Filter by
SISTEM PENDETEKSI KUALITAS FISIK PADA BIJI KOPI SECARA REALTIME MENGGUNAKAN M…
In this global era, the demand for coffee beans has been increasing year by year. The economic value of coffee beans is greatly influenced by the quality of the beans being marketed. Manual assessment of coffee quality by experts or farmers tends to be subjective and time-consuming. Based on these conditions, this study developed software to detect coffee bean quality in real-time using the YOL…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xvii, 73 hlm.; ilus.; tab.; 29 cm.
- Series Title
- -
- Call Number
- T1999962026
PERBANDINGAN DEEP LEARNING CNN DENSENET DENGAN TRANSFORMER PADA KLASIFIKASI C…
Klasifikasi citra makanan merupakan tantangan penting dalam computer vision dengan aplikasi luas pada sistem pencatatan gizi dan pelacakan kalori, namun penelitian yang secara langsung membandingkan arsitektur CNN dan Transformer pada dataset makanan multikelas dengan variasi visual tinggi masih terbatas. Penelitian ini membandingkan dua arsitektur deep learning, yaitu DenseNet121 dan Vision Tr…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xv, 333 hlm.; ilus.; tab.; 29 cm.
- Series Title
- -
- Call Number
- T1990412026
PENERAPAN PARTICLE SWARM OPTIMIZATION UNTUK PROCEDURAL CONTENT GENERATION PAD…
The rice farming sector plays a strategic role in supporting national food security; however, its sustainability is under serious pressure due to the conversion of paddy fields into oil palm plantations. In Banyuasin Regency, the sustainability of rice farming has been increasingly threatened by this land-use conversion. The binary logistic regression analysis shows that farm income is the only…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xvii, 122 hlm.; ilus.; tab.; 29 cm.
- Series Title
- -
- Call Number
- T1985432026
INTERAKSI DIALOG NPC BERBASIS SPEECH RECOGNITION DAN FUZZY STRING MATCHING DE…
This study aims to develop an interactive dialogue system based on Speech Recognition in games, allowing players to interact with Non-Player Characters (NPCs) through real-time voice commands. The method used is Fuzzy String Matching with the Levenshtein Distance algorithm to calculate the similarity between speech-to-text conversion results and predefined system commands. The research stages i…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xvi, 130 hlm.; ilus.; tab.; 29 cm.
- Series Title
- -
- Call Number
- T1989662026
PENERAPAN HAND GESTURE RECOGNITION UNTUK KONTROL GAME MAZE MENGGUNAKAN YOLOV11
Advances in computer vision and artificial intelligence technology are driving innovation in human computer interaction, one example of which is Hand Gesture Recognition. In maze games, control systems generally still use conventional devices such as keyboards, mice, or joysticks, thus not providing a fully natural and interactive gaming experience. Therefore, this study implemented a Hand Gest…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xv, 135 hlm.; ilus.; tab.; 29 cm.
- Series Title
- -
- Call Number
- T1989682026
KLASIFIKASI UJARAN KEBENCIAN BAHASA INDONESIA MENGGUNAKAN INDOBERT DAN CONVOL…
The increasing spread of hate speech on social media, particularly on Twitter (X), poses challenges for manual content moderation, which tends to be ineffective. This condition highlights the need for an automated classification system capable of accurately detecting hate speech. This study aims to develop a hate speech classification model using approach combining IndoBERT and Convolutional Ne…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xvii, 147 hlm.; ilus.; tab.; 29 cm.
- Series Title
- -
- Call Number
- T1988762026
PERBANDINGAN RANDOM MAJORITY UNDERSAMPLING DAN RANDOM OVERSAMPLING PADA ANALI…
Game application reviews, such as those for Roblox, often exhibit significant class imbalance, where positively labeled reviews vastly outnumber negative ones. This study aims to compare the effectiveness of Random Majority Under-sampling (RMU) and Random Over-sampling (ROS) data balancing techniques on the Bidirectional Encoder Representations from Transformers (BERT) architecture for automati…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xiv, 90 hlm.; ilus.; tab.; 29 cm.
- Series Title
- -
- Call Number
- T1990722026
ANALISIS PERBANDINGAN KINERJA CONVOLUTIONAL NEURAL NETWORK DAN LONG SHORT-TER…
The rapid development of digital technology has increased the spread of online gambling promotional content across various digital platforms, making manual moderation less effective. This study aims to develop an online gambling text classification model using Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM). CNN was selected for its ability to recognize important word patte…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- x, 44 hlm.; ill.; tab.; 29 cm.
