Alumni
DONI SUMITO SUKISWO
Class of 2022, AI-016
THESIS:
Development Of Bert and Hybrid Models for Sentiment Analysis Using Acehxfine-Tuning and Tokenizer Adaptation
ABSTRACT :
In the digital era. sentiment analvsis has become one of the kev areas in natural language processing (NLP). NLP development for regional languages in Indonesia remains very limited, including for the Acehnese language which possesses rich lexical diversity and unique morphological structures. One of the main challenges in developing sentiment analysis for Acehnese is the lack of a representative dataset for sentiment analvsis tasks. Moreover, there is currently no BERT-based model utilizing the Masked Language Modeling (MLM) approach that has been specifically optimized for the Acehnese language. Existing pretrained models such as IndoBERT still relv on Indonesian-language data and have yet to fully capture the distinctive linguistic characteristics of Acehnese. Therefore, this studv aims to construct an AcehX Sentiment dataset in the Acehnese language ana develop the AcehXBERT model by re-training the IndoBERT-base model using the MLM approach on the AcehX corpus. This aims to enhance the model’s semantic and contextual understanding of the Acehnese language. The study also investigates the fine-tuning process for sentiment classification tasks on Acehnese text to evaluate the model’s comprehension of local context and its sentiment classification performance. Experimental results on the Acehx dataset using test data show that the AcehXBERT model for sentiment classification successfully achieved an F1- macro of 82.50% and the AcehXBERT+BiLSTM model achieved an F1-macro of 81.62% while for the NusaX dataset using test data, AcehXBERT achieved an F1- macro of 81.89% and AcehXBERT+BiLSTM achieved an F1-macro of 82.29% outperforming the model from NusaBERT. This study shows that an adaptive approach to pre-trained models and tokenizers is very important in the developmen of NLP for regional languages, especially in efforts to support the preservation and utilization of the Acehnese language in modern technology.
Keywords: AcehX, analisis sentimen, NLP, transformers, BERT, deep learning
FARRASA RANI FAISYAL
Class of 2022, AI-015
THESIS:
A Comparative Study of Machine Learning Techniques for Optimizing Near Infrared Spectroscopy (NIRS) Models in Predicting Total Acidity and Vitamin C Content in Mangoes
ABSTRACT :
Mango is one of the most popular fruits, highly favored by the public due to its high vitamin content and significant economic value. As such, evaluating anc identifying fruit quality is a critical aspect of post-harvest processing. The quality of mango fruit can be determined by parameters such as total acidity and vitamin C content. Near Infrared Spectroscopy (NIRS) is a rapid and non-destructive analytical technique that serves as an alternative method for assessing quality parameters. Non-linear methods such as Machine Learning (ML) regressior algorithms are required to enhance data analysis accuracy. This study aims to compare various ML techniques for optimizing NIRS models to predict tota acidity and vitamin C content in mangoes, using spectral correction methods including Multiplicative Scatter Correction (MSC) and Standard Normal Variate (SNV). The best prediction results for total acidity were achieved using the GBR algorithm, with MSC-GBR values (R2: 0.99. r: 0.97. RMSEC: 0.00, RPD: 7.06 and SNV-GBR values (R2: 0.99. r: 0.99. RMSEC: 0.01, RPD: 17.23). The RFR algorithm also provided satisfactory results after GBR. For vitamin C prediction the GBR and RFR algorithms demonstrated the best performance. with MSC GBR values R2: 0.97. r: 0.99. RMSEC: 0.00. RPD: 5.82) and SNV-RFR values (R2: 0.99, R: 0.99, RMSEC: 0.18, RPD: 12.41). On the other hand, the SVR and KNN-R algorithms showed poor performance in predicting total acidity and vitamin C in mangoes. This study successfully demonstrates that NIRS combined with ML techniques can predict total acidity and vitamin C content in mangoes with high accuracy, particularly using GBR and RFR models. The selection of appropriate spectral correction methods and ML algorithms is crucial to achieving optimal results.
