Garuda - Garba Rujukan Digital

p-Index From 2021 - 2026

5.19

P-Index

This Author published in this journals

Nanik Suciati

Departemen Informatika Institut Teknologi Sepuluh Nopember

Author-ID : 387460

Arts Humanities Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Dentistry Education Electrical & Electronics Engineering Engineering Environmental Science Industrial & Manufacturing Engineering Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Library & Information Science Mathematics Mechanical Engineering Public Health Social Sciences Other

Published : 91 Documents Claim Missing Document

Claim Missing Document

Articles

6 7 8 9 10

Road Damage Detection Using YOLOv7 with Cluster Weighted Distance-IoU NMS Rachman, Rudy; Suciati, Nanik; Hidayati, Shintami Chusnul
JOIN (Jurnal Online Informatika) Vol 10 No 1 (2025)
Publisher : Department of Informatics, UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15575/join.v10i1.1481

Road damage can occur everywhere. Potholes are one of the most common types of road damage. Previous research that used images as input for pothole detection used the Faster Regional Convolutional Neural Network (R-CNN) method. It has a large inference time because it is a two-stage detection method. The object detection method requires post-processing for its detection results to save only the best prediction from the method, namely, non-maximum suppression (NMS). However, the original NMS could not properly detect small, far, and two objects close to each other. Therefore, this research uses the YoloV7 method as the object detection method because it has better mean Average Precision (mAP) results and a lower inference time than other object detection methods; with an improved NMS method, namely Cluster Weighted Distance Intersection over Union (DIoU) NMS (CWD-NMS), to solve small or close potholes. When training YoloV7, we combined a new, independently collected pothole dataset, with previous public research datasets, where the detection results of the YoloV7 method were better than those of Faster R-CNN. The YoloV7 method was trained using various scenarios. The best scenario during training is using the best checkpoint without using a scheduler. The mAP.5 and mAP.5-.95 value of CWD-NMS was 89.20% and 63.30% with 10.30 millisecond per image for inference time.

THE EFFECT OF FACIAL ACCESSORY AUGMENTATION ON THE ACCURACY OF DEEP LEARNING-BASED FACIAL RECOGNITION SYSTEMS Hidayat, Ahmad Nur; Suciati, Nanik; Saikhu, Ahmad
JURTEKSI (Jurnal Teknologi dan Sistem Informasi) Vol 11, No 3 (2025): Juni 2025
Publisher : Universitas Royal

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33330/jurteksi.v11i3.3846

Abstract: Face recognition based on deep learning has become an important technology in many areas. However, these systems often face challenges in real-world conditions, such as when the face is partially covered by accessories such as masks or glasses. This study aims to evaluate the effect of data augmentation by adding facial accessories (masks, glasses, and a combination of both) and geometric augmentation on the accuracy of face recognition systems. There are three types of datasets used in this method: the original dataset (category 1), the dataset with facial accessories augmentation (category 2), and the dataset with geometric augmentation (category 3). Data augmentation was performed on the training dataset to increase diversity, followed by the face detection process using SCRFD and feature extraction with ArcFace. The model was then trained using Multi-Layer Perceptron (MLP). Based on the results, adding face accessories (category 2) made the model a lot more accurate, hitting 99% accuracy. In category 3, adding geometric features improved accuracy to 91%. Other evaluation metrics, such as precision, recall, and F1-score, also showed improvement after augmentation. This study concludes that facial accessories augmentation is more effective in improving the accuracy and robustness of face recognition models compared to geometric augmentation.Keywords: augmentation; deep learning; face recognition; glasses. Abstrak: Pengenalan wajah berbasis deep learning telah menjadi salah satu teknologi penting dalam berbagai aplikasi. Namun, sistem ini sering kali menghadapi tantangan dalam kondisi dunia nyata, seperti saat wajah tertutup sebagian oleh aksesori seperti masker atau kacamata. Penelitian ini bertujuan untuk mengevaluasi pengaruh augmentasi data dengan menambahkan aksesori wajah (masker, kacamata, dan kombinasi keduanya) serta augmentasi geometris terhadap akurasi sistem pengenalan wajah. Metode yang digunakan melibatkan tiga kategori dataset: dataset asli tanpa augmentasi (kategori 1), dataset dengan augmentasi aksesoris wajah (kategori 2), dan dataset dengan augmentasi geometris (kategori 3). Augmentasi data dilakukan pada dataset pelatihan untuk meningkatkan keberagaman, diikuti dengan proses deteksi wajah menggunakan SCRFD dan ekstraksi fitur dengan ArcFace. Model kemudian dilatih menggunakan Multi-Layer Perceptron (MLP). Hasil penelitian menunjukkan bahwa augmentasi aksesoris wajah (kategori 2) memberikan peningkatan signifikan pada akurasi model, mencapai 99%, sedangkan kategori 3 dengan augmentasi geometris mencapai akurasi 91%. Metrik evaluasi lainnya, seperti precision, recall, dan F1-score, juga menunjukkan peningkatan setelah augmentasi. Penelitian ini menyimpulkan bahwa augmentasi aksesoris wajah lebih efektif dalam meningkatkan akurasi dan ketahanan model pengenalan wajah dibandingkan dengan augmentasi geometris.Kata kunci: augmentasi; deep learning; kacamata; pengenalan wajah.