- Series Title
- -
- Call Number
- T1989422026
KLASIFIKASI PNEUMONIA MELALUI CITRA CHEST X-RAY MENGGUNAKAN CONVOLUTIONAL NEU…
Pneumonia is a lung infection with a high mortality rate, ranking as the fourth leading cause of global deaths in 2020 according to the WHO. In this study, Chest X-Ray images were classified into pneumonia and normal categories using the Convolutional Neural Network (CNN) method with EfficientNetV2-B0, EfficientNetV2-B1, EfficientNetV2-B2, and EfficientNetV2-B3 architectures, which were subsequ…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xiii, 130 hlm.; tab.; 29 cm.
- Series Title
- -
- Call Number
- T1989592026
OPTIMASI PARAMETER SUPPORT VECTOR MACHINE (SVM) BERBASIS GRID SEARCH DENGAN S…
The banking industry faces significant challenges in credit risk management due to the subjectivity of manual analysis. This study aims to develop an objective credit risk classification model using Support Vector Machine (SVM) optimized with Grid Search (GS) and Synthetic Minority Over-sampling Technique (SMOTE) on the German Credit Dataset. Recall is designated as the primary evaluation metri…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xvii, 64 hlm.; ilus.; tab.; 29 cm.
- Series Title
- -
- Call Number
- T1989442026
KLASIFIKASI PENYAKIT JANTUNG BERBASIS SELEKSI FITUR CHI-SQUARE DAN MACHINE LE…
Heart disease is one of the leading causes of death worldwide, making accurate early detection methods essential. Machine learning can be used to classify patients conditions, but data imbalance and feature selection can affect model performance. This study applied Chi-Square feature selection and the Synthetic Minority Over-sampling Technique (SMOTE) to 918 patient records with 11 clinical att…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xviii, 112 hlm.; ilus.; tab, 29 cm
- Series Title
- -
- Call Number
- T1975272026
KLASIFIKASI KOMPOSISI PADA PRODUK KEMASAN MAKANAN UNTUK IDENTIFIKASI ALERGEN …
Eczema is a skin condition that can be triggered by food allergens, requiring sufferers to carefully examine the composition of packaged food products. The complexity of ingredient labels often makes manual identification difficult. This study aims to develop a food composition classification system to detect allergens that trigger eczema using the Convolutional Recurrent Neural Network (CRNN) …
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xix, 106 hlm.; ilus.; tab, 29 cm
- Series Title
- -
- Call Number
- T1973212026
ANALISIS GAME ENGAGEMENT EXPERIENCE DISGASTER MENGGUNAKAN KUESIONER GAME USER…
The success of a game no longer depends solely on its technical functionality, but also on its ability to establish player engagement. This study aims to develop Disgaster 2 as a gas leak disaster simulation game, as well as to measure the game engagement experience using the GUESS-18. The game was developed using the Unity 2022 game engine with the GDLC methodology. Testing was conducted throu…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xiv, 134 hlm.; ilus.; tab.; 29 cm.
- Series Title
- -
- Call Number
- T2019572026
ANALISIS SENTIMEN ULASAN TRIPADVISOR MENGGUNAKAN MODEL MULTILINGUAL XLM-ROBERTA
Sentiment analysis is one of the applications of Natural Language Processing (NLP) used to identify users’ opinions or evaluations toward a particular service or object. This study aims to develop a sentiment analysis system for Tripadvisor reviews using the Transformer-based XLM-RoBERTa model. The dataset used consists of 28.754 Italian-language Tripadvisor reviews classified into two sentim…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xiv, 160 hlm.; ilus.; tab.; 29 cm.
- Series Title
- -
- Call Number
- T2019112026
EVALUASI EFEKTIVITAS DESAIN USER EXPERIENCE PADA GAME DISGASTER DENGAN SYSTEM…
Gas plant leaks are a disaster that can be fatal, not only for factory owners and workers, but also for communities living around industrial areas, making evacuation training for nearby residents essential. One effective, fun, and interactive medium for delivering information is through serious game, which are now widely used by both young people and adults. However, for the message to be conve…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xviii, 195 hlm.; ilus.; tab.; 29 cm.
