Colorectal cancer is a leading cause of cancer-related deaths globally, predominately developing from adenomatous polyps. While colonoscopy is the gold standard for polyp detection, it suffers from a miss rate of 26% due to operator fatigue and visual variability. This study aims to develop a real-time Computer-Aided Diagnosis (CADx) system for polyp segmentation using the efficient YOLOv11-Seg…
Job vacancy information is now more accessible to the public. However, this ease of access also has a negative impact, namely the increased ease with which fake job vacancies can be spread, causing harm to job seekers. Therefore, this study aims to develop a model based on Bidirectional Encoder Representation from Transformers (BERT) to classify genuine and fake job vacancies. The dataset used …
The massive growth of digital information, especially in the form of news articles, demands a system that is able to filter important information efficiently. Abstractive text summarization is a strategic solution in summarizing information by producing new sentences that still represent the main content of the source text. This study aims to apply the Bidirectional and Auto-Regressive Transfor…
Security in authentication systems is a crucial aspect of protecting user data and identity. One approach that can be applied is keystroke dynamics, which analyzes typing patterns as a form of behavioral biometric characteristics. This study aims to develop and evaluate a machine learning based classification model using the Random Forest algorithm to identify user typing patterns. The dataset …
This study develops a food image classification system for Indonesian traditional dishes based on Convolutional Neural Network to support the digitalization of internal processes in the culinary sector, particularly for automatic food identification and menu management. The model is built using the EfficientNet family and evaluated on a dataset consisting of 30 classes of Indonesian traditional…
Communication is very important for humans in their daily lives. However, for people with hearing impairments, this can be a problem. Many people do not understand the sign language used by people with hearing impairments. In order to communicate with people with hearing impairments, people must learn sign language. Therefore, a system that predicts sign language movements based on sound can he…
Pneumonia is a lung infection caused by various pathogens and poses a global health threat with a high mortality rate. The World Health Organization (WHO) reports that pneumonia caused approximately 740,180 deaths among children under five years of age in 2019, making early detection essential. Pneumonia is generally diagnosed using chest X-ray images because they are inexpensive, easily access…
The increasing problem of drug abuse has made it difficult for families to determine whether noticeable changes in a person’s behavior are normal or early signs of addiction. To address this challenge, this study develops an addiction-screening system using the Fuzzy Tsukamoto method combined with Particle Swarm Optimization (PSO). The system assesses four main indicators—physical, psycholo…
Artificial intelligence in video games plays an important role in managing the behavior of non-player characters (NPC) to make them appear dynamic and unpredictable. One method often used is the Finite State Machine (FSM), but this method has limitations because the transitions between states are deterministic and not varied. This study aims to design and implement a Probabilistic Finite State …
Intrusion Detection System (IDS) plays a critical role in maintaining network security by identifying abnormal or potentially malicious activities. However, the increasing complexity and volume of network traffic in distributed environments pose challenges for conventional detection systems in terms of accuracy and real-time capability. This study proposes an unsupervised learning-based IDS usi…
Indonesia has a high prevalence of active smokers, reaching 70.6% among adult males in 2019. Smoking is known to cause various physical changes in facial skin, such as accelerated aging and wrinkles. This research aims to develop and test an automated system for classifying smokers based on facial physical characteristics using a deep learning approach. The method used is Transfer Learning, uti…
MSMEs need to understand consumer behavior and improve business efficiency through transaction data analysis. This research focuses on purchasing patterns for chips and crackers products using the Apriori algorithm on transaction data for Rizan 858 Snack MSMEs. The transaction data includes 200 transactions recorded between October 2023 and May 2024. The analysis process was conducted in severa…
Lettuce (Lactuca sativa L.), is a commodity crop that is frequently consumed around the world. During cultivation, lettuce often faces challenges such as diseases that can cause losses. Classification of diseases on lettuce leaves is an important challenge in maintaining the quality and quantity of crop yields.. This study compares the performance of Convolutional Neural Network (CNN) and Visio…
Climate change and air pollution drive the search for environmentally friendly solutions, such as electric cars. Although they offer benefits like reduced emissions and energy efficiency, public acceptance varies depending on factors like price, infrastructure, and environmental awareness. This study employs the Support Vector Machine (SVM) algorithm as a machine learning model and the ADASYN m…
In preparing the subject schedule, it must be done correctly because all teaching and learning activities are between teachers and students. So far, the subject scheduling process at MAN 3 Palembang is carried out manually so that clashes often occur between subjects and teachers who teach can teach in different classes at the same time resulting in the teaching and learning process being sligh…
