1,721,050 research outputs found
Implementasi Konsep Kecerdasan Buatan dengan Metode Best First Search (BSF) untuk Pembuatan Game Ular Tangga Modifikasi
No full textGames as a medium of entertainment has grown rapidly along with the development
of technology as well. One of the elements that play an important role in a games is
artificial intelligence. With artificial intelligence, expected elements in the games can
act as natural as possible like human being. In this Ular Tangga games, the author
apply the concept of Best First method to get the fastest way to finish. Using the
software Adobe Flash CS3. Expected with the making of this game will give a new
discourse about the game and increase the knowledge of author about how to design a
games.53 HalamanSkripsi Sarjan
Implementasi Konsep Kecerdasan Buatan dengan Metode Best First Search (BSF) untuk Pembuatan Game Ular Tangga Modifikasi
No full textGames as a medium of entertainment has grown rapidly along with the development
of technology as well. One of the elements that play an important role in a games is
artificial intelligence. With artificial intelligence, expected elements in the games can
act as natural as possible like human being. In this Ular Tangga games, the author
apply the concept of Best First method to get the fastest way to finish. Using the
software Adobe Flash CS3. Expected with the making of this game will give a new
discourse about the game and increase the knowledge of author about how to design a
games.53 HalamanSkripsi Sarjan
Analisis dan Implementasi Keamanan E-Mail menggunakan Algoritma sebagai Enkripsi dan Dekripsi pada Mozilla Thunderbird
No full textABSTRACT
The development of Internet applications has progressed very rapidly and one of them
is sending messages electronically, called e-mail. Along with the times, sending
messages via e-mail has increased so that the necessary security systems capable of
maintaining the confidentiality of messages sent via e-mail. Frequent occurrence of
intercepts messages by parties who are not responsible for the author to analyze and
create application add-ons that are capable of maintaining the security of e-mail. The
application made by using the RSA algorithm (Ron Rivest, Adi Shamir, and Leonard
Adleman) to encrypt and decrypt a message in Thunderbird. Thunderbird was chosen
because it is a special application to send and receive messages and open source
applications that always make progress against this application users. In this paper the
author will analyze how the implementation of the RSA algorithm against sending
messages and receiving messages, allowing users to maintain the confidentiality of
messages.62 HalamanSkripsi Sarjan
Perancangan dan Pembuatan Aplikasi Sistem Pakar dengan Menggunakan Metode Backward Chaining untuk Menentukan Interaksi Obat Meloksikam dengan Obat Penyakit Lainnya
No full textIn this era of globalization, everything is can not be separated from technology.
Technological dependence is due to the very rapid development of computer functions
that cover all aspects of life. One of them is pharmacy or medicine. In this day, drugs
– medicine is one of the most widely consumed by humans. Not infrequently there are
frequent cases of mistakes in administering drugs to patients with no little cause of
death. That's why I use the computer advances to create an expert system application
to determining drug interactions. This application is made by using the method of
backward chaining. This system can provide information about drug interactions that
are prescribed by doctors especially for drugs Non Steroid Anti-inflammatory drugs
that Meloksikam with other drug diseases. Hopefully, this application can facilitate
users to understand the interaction of drugs without doing it manually, so that
efficiency of time can be achieved without sacrificing the result, and expected this
application will greatly assist the patient in the healing process because one of the
healing process is to provide the right medicine. This application is creating with PHP
and MySql as a database server.101 HalamanSkripsi Sarjan
Combination of 2DPCA (Two-Dimensional Principal Component Analysis), sPCA (sparse Principal Component Analysis), and Ridge Regression Model for Face Recognition
No full textThe 2DPCA (Two-Dimensional Principal Component Analysis) method, an
extension of the PCA (Principal Component Analysis) method, is commonly used in pattern recognition to represent complex data in lower dimensions. One of its applications is in face recognition, where it encounters the challenge of ensuring