Keywords: Mango, Total acidity, Vitamin C, Near Infrared Spectroscopy (NIRS)., regression, SVR, RFR, GBR, KNN-R
M. HAEKAL ALFANSHURY
Class of 2022, AI-014
THESIS:
Exploring Machine Learning Techniques for Enhanced Chlorogenic Acid Quantification in Coffee Beans using Near Infrared Spectroscopy
ABSTRACT :
This research aims to develop a deep learning-based object detection system using the YOLOv11 architecture to identify types of rice sacks: “Udang Premium,” “Udang Kuning,”‘ and “Udang Hitam”. The innovation of this study lies in exploring the impact of two different annotation methods, namely polygon and bounding box, on detection accuracy. The dataset comprised 169 images with 325 annotations, initially exhibiting significant class imbalance, particularly for the “Udang Kuning” class. The model was trained and evaluated using mAP (mAP), Precision, and Recall metrics. Experimenta results demonstrate excellent performance, with mAP@50 reaching 94% on the data validasi and 97% on the data pengujian. Per-class detection showed an Average Precision (AP) of 100% for “Udang Kuning” and “Udang Premium” on the data pengujian, and 91% for “Udang Hitam”. Confusion Matrix analysis confirmed a higl number of True Positives and minimal False Negatives, although some False Positives and one misclassification case were observed for the “Udang Hitam” class. A comparison with other object detection methods, namely RF-DETR (Base) and YOLOv12 (Fast), revealed that YOLOv11 consistently outperformed both in terms of mAP @50 (93.8% vs. 92.2% and 91.6%), Precision (95.8% vs. 95.1% and 93.5%). and Recall (92.6% vs. 81.0% and 89.7%). These findings affirm that YOLOv11 is a highly effective solution for rice sack detection, with the potential to enhance automation, and efficiency in identification and sorting processes within related industries.
Keywords: Object Detection, YOLOv11, Rice Sacks, Deep Learning, Computer Vision.
MUHAMMAD IHSAN
Class of 2021, AI-013
THESIS:
Performance Analysis of YOLOv11 in Rice Sack Detection and Classification
ABSTRACT :
This research aims to develop a deep learning-based object detection system using the YOLOv11 architecture to identify types of rice sacks: “Udang Premium,” “Udang Kuning,”‘ and “Udang Hitam”. The innovation of this study lies in exploring the impact of two different annotation methods, namely polygon and bounding box, on detection accuracy. The dataset comprised 169 images with 325 annotations, initially exhibiting significant class imbalance, particularly for the “Udang Kuning” class. The model was trained and evaluated using mAP (mAP), Precision, and Recall metrics. Experimenta results demonstrate excellent performance, with mAP@50 reaching 94% on the data validasi and 97% on the data pengujian. Per-class detection showed an Average Precision (AP) of 100% for “Udang Kuning” and “Udang Premium” on the data pengujian, and 91% for “Udang Hitam”. Confusion Matrix analysis confirmed a higl number of True Positives and minimal False Negatives, although some False Positives and one misclassification case were observed for the “Udang Hitam” class. A comparison with other object detection methods, namely RF-DETR (Base) and YOLOv12 (Fast), revealed that YOLOv11 consistently outperformed both in terms of mAP @50 (93.8% vs. 92.2% and 91.6%), Precision (95.8% vs. 95.1% and 93.5%). and Recall (92.6% vs. 81.0% and 89.7%). These findings affirm that YOLOv11 is a highly effective solution for rice sack detection, with the potential to enhance automation, and efficiency in identification and sorting processes within related industries.
Keywords: Object Detection, YOLOv11, Rice Sacks, Deep Learning, Computer Vision.