Attention-Driven U-Net with Ensemble Strategy for Inferior Alveolar Nerve Segmentation on 2.5D CBCT Data Arsy Bilahi Tama; Suciati, Nanik
Jurnal Nasional Pendidikan Teknik Informatika: JANAPATI Vol. 14 No. 2 (2025)
Publisher : Prodi Pendidikan Teknik Informatika Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/janapati.v14i2.96058

Image segmentation plays a crucial role in medical analysis, particularly in accurately identifying anatomical structures. In dental implant planning, the identification of the Inferior Alveolar Nerve (IAN) is critical to avoid complications resulting from nerve injury. However, the manual annotation process on CBCT images is time-consuming and labor-intensive. Recent studies utilizing deep learning for IAN segmentation in 3D images often face two main challenges: limited availability of annotated data and high computational requirements.To address these challenges, this study proposes a more efficient segmentation approach based on 2.5D images. We implemented a U-Net architecture enhanced with attention gates to improve the model's focus on relevant nerve structures and increase segmentation accuracy. Furthermore, to maximize performance, predictions from multiple models were combined using ensemble learning techniques, which enhance robustness and final accuracy by leveraging the predictive strengths of diverse training samples.Experimental results demonstrate that the proposed approach achieves an average Dice score of 87.7%. These findings indicate that the combination of an attention-enhanced U-Net, the use of 2.5D imaging, and ensemble learning effectively yields accurate IAN segmentation while providing a practical solution to the challenges of data scarcity and computational complexity.

Transfer Learning Menggunakan LoRA+ pada Llama 3.2 untuk Percakapan Bahasa Indonesia Kautsar, Faiz; Wicaksono, Farhan; Hafidz, Abdan; Purwitasari, Diana; Suciati, Nanik; Adni Navastara, Dini; Gurat Adillion, Ilham
Techno.Com Vol. 24 No. 2 (2025): Mei 2025
Publisher : LPPM Universitas Dian Nuswantoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62411/tc.v24i2.12508

Penelitian ini mengeksplorasi penerapan dari metode Parameter-Efficient Finetuning (PEFT) Low-Rank Adaptation+ (LoRA+) pada transfer learning model Llama 3.2 1B, sebuah model bahasa besar. Seiring bertambahnya ukuran model bahasa, finetuning yang dilakukan secara konvensional dalam transfer learning semakin tidak fisibel untuk dilakukan tanpa menggunakan komputasi skala besar. Untuk menangani hal tersebut, dapat dilakukan finetuning pada beberapa komponen saja, menggunakan komputasi yang relatif minimal berbanding dengan finetuning konvensional, metode-metode yang menerapkan prinsip ini disebut juga sebagai PEFT. Penelitian menguji efektifitas metode PEFT, yakni LoRA+, pada transfer learning model bahasa besar terhadap domain baru, yakni bahasa Indonesia, menggunakan metrik BLEU, ROUGE, serta Weighted F1. Hasil penelitian menunjukkan bahwa penerapan LoRA+ menghasilkan performa kompetitif dan unggul terhadap baseline dalam kemampuan berbahasa Indonesia, dengan peningkatan 112% pada skor BLEU dan 21.7% pada skor ROUGE-L, dengan standar deviasi yang relatif rendah sebesar 3.72 dan 0.00075. Meskipun terjadi penurunan pada skor Weighted F1 sebesar 13% yang disebabkan oleh domain shift, model menunjukkan kemampuan transfer lintas-bahasa yang baik. Kata kunci: Finetuning, Model Bahasa Besar, Parameter-Efficient Finetuning, Low-Rank Adaptation, Transfer Learning