good performance in different shooting conditions while avoiding overfitting to
allow for good generalizations. However, it still faces the issue of overfitting. To
overcome this problem, the authors researched by combining the 2DPCA method with the sPCA (sparse Principal Component Analysis) and the Ridge Regression
Model. In this research, the 2DPCA method performs feature extraction, sPCA selects the most informative features or feature selection, and the Ridge Regression
Model performs regularization. The results demonstrate that combining these three methods in face recognition can overcome overfitting and provide better accuracy than using the 2DPCA conventional, with an accuracy rate of 99.38%.79 PagesTesis Magiste
Prototipe Sistem Monitoring Energi Listrik dari Sedimen Mangrove Berbasis Android dan IoT (Internet Of Things)
No full textElectricity has become a basic need of today's society. Alternative energy sources through mangrove sediments are one way to help communities provide electrical energy. Therefore, the Internet of Things (IoT) allows it to be developed as a voltage and current control system in electrical energy systems so as to maximize efficiency in the use and increase safety of the applied system. The ESP32 system as a control is expected to be integrated with INA219 voltage and current sensor. This test is carried out by measuring the amount of voltage and electric current from the output of mangrove sediment deposits that are integrated on-grid with batteries. The magnitude of the voltage measurement that can be measured by INA219 is 26V and the current is 3.2A with a truth level of up to 8mA. This setting system will be limited to a specific power usage. The flow of electricity from the power supply to the load will only apply if the power is greater than 100 mW which will further disable the flow of power through the relay. The measurement results in this test showed that the system was running with an average error of 0.87%. The INA219 sensor performed in this test is able to provide good values for the measurement of voltage and electric current.68 HalamanSkripsi Sarjan
Development of Mamdani Fuzzy Rule-Based System Using the Gray Wolf Optimizer (GWO) Algorithm for Maritime Safety Prediction
No full textThe Gray Wolf Optimizer (GWO) algorithm is an optimization method inspired by the social hierarchy and hunting behavior of gray wolves. Gray wolves are categorized as alpha, beta, delta, and omega. From the initial population of wolves ‘P’, the one closest to the prey becomes the alpha. In the fuzzy rule development system, the Gray Wolf Optimizer algorithm is used to optimize Mamdani fuzzy rules, contributing to improved accuracy in predicting maritime safety. This is essential to ensure operational safety in maritime navigation by considering various factors that influence sailing conditions. The Mamdani fuzzy system used in this model is capable of handling uncertainty and ambiguity in the data related to factors affecting maritime safety. The Gray Wolf Optimizer algorithm aids in optimizing the fuzzy rules to produce more accurate predictions. Based on a comparison of optimization results, the model developed using the Gray Wolf Optimizer and Mamdani fuzzy system demonstrates better performance than conventional methods in terms of maritime safety prediction accuracy.105 PagesTesis Magiste
Optimalisasi Hyperparameter Tuning pada Algoritma CNN dengan Standard Freeze Layer
Full text linkConvolutional Neural Network is one of neural network commonly used on data image. CNN can be use to detect and identify object in an image. To run the CNN algorithm, it requires hyperparamters that match the target dataset in order to achieve good accuracy results. Currently the process of setting the hyperparameters, including the neural network layer requires reliability and some trial and error and is usually more down to luck than calculation. There are still many problems found in hyperparameter tuning that most of the time is spent on setting up the CNN layer. Therefore we need a standard that can classify datasets based on their complexity. In this study, Ktrain, CNN, and ResNet-50 were used to identify 9 body parts in the form of x-ray images. The CNN model was trained on a dataset containing 874 data. The CNN method is implemented using python language and Ktrain library for 36 data with an accuracy of 89% during testing. These results are very good especially in the tuning process during model training, so the system is expected to become a standard for machine learning developers so that it does not require a lot of tuning in the models they train.72 HalamanSkripsi Sarjan