REZA AULIA
Class of 2021, AI-012
THESIS:
Comparison of Machine Learning Algorithm Performance in Detecting Stunting with the Application of Recursive Feature Elimination and Smote
ABSTRACT :
Stunting remains a major global health issue that significantly impacts child growth and development. In Indonesia, the prevalence of stunting is still high, with Aceh Province reporting a rate of 31.2%, which is considerably above the national average of 21.6%. This condition necessitates a data-driven approach to support early detection and more effective nutritional interventions. This study aims to compare the performance of several machine learning algorithms in predicting children’s nutritional status, identify the most influential features affecting nutritional outcomes and develop an accurate and reliable predictive model. The dataset used is derived from the 2022 Indonesian Nutritional Status Survey (SSGI) for Aceh Province. Five machine learning algorithms were employed: Decision Tree, Random Forest, XGBoost, Support Vector Machine (SVM), and Logistic Regression. Data preprocessing involved simple imputation to handle missing values and nominal encoding for categorical variables. Furthermore, Recursive Feature Elimination (RFE) was applied for feature selection, and the Synthetic Minority Oversampling Technique (SMOTE) was utilized to address class imbalance. The experimental results indicate that models combining RFE and SMOTE significantly enhance recall and AUC-ROC performance. Among the evaluated models, the SVM model achieved the best performance with a recall of 96.75% and an AUC-ROC of 92.34%. Feature importance analysis using the SVM model with RFE and SMOTE identified Height/Length and Child Age as the two most influential features, with contribution scores of 0.436 and 0.370, respectively. The proposed predictive model demonstrates strong potential as a decision-support tool for early stunting detection in Aceh Province and can facilitate more targeted nutritional interventions.
Keywords: Stunting, Machine Learning, Recursive Feature Elimination, SMOTE, Nutritional Status Prediction
MARIA ULFA
Class of 2023, AI-011
THESIS:
Determination Of Ripeness Stages and Shelf-Life Estimation of Avocados using YOLOv8, Hybrid Machine Learning, and Additional Local Avocado Datasets
ABSTRACT :
Post-harvest management of ‘Hass’ avocados faces significant challenges due to unpredictable ripening, leading to substantial losses. While recent studies have utilized deep learning for ripening assessment, gaps persist in accuracy, efficiency for resource-constrained devices, and optimal exploitation of internal model features. This research aims to develop and evaluate a system for detecting avocado ripening stages and estimating shelf life using YOLOv8 and hybrid machine learning approaches. The system was adapted to a local avocado dataset, and a mobile application prototype was developed. The dataset comprises avocado images across five ripening stages (unripe, breaking, ripe1, ripe2, and overripe), including additional data from local Central Aceh avocado varieties. The YOLOv8 model was initially trained using the dataset by Xavier and subsequently fine-tuned with local data. Evaluation results indicate that the standalone model accurately classifies ripening stages with a mean Average Precision (mAP) of 0.93 and a classification accuracy of 0.88. The hybrid approach involved extracting features from optimal YOLOv8 layers, followed by Random Forest feature selection, class balancing with SMOTE, and classification using Logistic Regression, SVM, and XGBoost algorithms. Among these, Logistic Regression achieved the highest accuracy at 0.99 Shelf life estimation demonstrated an overall Mean Absolute Error (MAE) of 0.43 days for the hybrid approach and 0.44 days for Y OLOv8, significantly outperforming previous research (0.96 days). Thus, this study successfully developed an effective and more accurate system for avocado detection and estimation of shelf life Implementing the model within a mobile application offers a practical solution that contributes to post-harvest efficiency by helping to reduce losses and improve avocado management.