ESI-YOLO: Enhancing YOLOv8 with Efficient Multi-Scale Attention and Wise-IoU for X-Ray Security Inspection Haq, Arinal; Suciati, Nanik; Bui, Ngoc Dung
International Journal of Robotics and Control Systems Vol 5, No 3 (2025)
Publisher : Association for Scientific Computing Electronics and Engineering (ASCEE)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31763/ijrcs.v5i3.1983

Security inspection is a priority for preventing threats and criminal activities in public places. X-ray imaging can help with the closed luggages checking process. However, interpreting X-ray images is challenging due to the complexity and diversity of prohibited items. This paper proposes ESI-YOLO, an enhanced YOLOv8-based model for prohibited item detection in X-ray security inspection. The model integrates Efficient Multi-Scale Attention (EMA) and Wise-IoU (WIoU) loss function to improve multi-scale feature representation and detection accuracy. EMA improves multi-scale feature representation, while WIoU enhances bounding box regression, particularly in cluttered and overlapping scenarios. Comprehensive experiments on the CLCXray and PIDray datasets validate the effectiveness of ESI-YOLO. A systematic exploration for the optimal placement of EMA integration on YOLOv8 architecture reveals that the scenario with direct integration in both backbone and neck sections emerges as the most effective configuration without introducing significant computational complexity. Ablation experiments demonstrate the synergistic effect of combining EMA and WIoU in ESI-YOLO, outperforming individual component additions. ESI-YOLO demonstrates notable advancements over the baseline YOLOv8 model, achieving mAP50 improvements of 0.9% on CLCXray and 3.5% on the challenging hidden subset of PIDray, with a computational cost of 8.4 GFLOPs. Compared to other nano-sized models, ESI-YOLO exhibits enhanced accuracy while maintaining computational efficiency, making it a promising solution for practical X-ray security inspection systems.

Handling Imbalance in Javanese Manuscript Character Dataset using Skeleton-based Balancing Generative Adversarial Networks Faizin, Muhammad 'Arif; Suciati, Nanik; Fatichah, Chastine
Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Vol 9 No 4 (2025): August 2025
Publisher : Ikatan Ahli Informatika Indonesia (IAII)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29207/resti.v9i4.6572

Javanese script is an important part of Indonesia’s cultural heritage, representing cultural values from the past. However, recognizing and classifying Javanese characters within manuscripts is challenging due to the limited availability of data and uneven distribution of character classes. The decline in formal use of Javanese script has drastically reduced the pool of manuscript samples, causing certain characters to appear rarely and skewing class frequencies. Existing methods that utilize Generative Adversarial Networks (GANs) attempt to address this problem. However, they often struggle to generate characters that are both consistent and visually accurate in terms of structural details. To address these issues, this study introduces a skeleton-based balancing GAN (SkelBAGAN), which improves the structural details of the previous method for generating characters. The proposed method introduces three main enhancements: (i) a layer for extracting the character skeleton structure, (ii) an optimized pretrained network using an autoencoder for learning the skeleton distribution, and (iii) refinement of the evaluation function, preserving both the distribution and structural fidelity in the adversarial process. The performance of the proposed model is evaluated against previous methods using the Fréchet Inception Distance (FID) to assess distribution quality and the Structural Similarity Index Measure (SSIM) to evaluate structural fidelity. The results indicate that the proposed methods outperform previous methods in balancing the FID and SSIM metrics. The integration of all enhancements in SkelBAGAN achieves the lowest FID, indicating improved generative quality while maintaining competitive SSIM values. The qualitative study indicates that SkelBAGAN outperforms previous methods in character generation. These results highlight how the skeleton-based improvement of the quality of generated characters enhances the recognition performance for underrepresented Javanese characters in imbalanced datasets. Ultimately, this work contributes to the broader effort to preserve the Javanese script as a vital element of Indonesia’s cultural identity.

