ISIrwan Sembiring

Irwan Sembiring

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Irwan Sembiring has a total of 34 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Improving Multilabel Classification Model Performance in Imbalanced Datasets with Group-Level Undersampling

    by Danny Sebastian, Hindriyanto Dwi Purnomo, Irwan Sembiring

    Multilabel classification presents unique challenges, particularly when dealing with imbalanced datasets. This study introduces a Group-Level Undersampling (GLU) technique designed to enhance the performance and efficiency of multilabel classification models. By converting multilabels into combined labels and categorizing them into group levels, the proposed method preserves the diversity of label combinations, which is crucial for accurate classification. Undersampling is particularly challenging in multilabel classification because removing a single instance can affect the combination of classes. This study utilized Google Maps reviews of tourist sites in Yogyakarta, Indonesia, manually labeled into 11 classes. The IndoBERTweet model was fine-tuned using the GLU generated dataset, and its performance was compared with models trained on randomly selected datasets. Experimental results demonstrate that the GLU method can maintain the variation of multilabel datasets by using combined labels and group levels. The experimental results show that the GLU dataset significantly improves model accuracy and F1 scores while reducing computational time and resources. The findings suggest that the proposed undersampling technique offers a robust solution to address imbalanced datasets in multilabel classification, paving the way for more efficient and accurate machine learning models. However, there are still weaknesses in the ratio calculation formula, allowing some datasets to be suboptimal at certain group levels. Future research should apply the GLU technique to various datasets and multilabel tasks to assess its generalizability. Investigating hybrid data-balancing methods could further enhance model performance and efficiency. A sensitivity analysis on ratioBaseLine values and refining the ratio calculation formula will improve the robustness of the GLU method.


  • Open Access

    Implementation of ARIMA with Min-Max Normalization for predicting the Price and Production Quantity of Red Chili Peppers in North Sumatra Province considering Rainfall and Sunlight Duration Factors

    by Ifan Prihandi, Sutarto Wijono, Irwan Sembiring, Evi Maria

    Red chili peppers are a vital agricultural commodity in the North Sumatra province, playing a significant role in Indonesia's economy. Fluctuations in chili prices affect farmers, consumers, and overall economic stability. This study leverages time series forecasting using the ARIMA model to predict red chili pepper prices and production, incorporating weather factors such as rainfall and sunlight duration. The dataset spans March 2021 to December 2023 and includes historical records of chili prices, production levels, and weather conditions. The analysis reveals a strong correlation between price fluctuations and production trends: Prices tend to rise when production declines and fall when yields increase. Additionally, production is influenced by weather conditions, where excessive rainfall damages crops and reduces yields, while balanced rainfall and sunlight duration support optimal growth. The ARIMA model demonstrates its effectiveness in capturing these patterns, providing actionable insights for farmers and policymakers to predict price changes and optimize production strategies. By integrating data-driven forecasting with weather analysis, this research contributes to more adaptive and informed decision-making in the agricultural sector.


  • Open Access

    Network Intrusion Detection Using Transformer Models and Natural Language Processing for Enhanced Web Application Attack Detection

    by Wowon Priatna, Irwan Sembiring, Adi Setiawan, Iwan Iwan Setyawan

    The increasing frequency and complexity of web application attacks necessitate more advanced detection methods. This research explores integrating Transformer models and Natural Language Processing (NLP) techniques to enhance network intrusion detection systems (NIDS). Traditional NIDS often rely on predefined signatures and rules, limiting their effectiveness against new attacks. By leveraging the Transformer's ability to capture long-term dependencies and the contextual richness of NLP, this study aims to develop a more adaptive and intelligent intrusion detection framework. Utilizing the CSIC 2010 dataset, comprehensive preprocessing steps such as tokenization, stemming, lemmatization, and normalization were applied. Techniques like Word2Vec, BERT, and TF-IDF were used for text representation, followed by the application of the Transformer architecture. Performance evaluation using accuracy, precision, recall, F1 score, and AUC demonstrated the superiority of the Transformer-NLP model over traditional machine learning methods. Statistical validation through Friedman and T-tests confirmed the model's robustness and practical significance. Despite promising results, limitations include the dataset's scope, computational complexity, and the need for further research to generalize the model to other types of network attacks. This study indicates significant improvements in detecting complex web application attacks, reducing false positives, and enhancing overall security, making it a viable solution for addressing increasingly sophisticated cybersecurity threats