- Series Title
- -
- Call Number
- T2019132026
IMPLEMENTASI PROXIMAL POLICY OPTIMIZATION (PPO) UNTUK PENGEMBANGAN AGENT MUSU…
Karakter non-pemain (NPC) pada game fighting 2D umumnya masih menggunakan scripted AI yang statis, sehingga mengurangi dinamika tantangan bermain. Oleh karena itu, penelitian ini bertujuan mengimplementasikan algoritma Proximal Policy Optimization (PPO) berbasis Deep Reinforcement Learning guna mengembangkan agent musuh yang adaptif. Simulasi dibangun menggunakan Unity Engine dan ML-Agents Tool…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xvi, 96 hlm.; ilus.; 29 cm
- Series Title
- -
- Call Number
- T1967442026
INSTANCE SEGMENTATION PADA CITRA MIKROSKOPIS TELUR CACING PARASIT MENGGUNAKAN…
Helminthiasis is classified by the World Health Organization (WHO) as one of the Neglected Tropical Diseases (NTDs) due to its wide distribution, yet it often remains a low priority in global health agendas. Control efforts for helminthiasis are generally carried out through manual microscopic examination. However, this method requires specialized expertise, has technical limitations such as lo…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xiv, 110 hlm.; ilus.; tab.; 29 cm.
- Series Title
- -
- Call Number
- T2015982026
KLASTERISASI BANDARA DI INDONESIA BERDASARKAN LALU LINTAS ANGKUTAN UDARA DOME…
Indonesia, as an archipelagic country, is highly dependent on air transportation to support interregional connectivity. Airports play a role in supporting mobility, economic activities, and equitable development. This study aims to cluster airports in Indonesia based on domestic air traffic activity using the K-Means algorithm and to analyze the characteristics and patterns of the clusters. The…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xiv, 175 hlm.; ilus.; tab, 29 cm
- Series Title
- -
- Call Number
- T1965052026
EVALUASI KEGUNAAN APLIKASI VIRTUAL REALITY UNTUK PELATIHAN PEKERJA GERAI MINU…
Indonesia’s rapidly expanding beverage industry demands efficient training methods for workers who must prepare drinks accurately under pressure. Traditional training often faces high costs, limited resources, and reliance on instructors, making it less accessible. This study introduces a Virtual Reality (VR) simulation developed with Unity to provide an immersive, cost-effective training alt…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xix, 123 hlm.; ilus.; tab, 29 cm
- Series Title
- -
- Call Number
- T1968892026
SEGMENTASI CITRA PARU-PARU MENGGUNAKAN ARSITEKTUR CUSTOM U-NET
Manual analysis of COVID-19 chest CT scans is often time-consuming and prone to subjectivity, thereby hindering diagnostic efficiency. Therefore, this study aims to develop an automatic segmentation system using a Custom U-Net architecture with a pre-trained DenseNet-169 encoder to accurately map infection areas, including Ground Glass Opacity (GGO), Consolidation, and Pleural Effusion classes.…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xx, V-20 hlm.; ilus.; 29 cm
- Series Title
- -
- Call Number
- T1961082026
KLASIFIKASI JENIS KASUS DI KEJAKSAAN NEGERI CILEGON MENGGUNAKAN METODE SUPPOR…
The development of text processing technology allows for automated data classification using machine learning methods. In the case classification process, manual text grouping is often time-consuming and error-prone. This study aims to build a text classification system capable of automatically grouping case descriptions into specific categories. The methods used in this study are Support Vecto…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xv, 96 hlm.; ilus.; tab.; 29 cm.
