translation Sample Pair

   Alham Fikri Aji, Radityo Eko Prasojo Tirana Noor Fatyanosa, Philip Arthur, and 5 more authors

   In Proceedings of the 35th Pacific Asia Conference on Language, Information and Computation, Nov 2021

2020

1. 

   Effects of the Number of Hyperparameters on the Performance of GA-CNN

   Tirana Noor Fatyanosa, and Masayoshi Aritsugi

   In , Dec 2020

2019

1. 

   DBMS-KU Interpolation for WMT19 News Translation Task

   Sari Dewi Budiwati, Al Hafiz Akbar Maulana Siagian, Tirana Noor Fatyanosa, and 1 more author

   In , Dec 2019

2. 

   Hybrid Modified Evolution Strategies and Linear Programming for Beef Cattle Feed Optimization

   Tirana Noor Fatyanosa, Wayan Firdaus Mahmudy, and Marjuki Marjuki

   International Journal on Electrical Engineering and Informatics, Mar 2019

3. 

   DBMS-KU at SemEval-2019 Task 9: Exploring Machine Learning Approaches in Classifying Text as Suggestion or Non-Suggestion

   Tirana Fatyanosa, Al Hafiz Akbar Maulana Siagian, and Masayoshi Aritsugi

   In , Mar 2019

2018

1. 

   Feature Selection using Variable Length Chromosome Genetic Algorithm for Sentiment Analysis

   Tirana Noor Fatyanosa, Fitra A. Bachtiar, and Mahendra Data

   In , Nov 2018

2. 

   A Comparison of Four Types of Evolution Strategies for Beef Cattle Feed Optimization

   Tirana Noor Fatyanosa, Wayan Firdaus Mahmudy, and Marjuki Marjuki

   Journal of Telecommunication, Electronic and Computer Engineering (JTEC), Feb 2018

3. 

   Linear Programming Initialization Method of Evolution Strategies for Beef Cattle Feed Optimization

   Tirana Noor Fatyanosa, Fitri Utaminingrum, and Mahendra Data

   Journal of Telecommunication, Electronic and Computer Engineering (JTEC), Jan 2018

2017

1. 

   Classification method comparison on Indonesian social media sentiment analysis

   Tirana Noor Fatyanosa, and Fitra A. Bachtiar

   In , Nov 2017

2. 

   Extracting fuzzy rules and parameters using particle swarm optimization for rainfall forecasting

   Tirana Noor Fatyanosa, Gusti Ahmad Fanshuri Alfarisy, Fatwa Ramdani, and 2 more authors

   In , Oct 2017

3. 

   Hybrid Genetic Algorithm and Simulated Annealing for Function Optimization

   Tirana Noor Fatyanosa, Andreas Nugroho Sihananto, Gusti Ahmad Fanshuri Alfarisy, and 2 more authors

   Journal of Information Technology and Computer Science, Feb 2017

   Abs

   The optimization problems on real-world usually have non-linear characteristics. Solving non-linear problems is time-consuming, thus heuristic approaches usually are being used to speed up the solution’s searching. Among of the heuristic-based algorithms, Genetic Algorithm (GA) and Simulated Annealing (SA) are two among most popular. The GA is powerful to get a nearly optimal solution on the broad searching area while SA is useful to looking for a solution in the narrow searching area. This study is comparing performance between GA, SA, and three types of Hybrid GA-SA to solve some non-linear optimization cases. The study shows that Hybrid GA-SA can enhance GA and SA to provide a better result

2016

1. 

   Implementation of Real Coded Genetic Fuzzy System for Rainfall Forecasting

   Tirana Fatyanosa, and Wayan Mahmudy

   In , Oct 2016