Temu Kembali Citra Tenun Nusa Tenggara Timur menggunakan Esktraksi Fitur yang Robust terhadap Perubahan Skala, Rotasi, dan Pencahayaan Baso, Budiman; Suciati, Nanik
Jurnal Teknologi Informasi dan Ilmu Komputer Vol 7 No 2: April 2020
Publisher : Fakultas Ilmu Komputer, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25126/jtiik.2020722002

Ragam motif pada tenun Nusa Tenggara Timur (NTT) seperti flora, fauna dan geometris menjadi suatu keunikan yang dapat membedakan daerah asal dan jenis dari tenun tersebut. Pada penelitian ini, sistem temu kembali citra berbasis isi atau Content-Based Image Retrieval (CBIR) diimplementasikan pada citra tenun NTT sehingga user dapat mencari citra tenun pada database menggunakan citra query berdasarkan fitur visual yang terkandung dalam citra. Seringkali citra query yang diinputkan user memiliki skala, rotasi dan pencahayaan yang bervariasi, sehingga diperlukan suatu metode ektraksi fitur yang dapat mengakomodasi variasi tersebut. Sistem temu kembali citra tenun pada penelitian ini menggunakan model Bag of Visual Words (BoVW) dari keypoints pada citra yang diekstrak dengan metode Speeded Up Robust Feature (SURF). BoVW dibangun menggunakan K-Means untuk menghasilkan visual vocabulary dari keypoints pada seluruh citra training. Representasi BoVW diharapkan dapat menangani variasi skala dan rotasi pada citra. Sedangkan untuk mengatasi variasi pencahayaan pada citra, dilakukan perbaikan kualitas citra dengan menggunakan Contrast Limited Adaptive Histogram Equalization (CLAHE). Percobaan dilakukan dengan membandingkan kinerja dari representasi BoVW yang dibangun menggunakan fitur SURF dengan Maximally Stable Extremal Regions (MSER) pada temu kembali citra tenun. Hasil uji coba menunjukkan bahwa metode SURF menghasilkan rata-rata akurasi 89,86% dan waktu komputasi 9,94 detik, sedangkan MSER menghasilkan rata-rata akurasi 84,04% dan waktu komputasi 1,95 detik. AbstractThe variety of motifs in East Nusa Tenggara tenun such as flora, fauna and geometric is an unique thing that can distinguish the region of origin and type of the tenun. In this study, the Content-Based Image Retrieval (CBIR) system is implemented in the tenun image. With Content-based techniques Users can search tenun images on the image database by using query images based on visual features contained in the image. Often the query image that the user enters has a different scale, rotation and lighting, so a feature extraction method is needed that can accommodate these differences. The tenun image retrieval system in this study used the Bag of Visual Words (BoVW) model of the keypoints in the extracted image using the Speeded Up Robust Feature (SURF) method. BoVW was built using K-Means to produce visual vocabulary from keypoints on all training images. The representation of BoVW is expected to be able to handle scale variations and rotations in images. Whereas to overcome the lighting variations in the image, image quality improvement is done by using Contrast Limited Adaptive Histogram Equalization (CLAHE). The experiment was conducted by comparing the performance of the BoVW representation which was built using the SURF feature with Maximally Stable Extremal Regions (MSER) at the tenun image retrieval. The results of the trial showed that SURF obtained higher accuracy in all conditions of tenun image data with an average value of 89.86% whereas MSER obtained an average accuracy value of 84.04%. But MSER's computation time is 1.95 seconds faster than SURF which is 9.94 seconds.

Pengenalan Karakter Tulisan Tangan Menggunakan Ekstraksi Fitur Bentuk Berbasis Chain Code Mawaddah, Saniyatul; Suciati, Nanik
Jurnal Teknologi Informasi dan Ilmu Komputer Vol 7 No 4: Agustus 2020
Publisher : Fakultas Ilmu Komputer, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25126/jtiik.2020742022

