AIAditya Firman Ihsan

Aditya Firman Ihsan

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Aditya Firman Ihsan has a total of 25 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Society-oriented AI governance: a parallel layered model and multi-actor coordination framework

    by Aditya Firman Firman Ihsan, Thomhert Suprapto Suprapto Siadari, Andry Alamsyah, Kemas Muslim Lhaksmana, Helni Mutiarsih Mutiarsih Jumhur

    The rapid and unrestricted public adoption of large language models and other openly accessible AI technologies has exposed a fundamental structural weakness in existing AI governance frameworks: the systematic underrepresentation of society as the ultimate recipient of AI’s consequences. Most current frameworks treat societal impact as a downstream consideration rather than a foundational design criterion. This paper proposes society-oriented governance — the principle that societal reception and utilization of AI must serve as the primary design criterion for governance frameworks — as a structural response to this failure.To operationalize this principle, we develop two interconnected analytical frameworks: a governance development framework that replaces the sequential, hierarchical structure of a prior layered governance model with a parallel and dynamic model in which the technical, ethical, and social layers are developed simultaneously and in continuous mutual interaction; and a multiple-actors framework that maps stakeholders across the full lifecycle of AI utilization and proposes coordination mechanisms suited to both centralized and federated governance paradigms. A key finding is the structural distinction between the 4-actors model, applicable to professionally mediated AI deployment, and the 3-actors model, which describes the dominant pattern of openly accessible consumer AI where no professional intermediary exists. This distinction reveals the social layer as the most critical and most underdeveloped layer in current AI governance, and motivates concrete recommendations for strengthening it through socially responsive design standards, national-level regulatory obligations, and federated multi-stakeholder coordination. These contributions are grounded in AI governance theory — in governance design, institutional architecture, and stakeholder coordination — and are not proposed as a sociological theory of society.

  • Open Access

    Investigating Shallow Learning Methods for Optical Character Recognition of Indonesia’s Nusantara Scripts

    by Mahmud Dwi Sulistiyo, Aji Gautama Putrada, Aditya Firman Ihsan, Prasti Eko Yunanto, Donny Richasdy, Hassan Rizky Putra Sailellah

    Indonesia has numerous regional scripts—or so-called Nusantara scripts—and recognizing them is important to preserve Indonesia's cultural heritage. The advances of AI and computer vision technologies make it possible for a machine to optically read the handwritten scripts through the Optical Character Recognition (OCR) technique. However, collecting some of the top OCR solutions and comprehensively investigating their performances on the Nusantara scripts is currently lacking. This study investigates and evaluates some shallow learning-based methods on our newly introduced datasets, consisting of more than 38,000-character images across 80 letter classes in total; here, we focus on three regional scripts: Javanese, Sundanese, and Balinese. The methods include Random Forest, SVM, Logistic Regression, and Gaussian Naïve Bayes, as well as boosting techniques such as XGBoost, Light GBM, and CatBoost. A 5-fold cross-validation approach assessed model performance based on accuracy, precision, recall, and F1-score. Based on the experimental results, the methods demonstrated their competitiveness in reaching the best models for scripts; in particular, XGBoost, Light GBM, and Random Forest-Gini were the winners for Javanese, Sundanese, and Balinese scripts, respectively. These findings demonstrate the effectiveness of ensemble learning methods for diverse handwritten scripts. Comparative analysis to prior deep learning studies is also discussed in this paper. In addition, this research also contributes to preserving Indonesian traditional scripts, as well as offers insights for future regional OCR in other countries.


  • Open Access

    On Data Driven SIRD Model of Delta and Omicron Variants of COVID-19

    by Aditya Firman Ihsan

    The compartmental model stands as a cornerstone in quantitatively describing the transmission dynamics of diseases. Through a series of assumptions, this model can be formulated and subsequently validated against real-world conditions. Leveraging the abundance of COVID-19 data presently available, this study endeavors to reverse engineer the model construction process. Specifically, we analyse the compartmental model governing two notable variants of COVID-19: Delta and Omicron, utilizing empirical data. Employing the SINDy method, we extract parameters that define the model by effectively fitting the available data. To ensure robustness, the obtained model undergoes validation via comparison with real-world data through numerical integration. Additionally, we conduct fine-tuning in regularization techniques and input features to refine model selection. The constructed model then undergoes thorough analysis to gain qualitative insights and interpretations regarding the transmission dynamics of COVID-19.

  • Anomaly detection on natural gas operational data using variational autoencoder

    by Aditya Firman Ihsan, Widi Astuti, Hasmawati

    In the natural gas industry, production monitoring has many challenging problems that need to be solved using machine learning tasks. One of such tasks is anoma In the natural gas industry, production monitoring has many challenging problems that need to be solved using machine learning tasks. One of such tasks is anoma

  • Open Access

    Clustering Content Types and User Motivation Using DBSCAN on Twitter

    by Made Mita Wikantari, Yuliant Sibaroni, Aditya Firman Ihsan

    We are currently in an era full of information and communication technology. One of the communication media used is Twitter. Twitter is a microblogging service that is used by its users to express their thoughts on a topic called a tweet. Tweets that are posted can be either positive tweets or negative tweets. One of the topics that is currently being discussed by Twitter users is Anies Baswedan as a 2024 Indonesian Presidential Candidate. Many people have tweeted this but it is not known how many users support or reject Anies Baswedan to run as a 2024 Indonesian presidential candidate. To assist the analysis, use the method clustering namely algorithm (Density-Based Spatial Clustering of Application with Noise). DBSCAN has the advantage of being able to detect data that is not included in a cluster and will be considered noise. This can improve the accuracy of the grouping because the data in the cluster will be cleaner. The TF-IDF Vectorizer is used to make it easier for programs to manage data because it can turn sentences into vectors that can be processed by the algorithm. To determine the evaluation of the program, the silhouette score method will be used. The results of calculating the silhouette score show a value of 0.29 with the formation of 3 clusters. Then an analysis is carried out based on the top words from each cluster and it can be identified that cluster 0 has a positive category supporting Anies Baswedan to run for the 2024 Presidential Candidate and cluster 1 has a negative category that does not support Anies Baswedan not advancing for the 2024 Presidential Candidate.