- Series Title
- -
- Call Number
- T2012912026
OPTIMASI PRECISION PADA LESI HISTOPATOLOGI HATI MENGGUNAKAN YOLOV11
Rapid and reliable pathological evaluation is made possible in large part by the automatic detection of liver histological abnormalities. Using YOLOv11 on the Roboflow Liver Disease public dataset, this work aims to maximize precision in the detection of four lesions: ballooning, fibrosis, inflammation, and steatosis. After pre-processing the data with a 640×640 resize, the model was trained u…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xx, 96 hlm.; ilus.; 29 cm
- Series Title
- -
- Call Number
- T1962742026
APLIKASI GENERASI OTOMATIS SPRITE SHEET KARAKTER 2D BERBASIS ZERO123++ UNTUK …
erkembangan pesat kecerdasan buatan, khususnya dalam model generatif, telah membuka peluang besar bagi otomatisasi dalam industri pengembangan game. Namun, pengembang indie dan studio kecil masih menghadapi kendala signifikan dalam pembuatan aset game 2D, terutama sprite sheet karakter, yang secara tradisional memerlukan waktu lama, biaya tinggi, dan upaya manual untuk menghasilkan berbagai sud…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xviii, 145 hlm.; ilus.; 29 cm
- Series Title
- -
- Call Number
- T1960572026
OPTIMASI KERNEL CUDA BERBASIS MEMORY-AWARE UNTUK SELF-ATTENTION PADA INFERENS…
The self-attention mechanism in Large Language Models (LLMs) is computationally intensive and memory-bound, posing significant challenges for inference on consumer-grade hardware. This research proposes a memory-aware optimization of CUDA kernels for the self-attention mechanism within the GPT-2 architecture, integrating Shared Memory Tiling, Streaming Softmax, and Kernel Fusion to minimize glo…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xviii, 68 hlm.; ilus.; 29 cm
- Series Title
- -
- Call Number
- T1955422026
PERBANDINGAN FUZZY TIME SERIES CHEN DAN LEE PADA PREDIKSI JUMLAH KUNJUNGAN PA…
The number of outpatient visits in hospitals often fluctuates, making it difficult for hospital management to plan administrative services, particularly in managing patient queues, as well as the allocation of resources, especially administrative staff, effectively. Therefore, a prediction method is needed to estimate the number of patient visits in the future. This study aims to apply and comp…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xix, 169 hlm.; ilus.; 29 cm
- Series Title
- -
- Call Number
- T1962402026
KLASIFIKASI PNEUMONIA DAN NORMAL BERDASARKAN CITRA X-RAY PARU-PARU MENGGUNAKA…
This study aims to develop a pneumonia classification model on chest X-ray images using a Convolutional Neural Network with the DenseNet-121 architecture for early detection of lung diseases. The dataset consists of 5,856 images divided into training, validation, and testing sets. The research stages include preprocessing, image augmentation, model training using a progressive fine-tuning strat…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xxiv, V-7 hlm.; ilus.; 29 cm
- Series Title
- -
- Call Number
- T1955372026
ANALISIS SENTIMEN BERBASIS TRANSFORMER (INDOBERT) PADA ULASAN PENGGUNA APLIKA…
This study implements and compares the performance of the IndoBERT deep learning model in analyzing the sentiment of Gojek application reviews using 3-class and 5-class schemes. Data was extracted via web scraping, yielding 66,709 raw reviews, which were then preprocessed and balanced into 23,115 (3-class) and 32,899 (5-class) datasets. Training was executed with an 80:10:10 ratio across six sc…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xix, V-9 hlm.; ilus.; 29 cm
- Series Title
- -
- Call Number
- T1955952026
KLASIFIKASI KOMENTAR JUDI ONLINE MENGGUNAKAN INDOBERT
The rapid growth of social media has enabled the misuse of comment sections as a medium for promoting online gambling. The high volume of comments makes manual moderation ineffective, thus requiring an automated classification system. This study aims to develop a classification system for online gambling comments using the IndoBERT model and to evaluate its performance based on accuracy, precis…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xxiii, VI-2 hlm.; ilus.; 29 cm
- Series Title
- -
- Call Number
- T1955942026
ANALISIS SENTIMEN TERHADAP ULASAN APLIKASI DANA MENGGUNAKAN WORD2VEC DAN LONG…
DANA has become one of the most popular e-wallet applications in Indonesia. This development has elicited various reactions and responses from users in the form of reviews. These reviews can provide important insights for improving the quality of the application's services. This study aims to conduct sentiment analysis on user reviews by utilizing Long Short-Term Memory (LSTM) as the model arch…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xvii, VI-2 hlm.; ilus.; 29 cm
- Series Title
- -
- Call Number
- T1955622026
PERBANDINGAN MACHINE LEARNING XGBOOST DAN DEEP LEARNING CNN 1D UNTUK MENDETEK…
Phishing is a form of cyberattack that exploits fraudulent websites to obtain users’ sensitive information. This study aims to compare the performance of the XGBoost machine learning algorithm and CNN 1D deep learning in detecting phishing URLs using tabular URL data. XGBoost was chosen because it is effective in handling tabular data and capable of modeling nonlinear relationships among feat…
- Edition
- -
- ISBN/ISSN
- -
- Collation
- xvi, 61 hlm.; ilus.; 29 cm
- Series Title
- -
- Call Number
- T1961092026
Computer Science, Information & General Works 
Philosophy & Psychology 
Religion 
Social Sciences 
Language 
Pure Science 
Applied Sciences 
Art & Recreation 
Literature 
History & Geography