Pengenalan karakter tulisan tangan pada citra merupakan suatu permasalahan yang sulit untuk dipecahkan, dikarenakan terdapat perbedaan gaya penulisan pada setiap orang. Tahapan proses dalam pengenalan tulisan tangan diantaranya adalah preprocessing, ekstraksi fitur, dan klasifikasi. Preprocessing dilakukan untuk merubah citra tulisan tangan menjadi citra biner yang hanya mempunyai ketebalan 1 pixel melalui proses binerisasi dan thining. Kemudian pada tahap ekstraksi fitur, dipilih fitur bentuk karena fitur bentuk memiliki peran yang lebih penting dibanding 2 fitur visual lainnya (warna dan tekstur) pada pengenalan karakter tulisan tangan. Metode ekstraksi fitur bentuk yang dipilih dalam penelitian ini adalah metode berbasis chain code karena metode tersebut sering digunakan dalam beberapa penelitian pengenalan tulisan tangan. Pada penelitian ini, dilakukan studi kinerja dari ekstraksi fitur berbasis chain code pada pengenalan karakter tulisan tangan untuk mengetahui metode terbaiknya. Tiga metode ekstraksi fitur berbasis chain code yang digunakan dalam penelitian ini adalah freeman chain code, differential chain code dan vertex chain code. Setiap citra karakter diekstrak menggunakan 3 metode tersebut dengan tiga cara yaitu ekstraksi secara global, lokal 3x3, 5x5, dan 7x7. Setelah esktraksi fitur, dilakukan proses klasifikasi menggunakan support vector machine (SVM). Hasil eksperimen menunjukkan akurasi terbaik adalah pada model citra 7x7 dengan nilai akurasi freeman chain code sebesar 99.75%, differential chain code sebesar 99.75%, dan vertex chain code sebesar 98.6%.AbstractThe recognition of handwriting characters images is a difficult problems to be solved, because everyone has a different writing style. The step of handwriting recognition process are preprocessing, feature extraction, and classification. Preprocessing is done to convert handwritten images into binary images that only have 1 pixel thickness by using binarization and thinning. Then, in the feature extraction we select shape feature because it is more important than two other visual features (color and texture) in handwriting character recognition. Shape feature extraction method chosen in this research is chain code method because this method is often used in several studies for handwriting recognition. In this study, a performance study of feature extraction based on chain codes was carried out on handwriting character recognition to know the best chain code method. The three shape feature extraction based on chain code used in this study are freeman, differential and vertex chain codes. Each character image is extracted using these 3 methods in three ways: extraction globally, local 3x3, 5x5, and 7x7. After the extraction feature, the classification process is carried out using the support vector machine (SVM). The experimental results show that the best accuracy is in the 7x7 image model with the value of freeman chain code accuracy of 99.75%, the differential chain code of 99.75%, and the vertex chain code of 98.6%.

Eliminasi Non-Topic Menggunakan Pemodelan Topik untuk Peringkasan Otomatis Data Tweet dengan Konteks Covid-19 Damayanti, Putri; Purwitasari, Diana; Suciati, Nanik
Jurnal Teknologi Informasi dan Ilmu Komputer Vol 8 No 1: Februari 2021
Publisher : Fakultas Ilmu Komputer, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25126/jtiik.0814324

Akun twitter, seperti Suara Surabaya, dapat membantu menyebarkan informasi tentang COVID-19 meskipun ada bahasan lainnya seperti kecelakaan, kemacetan atau topik lain. Peringkasan teks dapat diimplementasikan pada kasus pembacaan data twitter karena banyaknya jumlah tweet yang tersedia, sehingga akan mempermudah dalam memperoleh informasi penting terkini terkait COVID-19. Jumlah variasi bahasan pada teks tweet mengakibatkan hasil ringkasan yang kurang baik. Oleh karena itu dibutuhkan adanya eliminasi tweet yang tidak berkaitan dengan konteks sebelum dilakukan peringkasan. Kontribusi penelitian ini adalah adanya metode pemodelan topik sebagai bagian tahapan dalam serangkaian proses eliminasi data. Metode pemodelan topik sebagai salah satu teknik eliminasi data dapat digunakan dalam berbagai kasus namun pada penelitian ini difokuskan pada COVID-19. Tujuannya adalah untuk mempermudah masyarakat memperoleh informasi terkini secara ringkas. Tahapan yang dilakukan adalah pra-pemrosesan, eliminasi data menggunakan pemodelan topik dan peringkasan otomatis. Penelitian ini menggunakan kombinasi beberapa metode word embedding, pemodelan topik dan peringkasan otomatis sebagai pembanding. Ringkasan diuji menggunakan metode ROUGE dari setiap kombinasi untuk ditemukan kombinasi terbaik dari penelitian ini. Hasil pengujian menunjukkan kombinasi metode Word2Vec, LSI dan TextRank memiliki nilai ROUGE terbaik yaitu 0.67. Sedangkan kombinasi metode TFIDF, LDA dan Okapi BM25 memiliki nilai ROUGE terendah yaitu 0.35. AbstractTwitter accounts, such as Suara Surabaya, can help spread information about COVID-19 even though there are other topics such as accidents, traffic jams or other topics. Text summarization can be implemented in the case of reading Twitter data because of the large number of tweets available, making it easier to obtain the latest important information related to COVID-19. The number of discussion variations in the tweet text results in poor summary results. Therefore, it is necessary to eliminate tweets that are not related to the context before summarization is carried out. The contribution to this research is the topic modeling method as part of a series of data elimination processes. The topic modeling method as a data elimination technique can be used in various cases, but this research focuses on COVID-19. The aim is to make it easier for the public to obtain current information in a concise manner. The steps taken in this study were pre-processing, data elimination using topic modeling and automatic summarization. This study uses a combination of several word embedding methods, topic modeling and automatic summarization as a comparison. The summary is tested using the ROUGE method of each combination to find the best combination of this study. The test results show that the combination of Word2Vec, LSI and TextRank methods has the best ROUGE value, 0.67. While the combination of TFIDF, LDA and Okapi BM25 methods has the lowest ROUGE value, 0.35.