  • Open Access

    Progress in Non-Invasive Cognitive Brain-Computer Interface and Implications for Mind-Uploading

    by I Wayan Aswin Dew Astawa, Hindriyanto Dwi Purnomo, Irwan Sembiring

    Mind-uploading, the vision of transferring human consciousness into a digital realm, relies on a profound comprehension of the brain and cutting-edge technology. Non-invasive cognitive Brain-Computer Interfaces (BCI) offer a promising avenue for delving into neural activity and bridging the brain-machine gap. This research explores the potential of non-invasive cognitive BCI in realizing mind-uploading through a systematic literature review (SLR), analyzing recent research that focuses on its current progress and implications for mind-uploading. The SLR unveils significant strides in non-invasive cognitive BCI, demonstrating increased precision in recording and decoding cognitive processes and fostering a deeper understanding of these processes. This progress is attributed to a diverse range of emerging feature extraction and decoding methods, transforming subtle neural signals into interpretable commands. Notably, advancements in signal processing and neuroimaging techniques enhance communication speed and clarity between the brain and computer. Furthermore, the development of cost-effective methods, frameworks, and hardware holds the promise of broader accessibility to BCI technology. However, significant hurdles remain. The computational demands of current cognitive BCI systems pose a substantial challenge, while the scarcity of high-quality training datasets hampers algorithm development and accuracy. The poor signal quality causes difficulties in recording neural complexity and hampers accuracy. In conclusion, non-invasive cognitive BCI has significant potential to pave the way for mind-uploading. However, its limitations, make their capabilities remain insufficient to fully realize this ambitious vision. This highlights the critical need for sustained research and innovation to bridge the gap between current understanding and the exciting realm of mind-uploading.


  • Open Access

    Implementation of Wireless User Authentication using WLC-Forti Framework

    by Ignatius Agus Supriyono, Irwan Sembiring, Adi Setiawan, Iwan Setyawan, Theophilus Wellem, Henderi, Ilham Hizbuloh

    Internet access at this time is a daily necessity that cannot be denied. It is certain that most institutions and business entities require internet access in carrying out their activities, including educational institutions. With the development of mobile computer technology in which more users use mobile devices to access the internet, wireless-based network infrastructure is a demand that cannot be postponed any longer. By using a wireless connection to connect to the network, authentication becomes something that must be considered, the use of access to the network by unwanted parties can harm other parties. Changing passwords regularly is important to avoid misuse of access to the network by other parties. This paper presents a problem where when an educational institution implements the Bring Your Own Device (BYOD) program, students and teachers cannot change passwords using the personal device used, this is because the personal device is not registered with the domain controller at the institution. The solution proposed in this article is to move the NPS RADIUS server function on the local site to LDAP in the cloud using a combination of WLC which handles Wi-Fi clients and Fortinet which handles authentication to the cloud. The implementation results show that the WLC-Forti framework functions well.


  • Open Access

    Perancangan sistem informasi penjualan berbasis CRM dengan menggunakan metode FAST dan framework PIECES

    by Rimes Jopmorestho Malioy, Irwan Sembiring, Ade Iriani

    Kios Tani Murni is a home industry that specializes in selling agriculture. The monthly sales rate at this store is low because people need to learn about the existence and function of this store. Kios Tani Murni has yet to utilize information technology in the sales process. The relationship between the customer and the seller is not going well because there is no forum for discussion. For this reason, this research created an information system design based on customer relationship management (CRM) in the sales process, using the Framework for the Application of System Technique (FAST) method and the PIECES framework for system analysis. Observations, interviews, and literature studies were used to collect data. The result of this research is that there is a sales information system design. The system is helpful for sales transactions, stock items, price updates, employee performance, report printing, and discussions between owners and customers.


