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…
Aplikasi ini dirancang untuk mengembangkan sistem pengelolaan manajemen kerja praktik di Perumda Tirta Musi Palembang agar memudahkan perusahaan dalam mengelola laporan manajemen kerja praktik lebih efisien untuk mengatasi permasalahan yang muncul pada proses pengelolaan manajemen kerja praktik yang dilakukan secara manual. Metode yang digunakan dalam pengembangan aplikasi ini adalah metode Wat…
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…
Sales evaluation application is intended to increase efficiency in terms of making sales reports. Therefore it is necessary to conduct research to develop sales evaluation applications. The purpose of this application is to provide convenience to employees and leaders of the echo store in making reports and being able to find out sales transaction targets. This research was conducted using obse…
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…
The author's goal in building an Outpatient Information System at Stikes Al-Su'Aibah Palembang is to make it easier for patients to register and administer outpatient registration. In addition, the system will make it easier for administrators to register and manage outpatient data. Currently, the waterfall model is the most common software development model. This development model uses a seque…
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…
This paper presents a coffee aroma detection system based on the Internet of Things (IoT) using gas sensors (MQ-3, MQ-4, MQ-5, MQ-135) and a DHT-11 humidity sensor. Data collection was conducted at the Robotics, Control, and Embedded Systems Laboratory, Faculty of Computer Science, Sriwijaya University, involving coffee aroma, temperature, and humidity readings. The gas sensors provided reading…
In today's modern era, the development of information technology is increasingly sophisticated and fast, no less in the medical world, technology is vert much needed, At the Al-Suaibah clinic, in managing daily operations, the As-Suaibah Clinic faces several challenges, including data management, patient medical record management, Therefore a patient data collection and reporting system was bui…
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…
Digital images are one of the most used for various applications that is always in our everyday life. These images always come with noises, an anomaly that distorts the image further from its original condition. This could be fatal to some of the everyday life usage that humans do, such as medical conditions. The method in this study, Non-Local Means will be used to restore these images to at l…
Tuberculosis (TB) is one of the world's deadliest infectious diseases, affecting millions of people each year. Rapid and accurate diagnosis of TB is essential for effective treatment and control of the spread of the disease. This study develops a TB classification model from lung X-ray images using the Convolutional Neural Network (CNN) method with MobileNetV2 architecture. The dataset used con…
Face detection is the basis of the recognition process used in various applications such as security, biometrics, and law enforcement. Multi-face detection becomes relevant for detecting face more than one at a time. The challenge in developing a multi-face detection system involves the complexity of managing several special conditions, one of which is the condition of different face orientatio…
Penyakit mata katarak merupakan salah satu penyebab utama kebutaan di dunia. Terbatasnya pengetahuan dan kewaspadaan masyarakat terkait penyakit ini, mengakibatkan semakin tingginya tingkat pengidap gangguan pengelihatan. Tujuan dari penelitian ini adalah untuk membangun sistem yang dapat mendeteksi penyakit mata katarak menggunakan Convolutional Neural Network (CNN). Diharapkan sistem ini dapa…
The implementation of the Motion Graphic Logo as the Visual Identity of the Rumah Tahfidz Amanah is a solution that can meet the needs of the owner of the Rumah Tahfidz Amanah and become a benchmark for other Rumah Tahfidz. This logo was repaired and modified or updated into an animation to create and recognize the visual identity of Rumah Tahfidz Amanah. The methods used in making this logo in…
Seramed Pharmacy in its daily activities still uses conventional systems, especially in recording transaction data, both incoming goods and sales data. So there are shortcomings, for example the potential for errors to occur in recording and calculating transaction data. The aim of this research is to assist in website-based applications for managing drug stock at the Seramed Palembang Pharmacy…
Tripadvisor is a website that provides information about restaurants, user-generated restaurant reviews.data in the form of comment text. However, there are many features in review makes textual data ambiguous, thus causing difficulties sentiment analysis. To overcome this challenge, this research uses Information Gain feature selection method to reduce high features dimensions in sentiment ana…
This study aims to enhance the dataset of lung cancer histopathological images. There is already exist CNN architecture for lung cancer classification, but only have accuracy of 71%. The dataset of lung cancer histopathological images suffers from poor color constancy and contrast due to variability in image capture, management, analysis, and data collection processes. To address these issues, …
The development and use of information technology is very rapid and plays a very important role in supporting human activities in terms of optimizing time. However, there are still institutions that have not maximized technological advances, one of which is at PT Ricky Kencana Sukses Mandiri in managing the performance of employee leave applications that are still manual such as filling out for…
Berdasarkan hasil observasi penulis pada PT Pertamina (Persero) Kertapati Palembang, penulis menjumpai bahwa proses pendataan pegawai yang berprestasi masih dilakukan secara terpisah-pisah pada file dan buku catatan, sehingga proses pengelolaan dan pencarian data menjadi lebih sulit. Adapun proses pendataan pegawai yang berprestasi dilakukan menggunakan kertas, formulir, atau spreadsheet, yang …
Pathfinding in games presents a complex challenge, particularly when the game environment is dynamic and unpredictable. The D* Lite algorithm is an advanced solution that offers advantages in efficiently finding optimal paths in environments that frequently change. Unlike conventional pathfinding algorithms, D* Lite can quickly update previously computed paths when changes occur in the map, suc…