Deteksi Kejadian Lalu Lintas Pada Teks Twitter Dengan Pendekatan Klasifikasi Multi-Label Berbasis Deep Learning Atikah, Luthfi; Purwitasari, Diana; Suciati, Nanik
Jurnal Teknologi Informasi dan Ilmu Komputer Vol 9 No 1: Februari 2022
Publisher : Fakultas Ilmu Komputer, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25126/jtiik.2022915206

Kemacetan merupakan salah satu kejadian yang sering terjadi di kota-kota besar. Hal ini dapat merugikan pengguna jalan, oleh karena itu perlu dilakukan pendeteksian kejadian lalu lintas. Saat ini, twitter digunakan sebagai sumber informasi untuk mendeteksi suatu kejadian. Namun, pengguna twitter cenderung membagikan beberapa informasi sekaligus, sehingga dalam satu tweet bisa memiliki lebih dari satu label. Pada penelitian ini dilakukan klasifikasi multi-label menggunakan 18.000 data dari akun twitter terverifikasi di Surabaya. Klasifikasi multi-label pada penelitian ini dilakukan untuk mengidentifikasi banyak situasi lalu lintas seperti kondisi cuaca, kecelakaan lalu lintas, kemacetan lalu lintas, lalu lintas padat, dan lalu lintas lancar. Klasifikasi dilakukan dengan menggunakan pendekatan deep learning (CNN dan LSTM) dan word embedding (word2vec dan fastText) dengan augmentasi dan non-augmentasi data. Eksperimen dilakukan dengan 3 skenario berbeda untuk melihat pengaruh data uji yang berbeda pada data latih yang sama. Selanjutnya dilakukan eksperimen untuk menguji pengaruh jumlah label terhadap klasifikasi multi-label pada data uji yang sama. Akurasi tertinggi pada non-augmentasi data adalah 0,75 dan pada augmentasi data adalah 0,95. Dari keseluruhan ujicoba akurasi tertinggi diperoleh dari kombinasi LSTM dan fastText. Abstract Congestion is one of the events that often occurs in big cities. This can be detrimental to road users, therefore it is necessary to detect traffic events accurately and efficiently. Currently, Twitter is used as a source of information to detect an incident. However, twitter users tend to share several information at once, so that in one tweet can have more than one label. Therefore, multi-label classification is necessary. This study utilizes 18,000 data from verified twitter accounts in Surabaya. Multi-label classification is carried out to identify many traffic situations, such as weather conditions, events, traffic jams, heavy traffic, and smooth traffic. Classification is performed using deep learning approach (CNN and LSTM) and word embedding (word2vec and fastText) with augmented and non-augmented . Experiments are carried out with 3 different scenarios to see the effect of different uji data on the same train data. Furthermore, the experiments are conducted to examine the effect of the number of labels on the multi-label classification on the same test data. The highest accuracy on non-augmented data is 0,75 and on augmented data is 0,95. All of the highest accuracy is obtained from the combination of LSTM and fastText