  • Open Access

    Clustering Content Types and User Roles Based on Tweet Text Using K-Medoids Partitioning Based

    by Raisa Benaya, Yuliant Sibaroni, Aditya Firman Ihsan

    In this modern era, the spread of information occurs rapidly through social media. One of the channels for disseminating information is through the Twitter platform. Many Twitter users respond to existing content with positive, negative and neutral responses. One of the hot content to respond to is political content. This content is currently being discussed considering the approaching election of the 2024 Presidential Candidate of the Republic of Indonesia. One of the candidate pairs discussed was Anies Baswedan. With so many responses from Twitter users, it will be difficult to track whether users support Anies Baswedan to run as a presidential candidate due to the large number of responses. This study aims to determine the response of twitter users to the advancement of Anies Baswedan as a presidential candidate. The method used in this study is the K-Medoids Partitioning-Based algorithm based on twitter user text. This algorithm was chosen because it is easy to implement considering the basis of K-Medoids development is the K-Means algorithm but the K-Medoids algorithm can overcome the shortcomings of the K-Means algorithm which is sensitive to outliners. The evaluation will be done using Silhouette Score which produces a value of 0.35 with the number of clusters is 2. Then an analysis of each cluster is carried out by looking at the words in the cluster. As a result, from the two clusters formed, both clusters contain positive content and show that Twitter users support Anies Baswedan to run as a 2024 presidential candidate.


  • Open Access

    Performance Analysis of Deep Learning Implementation in Operational Condition Forecasting of a Gas Transmission Pipeline Network

    by Aditya Firman Ihsan, Darmadi, Saladin Uttunggadewa, Silvya Dewi Rahmawati, Irsyad Giovanni, Salamet Nur Himawan

    Monitoring natural gas transmission in a pipeline network is important to maintain the supply and demand balance in natural gas transactions and distribution. Gas pressure, temperature, flowrate, and gas properties must be monitored during the transmission process. These variables, also known as operational conditions, need to be simulated carefully to understand the dynamics and behavior over time. Commonly used physical equations, such as thermodynamic or hydraulic equations, have limitations in simulating future trends because they need some known boundary conditions to be solved. In that case, data-driven method is needed, especially nowadays when data management is widely implemented. This paper implements a deep Recurrent Neural Network (RNN) to forecast the future behavior of gas pressure as an operational condition in a gas pipeline network receiving platform. Different types of recurrent cells are used, i.e., Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU). The model is trained in 8 years of a gas pipeline network operational data. Historical flowrate data in the end-nodes become the forecast input in addition to the past pressure data. The sensitivity of the model and learning parameters is experimented with and analyzed to understand the capacity of the RNN in the given task. Mean absolute error is set as the satisficing metric, whereas the training time is set as optimizing metrics. The obtained best model successfully forecasts the future pressure of one day ahead with only around 2% relative error.


  • Open Access

    Analysis Content Type and Emotion of the Presidential Election Users Tweets using Agglomerative Hierarchical Clustering

    by Cika Carissa Sujadi, Yuliant Sibaroni, Aditya Firman Ihsan

    Over the past few years, social media has become essential for getting up-to-date information and interacting online. During presidential elections in Indonesia, Twitter has grown as a crucial platform for expressing opinions and sharing information. This study focuses on analyzing the content types and emotions of tweets related to Anies Baswedan, one of the presidential candidates. The results show a variety of discussions, including support, criticism, and discussion of policies for the 2024 presidential candidate. Clustering enables meaningful information extraction from vast Twitter data. Data were clustered using Agglomerative Hierarchical Clustering, which resulted in the identification of 10 clusters. With 4 clusters containing opinion content and 6 clusters containing information content. In addition, 6 clusters reflect excitement, 3 reflect expectations, and 1 reflect doubt. This research provides insights into the Twitter conversation around the 2024 presidential election, providing an understanding of content and emotions expressed by users.


  • Open Access

    The Organization Entity Extraction Telkom University Affiliated using Recurrent Neural Network (RNN)

    by Muhammad Daffa Regenta Sutrisno, Donni Richasdy, Aditya Firman Ihsan

    In the news portal text, there is a lot of important information such as the name of the person, the name of the organization, or the name of the place. To obtain information in text documents manually, humans must read the contents of the entire news text. To overcome this issue, information extraction, commonly called Named Entity Recognition (NER) was used. The extraction of information expressly for the NER field is used to make it easier to process word or sentence data. It helps search engines and also helps to classify places, times, and organizations. There is a limited number of NER in Indonesian texts using only the Recurrent Neural Network (RNN) method. Similar previous studies only employed other versions of RNN such as LSTM (Long Short Term Memory), BiLSTM (Bidirectional Long Short Term Memory), and CNN (Convolutional Neural Network). NER is one of the answers to the problems that exist in a large number of news portal texts to obtain information effectively and efficiently. The results of this study indicate that the NER system using the RNN method in Indonesian news texts has an F1 -Score of 80%