  • Open Access

    A Note on the Combination of the New Similarity Formula with Feedback to Better Handle Complaints of In Vitro Fertilization (IVF) Patients

    by Irwan Sembiring, Paminto Agung Christianto, Eko Sediyono

    In Vitro Fertilization (IVF) patients frequently experience anxiety and report it right away to their fertility doctors. But because the fertility specialists took their time to respond, IVF patients eventually became more anxious. The research's findings demonstrated that using the standard feasibility value of 80% along with the Chris Case-Based Reasoning (CCBR) similarity formula led to an accurate system recommendation (100%) with a higher precision value (83.15%). The F value of the ANOVA test was 9.902 with a significance level of 0.007


  • Open Access

    Pelatihan Data Science Pada 2024 Guru dan Siswa SMA/SMK Provinsi Nusa Tenggara Timur

    by Danny Manongga, Ade Iriani, Budhi Kristianto, Irwan Sembiring, Hendry, Evangs Mailoa, Nina Setiyawati, Dwi Hosanna Bangkalang

    Data menjadi aset paling berharga untuk organisasi mana pun karena dapat memandu pengambilan keputusan. Oleh karena itu kemampuan data science merupakan salah satu skill penting. Data science tergambarkan sebagai proses yang dimulai dari pengumpulan dan pengolahan, kemudian disajikan sebagai informasi yang berguna untuk pengambilan keputusan atau bermanfaat bagi pihak yang berkepentingan dengan data. Data science memiliki banyak fungsi dan manfaat dimana beberapa diantaranya adalah membantu menciptakan budaya keputusan berbasis data, mengurangi ketidakpastian dan meningkatkan konsistensi dan keandalan data. Melihat pentingnya kemampuan data science, maka Fakultas Teknologi Informasi bekerja sama dengan ASEAN Foundation serta Dinas Pendidikan dan Kebudayaan Provinsi Nusa Tenggara Timur (NTT) melakukan Pengabdian kepada Masyarakat (PkM) dalam bentuk pelatihan kepada 2024 guru dan siswa SMA/SMK Provinsi NTT sebagai bagian untuk mencetak talenta digital Indonesia. Pelatihan didukung oleh SAP yang merupakan perusahaan software dan teknologi yang berbasis di Jerman melalui platform SAP Analytics Cloud (SAC). PkM dilaksanakan secara daring dan luring. Guru dan siswa antusias mengikuti pelatihan ini terlihat dari hasil evaluasi yang bisa dikerjakan dengan baik oleh para peserta.


  • Open Access

    Security Level Significance in DApps Blockchain-Based Document Authentication

    by Qurotul Aini, Danny Manongga, Untung Rahardja, Irwan Sembiring, Vonda Elmanda, Adam Faturahman, Nuke Puji Lestari Santoso

    In the development of the Industrial revolution 4.0 to improve and modify the world's industry by integrating production lines, and extraordinary results in the field of technology and information marked its emergence. It can be used to enhance document security systems using Blockchain technology. Blockchain Innovation Authentication has attracted great attention in the world of science and capital markets. The persistent problems of the many available digital currencies and the various tricks of early coin offerings also welcome the well-known discussion of emerging innovations in the field of education. The importance of this paper follows the improvement of the blockchain framework to reveal the importance of decentralized applications (dApps) and blockchain on the future value in education. This study uses a descriptive method, which is a research method used to describe problems that occur in the present or ongoing, aiming to describe what happened as it should when the research was conducted. the novelty of cutting-edge dApps and talk about the title of blockchain progress to meet the positive attributes of future dApps. Readers will come to the conclusion of dApp research and know the continuous improvement in blockchain .


  • Open Access

    Analisis Pengguna Media Sosial Terhadap Isu UU Cipta Kerja Menggunakan SNA dan Naive Bayes

    by Stevanus Dwi Istiavan Mau, Irwan Sembiring, Hindriyanto Purnomo

    In research about analysis issue analysis, UU Cipta Kerja aims to investigate aktor which has the most influence in the discussion on the network is based on the analysis of the most popular centrality values on the issues law copyright this working. Research is in use of the method of Social Network Analysis ( SNA ). The data in minutely as many as 1686 nodes and 1403 edges in extract through the API Twitter with the help of application WinPython and Netlytic with a period 08 October 2020 - 05 July 2021. The result of this research showed that account @BEMUI_ Official is account the most popular with the Degree Centrality 944, value Betweenness Centrality 640042.0. But on a calculation Closeness Centrality aktor @BEMUI_Official having value 0.701235, therefore nodes who have a centrality highest do not necessarily have the value that both in terms of the dissemination of information.