Co-Authors Adhira Riyanti Amanda Adni Navastara, Dini Agus Eko Minarno Agus Priyono Agus Zainal Arifin Agus Zainal Arifin Ahmad Saikhu Ahmad Syauqi Ahmad Syauqi Akwila Feliciano Akwila Feliciano Akwila Feliciano Pradiptatmaka Alam Ar Raad Stone Aldinata Rizky Revanda Altriska Izzati Khairunnisa Hermawan Amelia Devi Putri Ariyanto Amirullah Andi Bramantya Andika Rahman Teja Anny Yuniarti Antonius Kevin Wiguna Ardian Yusuf Wicaksono Ari Wijayanti Aris Fanani Arrie Kurniawardhani Arsy Bilahi Tama Ary Mazharuddin Shiddiqi Arya Yudhi Wijaya Atika Faradina Randa Atikah, Luthfi Avin Maulana Awangditama, Bangun Rizki Ayu Kardina Sukmawati Ayu Septya Maulani Baso, Budiman Bryan Nandriawan Bui, Ngoc Dung Chastine Fatichah Chastine Fatichah Chilyatun Nisa' Damayanti, Putri Daniel Sugianto Darlis Herumurti Davin Masasih Diana Purwitasari Dimas Rahman Oetomo Dini Adni Navastara Dini Adni Navastara, Dini Adni Dion Devara Aryasatya Eko Prasetyo Eva Yulia Puspaningrum Evelyn Sierra Fairuuz Azmi Firas Faishal Azka Jellyanto Faizin, Muhammad 'Arif Fajar Astuti Hermawati Fandy Kuncoro Adianto Fandy Kuncoro Adianto Febri Liantoni, Febri Fiqey Indriati Eka Sari Fitri Bimantoro Ginardi, R.V. Hari Glenaya Gou Koutaki Gurat Adillion, Ilham Hafidz, Abdan Handayani Tjandrasa Handayani Tjandrasa Hani Ramadhan Haq, Arinal Hidayat, Ahmad Nur Hidayati, Shintami Chusnul Hilya Tsaniya Imagine Clara Arabella Imam Kuswardayan Imam Mustafa Kamal Irawan Rahardja, Agustinus Aldi Isye Arieshanti Isye Arieshanti Januar Adi Putra Januar Adi Putra Kautsar, Faiz Keiichi Uchimura Kevin Christian Hadinata Kevin Christian Hadinata M. Bahrul Subkhi Maulidan Bagus A.R Maulidiya, Erika Mawaddah, Saniyatul MIFTAHOL ARIFIN, MIFTAHOL Mochammad Zharif Asyam Marzuqi Muchamad Kurniawan Muchamad Kurniawan Muchamad Kurniawan, Muchamad Muhamad Nasir Muhammad 'Arif Faizin Muhammad Alif Satriadhi Muhammad Farih Muhammad Fikri Sunandar Mutmainnah Muchtar Nafa Zulfa Ni Luh Made ITS Novrindah Alvi Hasanah R Dimas Adityo R. Dimas Adityo Rachman, Rudy Rahma Fida Fadhilah Rangga Kusuma Dinata Rangga Kusuma Dinata Rayssa Ravelia Rizal A Saputra Rizal A Saputra, Rizal A Rohman Dijaya Romario Wijaya Safhira Maharani Safhira Maharani Salim Bin Usman Salim Bin Usman Salsabiil Hasanah Sarimuddin, Sarimuddin Septiana, Nuning Sherly Rosa Anggraeni Sherly Rosa Anggraeni Shintami Chusnul Hidayati Shofiya Syidada Sjahrunnisa, Anita Suastika Yulia Riska Sugianela, Yuna Surya Fadli Alamsyah Syavira Tiara Zulkarnain Tanzilal Mustaqim Tiara Anggita Tiara Anggita Tsaniya, Hilya Wahyu Saputra, Vriza Wan Sabrina Mayzura Wibowo, Della Aulia Wicaksono, Farhan Wijayanti Nurul Khotimah Yulia Niza Yulia Niza Yuna Sugianela Yuna Sugianela Yuslena Sari, Yuslena Yuwanda Purnamasari Pasrun Zakiya Azizah Cahyaningtyas Zakiya Azizah Cahyaningtyas

Title

Found 91 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 91 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 91 Documents
Search