Keywords: YOLOv8, hybrid model, avocado ripening, shelf-life estimation
MUHAMMAF FADHIL AL AMAL
Class of 2021, AI-010
THESIS:
Hybrid Transformer-RNN Model for Classofication of Indonesian Regional Language
ABSTRACT :
Transformer-based language models have been widely adopted in natural language processing tasks, yet their ability to capture sequential dependencies remains limited-particularly in Indonesia’s regional languages, which are low-resource and morphologically hybrid architecture that integrates rich. This study proposes a NusaBERT with BiLSTM and BiGRU layers on top of Transformer representations, while exploring various pooling strategies (CLS token, last hidden state, mean, and Experiments were conducted on three benchmark datasets-NusaParagraph max) Emotion, Rhetoric. Topic) NusaTranslation (Emotion, Sentiment), and Nusax Sentiment)-for multi-class and cross-lingual classification tasks. Results show that the hybrid models consistently outperform baselines, with the best variant NusaBERTLarge BiGRU, mean pooling, batch size 8) achieving macro FI scores of 77.09% (Emotion), 53.38% (Rhetoric), and 88.81% (Topic) on NusaParagraph, 71.03% (Emotion) and 88.71% (Sentiment on NusaTranslation; and 83.26% on NusaX (average across languages). Additional evaluation on the phenomenon of catastrophic forgetting shows that the hybrid model maintains more stable performance when sequential fine-tuning is applied across languages. These findings demonstrate that combining contextual representations from Transformers with the sequential modeling capabilities of RNNs can improve both performance and robustness in multilingual NLP scenarios, while supporting the development of more inclusive and adaptive language technologies for regional language preservation in Indonesia.
Keywords: Hybrid Model, Transformer, Classification, Catastrophic Forgetting Local Languages, Multilingual Text resource efficiency.
GANDHI RIFAL
Class of 2021, AI-009
THESIS:
Hybrid Smote-Adaboost-Evolutionary Algorithm to Overcome Class Imbalancies and Noisy Attributes in Well Log Data for Lithology Prediction with K-Nearest Neighbor
ABSTRACT:
Lithology prediction using well log data is a critical process in the oil and gas industry to accurately determine geological formations. Various machine learning methods including K-Nearest Neighbor (KNN), Random Forest, and Support Vector Machine have been applied to this task. While these methods offer advantages such as handling multi-class classification, they face significant challenges related to class imbalance and noisy attributes in the dataset. Class imbalance can cause bias toward majority classes, while noisy attributes reduce the model’s ability to detect relevant patterns thereby lowering prediction accuracy. Conventional approaches such as SMOTE (Synthetic Minority Oversampling Technique) to address class imbalance offer generate synthetic data that amplify noisy attributes, increasing the risk of overfitting. Furthermore, boosting methods like AdaBoost strengthen predictions by combining weak models but remain vulnerable to noise in the data. To address these limitations this study proposes the integration of Hybrid SMOTE, AdaBoost, and Evolutionary Algorithm. SMOTE is used to balance class distributions by generating more meaningful synthetic data. The Evolutionary Algorithm is applied for feature selection to minimize noisy attributes, while AdaBoost enhances the model’s robustness against overfitting. The proposed approach was tested on two public datasets, FORCE 2020 and KAGGLE. Experimental results show that this integration significantly improves prediction accuracy, achieving 96.75%, precision 88,83%, recall 84,55%, FI-Score 86,46% on the FORCE 2020 dataset and 76.37%, 70.1 1%, precision 57,06%, recall. 55,05%, F1-Score 55.86% on the KAGGLE dataset, outperforming conventional methods. This study aims to provide an innovative and robust solution to address geological data challenges, improve lithology prediction accuracy, and make substantial contributions to applications in the oil and gas industry