Implementasi Algoritma Tabu Search Dan Sistem Informasi Geografis (Sig) Dalam Pencarian Halte Bus Trans Mebidang Berbasis Web
No full textThe Trans Mebidang Bus is a bus rapid transit system in the province of North Sumatra which operates in Medan city, Binjai city, and Deli Serdang district. To travel by bus, bus users must wait at the stops that have been provided. However, users often experience difficulties in determining the location of the nearest Trans Mebidang Bus stop according to the user's location, causing users to often miss passing buses when they arrive at their destination stops. Therefore, we need a system that can make it easier for users to find the nearest stop with a route that is closer to the optimum route to the Trans Mebidang Bus Stop by utilizing the Geographical Information System (GIS) using the Google Maps API. This study uses the Haversine Distance method to find the nearest stop from the user's location and the Tabu Search algorithm to find a route that is close to the optimum route from the user's location to the destination stop. Based on the test, it is found that the Haversine Distance method produces a stop that has the closest distance to the user's location, the Tabu Search algorithm produces a route that is close to the optimum route with an average running time of 0.526 ms, and the Snap to Roads feature on the Google Maps API can be implemented for lane creation on Geographical Information Systems (GIS).Bus Trans Mebidang adalah sebuah sistem bus rapid transit di provinsi Sumatera Utara yang beroperasi di kota Medan, kota Binjai, dan kabupaten Deli Serdang. Untuk melakukan perjalanan dengan bus tersebut, pengguna bus harus menunggu di halte-halte yang telah disediakan. Namun, pengguna sering mengalami kesulitan dalam menentukan letak halte Bus Trans Mebidang terdekat yang sesuai dengan lokasi pengguna, sehingga menyebabkan pengguna sering ketinggalan bus yang melintas ketika tiba di halte tujuan. Oleh karena itu dibutuhkan suatu sistem yang dapat mempermudah pengguna dalam mencari halte terdekat dengan rute yang mendekati optimum menuju Halte Bus Trans Mebidang dengan memanfaatkan Sistem Informasi Geografis (SIG) menggunakan Google Maps API. Penelitian ini menggunakan metode Haversine Distance untuk mencari halte terdekat dari lokasi pengguna dan algoritma Tabu Search untuk pencarian rute yang mendekati optimum dari lokasi pengguna menuju halte tujuan. Berdasarkan pengujian, diperoleh metode Haversine Distance menghasilkan halte yang memiliki jarak terdekat dengan lokasi pengguna, algoritma Tabu Search menghasilkan rute yang mendekati optimum dengan running time rata-rata sebesar 0.526 ms, dan fitur Snap to Roads pada Google Maps API dapat diimplementasikan untuk pembuatan jalur pada Sistem Informasi Geografis (SIG).125 HalamanSkripsi Sarjan
Spelling Correction For Non-Word Errors In Indonesian Documents Using The Fasttext Approach
No full textTechnological advancements have influenced document writing, where critical issues such as typographical errors or spelling mistakes are commonly found. Proper document writing relies heavily on the use of correct spelling that adheres to the official language guidelines outlined in the Ejaan Yang Disempurnakan (EYD) based on the Kamus Besar Bahasa Indonesia (KBBI). Therefore, a system capable of checking and automatically correcting spelling errors is essential to address such issues in a document. Most existing spell correction systems are limited to processing a few sentences and require significant processing time. This study develops a spell correction system that implements the FastText approach and Peter Norvig's method, designed for Android-based platforms. A pre-trained FastText model is utilized to generate word candidates in the application of Peter Norvig's method. The Kamus Besar Bahasa Indonesia (KBBI) is employed as the word dataset in this research. The testing results of the developed spell correction system achieved a recall of 79.57%, precision of 91.79%, accuracy of 91.05%, and an F1 score of 85.24%. The system accepts input in the form of Indonesian documents with file extensions Microsoft Word (.docx) and PDF (.pdf).69 PagesSkripsi Sarjan
- …