The increasing number of scientific articles presents a challenge for researchers in quickly accessing relevant information. One of the main challenges is determining efficient keywords. To address this, an automatic keyword extraction system becomes an essential solution, aimed at developing and evaluating methods for keyword extraction to accelerate the search and management of information fr…
The rapid advancement of the internet and information technology has transformed many aspects of life, including communication through social media. Application X, previously known as Twitter, has become a primary platform for sharing information. However, this convenience also presents challenges, such as cyberbullying, a harmful behavior including threats that affect users' mental health. Thi…
Image classification is a major challenge in the digital world, especially in the field of deep learning. so this research develops a classification system using Convolutional Neural Network (CNN) with five architectures namely GoogLeNet (InceptionV3), MobileNet, ResNet50, SqueezeNet, and Visual Geometry Group (VGG16) to classify papaya fruit. With the number of data for ripe papaya 267, unripe…
The availability of Indonesian news articles on the internet has greatly increased, making it more challenging to recognize and categorize news accurately. Therefore, a solution to this issue is to develop a classification system for Indonesian news article categories. This research aims to classify Indonesian news category using fine-tuning on the pre-trained IndoBERT model. The dataset consis…
The development of Ibu Kota Nusantara (IKN) has become a topic of public interest, generating various opinions reflecting societal sentiment. This study aims to analyze public sentiment toward the development of IKN using a fine-tuned IndoBERT-based deep learning model. The dataset was collected from platform X, consisting of 18,264 training data, 2,283 validation data, and 2,283 test data, wit…
Emotion recognition based on Electroencephalogram (EEG) signals has become an intriguing research field in human-computer interaction and medical applications with the potential to provide deep insights into patients' mental conditions. This study aimed to compare the performance of three machine learning algorithms, XGBoost, LightGBM, and Random Forest, in classifying EEG signals for emotion r…
The development of music streaming platforms has created a "paradox of choice" phenomenon, where the abundance of music options makes it difficult for users to find songs that match their preferences. This research develops a music recommendation system using Content-Based Filtering method and Term Frequency-Inverse Document Frequency (TF-IDF) algorithm to address this issue. The study aims to …
The classification of undergraduate thesis documents is a crucial process for organizing and categorizing student research into appropriate fields of study. This research develops an ontology-based document classification system for final projects by integrating the FastText word embedding method to improve classification accuracy. The system utilizes BidangIlmuInformatika.owl ontology as a kno…
"Timun Mas Adventure: Learn & Leap" adalah permainan edukasi bergenre Endless Runner. Dalam gameplay, pemain menyelesaikan objektif dalam permainan ini memanfaatkan agen cerdas untuk NPC. Selain sebagai hiburan, game ini mendidik masyarakat tentang cerita rakyat Timun Mas. Penelitian ini menggunakan Finite State Machine (FSM) untuk menguji interaksi agen cerdas dengan pemain. Hasil pengujian me…
SMS Spam sangat membahayakan sehingga dapat menyebabkan kerugian bagi pengguna layanan SMS. Untuk mengatasi hal tersebut, dibutuhkan metode yang dapat membantu mengelompokan SMS sesuai dengan ketegorinya, yaitu Spam dan non spam. Klasifikasi merupakan salah satu proses mengelompokkan data kedalam kelas yang telah ditentukan sebelumnya. Pengklasifikasian melewati beberapa tahapan yaitu, pra peng…
Customer segmentation is a significant application of data analysis in business. This research uses the K-Means algorithm to group customer data based on transaction habits, with parameters from the dataset as cluster determinants. To determine the optimal number of clusters, the Elbow Method is applied, which is based on the highest difference of inertia values. The results show that 2 cluster…
A Question Answering System (QAS) is an automated system designed to provide accurate answers to questions posed using natural language. This research aims to determine the ability of the Indonesian QAS built to perform answer finding or predict the answers submitted to the system using regular expression and cosine similarity methods. The steps taken include filtering the system's documents th…
ATechnological advances drive the need for accurate automatic classification systems, such as in flower type identification, which is still often done manually. This study aims to measure the performance of the DenseNet201 architecture in classifying five types of flowers (Daisy, Dandelion, Rose, Sunflower, and Tulip) using a Convolutional Neural Network (CNN). The model was developed with a fu…
Stroke is a leading cause of disability and death worldwide, making accurate early diagnosis essential to reduce long-term impacts. This study aims to implement K-Nearest Neighbors (KNN) for Stroke diagnosis Classification and to optimize the value of K using Particle Swarm Optimization (PSO). The dataset was obtained from Kaggle, consisting of 5,110 entries and 12 demographic and clinical feat…
The rapid growth of Indonesian online news articles presents challenges in clustering based on content similarity. This study aims to develop a clustering system for online news articles using the k-medoids method. A total of 500 articles from Kaggle were processed through text preprocessing (case folding, tokenizing, stopword removal, and stemming), TF-IDF weighting, and clustering with k-medo…