Keywords: Lithology Prediction, Well Log Data, Class Imbalance, Noisy Attributes SMOTE, AdaBoost, Evolutionary Algorithm, K-Nearest Neighbor (KNN), Geological Data Analysis
TEUKU YUSRANSYAH TANDI
Class of 2021, AI-008
THESIS:
Integration of IndoBERT and Machine Learning Features to Improve the Performance of Indonesian Recognizing Textual Entailment
ABSTRACT :
This research aims to develop a Recognizing Textual Entailment (RTE) model in the Indonesian language, named Hybrid-IndoBERT-RTE. The model is designed to address challenges in recognizing textual entailment, which is a critical task in Natural Language Processing (NLP). The architecture of Hybrid-IndoBERT-RTE is built by modifying IndoBERT-large-p1, a language model that has proven effective in various NLP tasks in the Indonesian language. In this modification, the output vectors generated by IndoBERT-large-p1 are combined with machine learning features from a feature rich classifiers, enabling the model to capture richer and deeper information The classification head of this model consists of 1 input layer, 3 hidden layers, 1 dropout layer, and 1 output layer, which are designed to enhance the model’s predictive performance. To test the model’s performance, this research uses the Wiki Revisions Edits Textual Entailment (WRETE) dataset, which consists of 450 data samples, with 300 data samples used for training, 50 for validation, and 100 for testing. Experimental results show that Hybrid-IndoBERT-RTE achieved an F1-score of 85%, indicating that the model has a strong capability in recognizing textual entailment in Indonesian In addition to good performance, the Hybrid-IndoBERT-RTE odel also demonstrates efficiency in computational resource usage. During the training process this model utilized Video Random Access Memory Graphics Processing Unit (VRAM GPU) resources 42 times more efficiently on average compared to IndoBERT-large- pl used in previous IndoNLU research. Moreover, the training time of this model is 44.44 times faster, allowing for quicker experimentation and more iterations. This efficiency is crucial in the context of RTE model development, where saving computational resources and training time can accelerate innovation and fiurther applications.
Keywords: Hybrid-IndoBERT-RTE, Wiki Revisions Edits Textual Entailment (WRETE), IndoBERT-large-p1, machine learning features, computational resource efficiency.
ANDRI DARNIUS
Class of 2021, AI-007
THESIS:
Implementation of Dimensionality Reduction on Word Embeddings Vector Generated by Bidirectional Encoder Representation From Transformers (BERT)
ABSTRACT :
This research focuses on improving the efficiency of complex artificial intelligence models, such as BERT, by applying dimension reduction techniques. The BERT model has millions of parameters, resulting in high computational and memory requirements during training. The approach taken utilizes BERT as a foundation applying whitening or sphering techniques as a dimension reduction method at the feature extraction stage. Two scenarios are evaluated: a standard benchmark using the original BERT and a modified scenario involving BERT feature extraction, whitening techniques (PCA, ZCA, BERT Whitening), and classification using Bi-LSTM or MLP. The AG News dataset, containing news headlines and descriptions with four topic classes. is the main focus of the research. Results show that the model in the modified scenario, which combines BERT features, J. Su whitening, and a Bi-LSTM classifier, provides the best performance in terms of accuracy, F1 score, training time, and Graphics Processing Unit (GPU) memory usage. These findings indicate that whitening dimension reduction can improve text classification efficiency without sacrificing accuracy. This research is expected to expand AI applications in resource- constrained environments by improving the efficiency of complex models like BERT through parameter optimization and dimension reduction.
Link of Publication
Under Review
HARY RACHMAT
Class of 2021, AI-006
THESIS:
Development of Large Language Model to Answer Academic Related Questions at Syiah Kuala University using Fine-Tuning and Retrieval-Augmented Generation Methods
ABSTRACT :
Right now, academic information at Universitas Syiah Kuala (USK) is distributed on a website or summarized in the form of Frequently Asked Questions (FAQ). Information in the form of a website and FAQ is not interactive. Certain information must be searched from the web or FAQ list. Therefore, a more interactive way to get information using a chatbot is needed. Chatbots can be built using a Large Language Model (LLM) such as Mistral 7B. Mistral 7B is a large language model that can be applied to answer questions such as academic information using data collected from universities. The fine-tuning method with the QLoRA and RAG techniques can be used to train the model and retrieve relevant information from external document sources. The results are then evaluated using the ROUGE score. The answers from the USK Mistral 7B model gave results with a score of >0.5 on 15 out of 56 questions using the RAG method, and the fine-tuning method was tested on 20 questions, producing a value with a score of >0.5. Testing was also conducted with different questions that had the same meaning, and response results were obtained with a ROUGE score of 0.4-0.5 from the questions asked. Using the USK Mistral 7B model in a chatbot, academic information at USK can be shared interactively
Keywords: Large Language Model, Fine-tuning, RAG
HUZAIR SAPUTRA
Class of 2022, AI-005
THESIS:
Hyperparameter Tuning Automation in YOLO (Case Study: Corn Leaf Disease)
ABSTRACT :
Corn cultivation is pivotal in Southeast Asia, significantly contributing to regional food sccurity and cconomics. However, leaf discases posc a threat, leading to substantial losses in production and harvest quality. To tackle this issue, artificial intelligence (AI) technology is leveraged for early detection of corn leaf diseases. One effective approach is the use of YOLO (You Only Look Once) based object detection models. This study aims to automate the hyperparameter tuning process in YOLO models for corn leaf disease detection, focusing on improving model performance Through meticulous evaluation utilizing precision, recall, mAP50, and mAP50-95 metrics, the study identifies YOLOv8m and YOLO-NAS-L as top-performing models YOLOv8m excels in mAP50 (98.5%) and mAP50-95 (67.8%), while YOLO-NAS-L demonstrates superior detection capabilities with mAP50 (70.3%) and mAP50-95 (38.9%). These findings underscore the potential of advanced AI-driven detection Systems in revolutionizing crop management, facilitating early disease identification, and enabling prompt preventive measures. By leveraging sophisticated object detection models, farmers can enhance crop yields, mitigate losses due to plant diseases, and boost agricultural productivity. The research lays a solid foundation for developing integrated, scalable disease detection systems, offering crucial support for global food security and farmer welfare.
Keywords: Object detection, YOLOv8, YOLO-NAS, corn leaf disease, hyperparameter tuning
AGA MAULANA
Class of 2022, AI-004
THESIS:
Performance Analysis of Deep Convolutional Neural Networks Architecture for Classification of Severity Score for Atopic Dermatitis Skin Disease
ABSTRACT :
The purpose of this research is to develop a system that iS capable of detecting and recognizing text contained in Indonesian ID Card (e-KTP) images accurately and efficiently. The first stage of this research involves selecting an ideal image from the e-KTP dataset. Furthermore, the pre-processing stage is carried out to cut the edges of the image and combine it with the background image to create a more varied dataset and obtain to generate an e-KTP image mask as training data. The total training data after selecting the ideal image 144 images. The U-Net architecture is the choice as the deep learning method used in this study as an image segmentation process. Meanwhile, for text detection and recognition, the Character-Region Awareness For Text detection (CRAFT) and TRBA framework (TPS-ResNet- BiLSTM-Attention) is used. The testing conducted is assessed based on the accuracy, Dice coefficient, and IoU score for the segmentation process with the percentage test results of the U-Net model obtaining an accuracy of 99.52%. Meanwhile for the text detection and recognition follows the confidence score.
Keyword: e-KTP; U-Net network; CRAFT network; TRBA network; Optical character recognition
TEUKU RIZKY NOVIANDY
Class of 2022, AI-003
THESIS:
Enhancing the Prediction of Inhibitor Activity Against Hepatitis C Virus NS5B Through the Utilization of LightGBM and Bayesian Optimization
ABSTRACT :
This study focuses on developing a predictive model for hepatitis C virus (HCV) NS5B inhibitor activity using the Light Gradient Boosting Machine (LightGBM) algorithm. The primary goal is to enhance the accuracy of inhibitor activity predictions, a crucial step in drug discovery for HCV. The study utilizes a molecular dataset comprising 3011 samples, collected from the ChEMBL database. This dataset is divided into 90% training data and 10% test data, resulting in 1503 compounds in the training process and 168 compounds for testing. The process of selecting molecular descriptors involved several stages, including selection based on variance values, multicollinearity, and Recursive Feature Elimination (RFE), resulting in the 50 most relevant molecular descriptors. The constructed LightGBM model employs Bayesian Optimization for hyperparameter tuning. Efforts to improve the model’s predictive performance involved combining several LightGBM models using a voting approach, with evaluation using the coefficient of determination (R’) and root mean squared error (RMSE). The model that has been built is used to predict test data. The results show an increase in the model’s predictive performance when the three LightGBM models are combined, proven by evaluation on test data which obtained the highest R2 value of 0.760 and the lowest RMSE of 0.637. Model validation was conducted through Y-Scrambling techniques, demonstrating that the model’s performance in predicting HCV NSSB inhibitor activity is based on real relationships and not coincidental. SHAP (SHapley Additive exPlanations) analysis was implemented to understand the contribution of each molecular descriptor to the model’s predictions. This analysis helped identify the most influential molecular descriptors, such as MDEC-33 and SpMaxl_Bh(e), providing insights into the molecular characteristics that play a role in inhibiting HCV NS5B. Utilizing SHAP visualizations, ncluding bar charts and bee swarm plots, offered deeper understanding of the influence of each descriptor on the model’s predictions. The study concludes that the combined approach of multiple LightGBM models, coupled with SHAP analysis, represents a significant advancement in predicting the activity of HCV NS5B inhibitors.
Keywords: QSAR, machine learning, drug discovery, model interpretation
ZAKI FUADI
Class of 2022, AI-002
THESIS:
The Deep Learning-Base Intelligent System for Extracting Information from E-KTP
ABSTRACT :
The purpose of this research is to develop a system that iS capable of detecting and recognizing text contained in Indonesian ID Card (e-KTP) images accurately and efficiently. The first stage of this research involves selecting an ideal image from the e-KTP dataset. Furthermore, the pre-processing stage is carried out to cut the edges of the image and combine it with the background image to create a more varied dataset and obtain to generate an e-KTP image mask as training data. The total training data after selecting the ideal image 144 images. The U-Net architecture is the choice as the deep learning method used in this study as an image segmentation process. Meanwhile, for text detection and recognition, the Character-Region Awareness For Text detection (CRAFT) and TRBA framework (TPS-ResNet- BiLSTM-Attention) is used. The testing conducted is assessed based on the accuracy, Dice coefficient, and IoU score for the segmentation process with the percentage test results of the U-Net model obtaining an accuracy of 99.52%. Meanwhile for the text detection and recognition follows the confidence score.
Keyword: e-KTP; U-Net network; CRAFT network; TRBA network; Optical character recognition
MUHAMMAD CHAIDIR
Class of 2021, AI-001
THESIS:
Face Synthesis Using Modified Hyperstyle Architecture
ABSTRACT :
Face recognition is the most stable and robust biometric technique for identifying and authenticating human faces. However, training face recognition models using a deep learing architecture requires large training images. In addition, labeling face image collections manually is a time-consuming and costly process. Augmenting face images using the HyperStyle architecture can solve these issues. Experimental results obtained using the ResNet-50 architecture trained on a dataset of 44,000 face images (the original FaceScrub dataset combined with the HyperStyle Age and HyperStyle Smile augmented images) demonstrate that the model achieved an F1-score of 79%, therefore outperforming the model trained using the original FaceScrub dataset without HyperStyle augmentation, i.e., F1-score of 63%. ResNet-50 with an FI-score of 82% outperformed other CNN models i.e., VGGNet-1 6 (60%), MobileNet V3 Small (65%). SEResNet18 (81%). HyperStyle modifications on the pre-processing and post- processing achieved promising results on model performance. ResNet-50 model trained using combination of the original FaceScrub dataset and the modified Hyper Style synthesized dataset obtained an F1-Score of 83%. Keywords: Deep Learning, Face Recognition, Data Augmentation, Hyperstyle ResNet-50