Journal of Informatics And Telecommunication Engineering
Not a member yet
373 research outputs found
Sort by
The Effectiveness Of OpenCV Based Face Detection In Low-Light Environments
Get PDFA face detection experiment based on OpenCV is a technology used to identify and find human faces in digital images or videos. This technology uses algorithms and image processing techniques to analyze the pixels in an image or video frame and determine if they contain a human face. The aim of this research is to determine the effectiveness of OpenCV-based face detection using the Viola-Jones algorithm with low-intensity brightness environmental conditions. The research was conducted with light intensity levels of 10 Lux, 30 Lux, and 50 Lux and was carried out using the camera on the ASUS TUF DASH 15 FX517ZC laptop. Data in evaluation using evaluation metrics containing the formula recall, precision, F-Score, and accuracy. The results of the study showed that experiments with higher light intensities up to 50 Lux showed the best level of efficiency according to the accuracy values (99.2%), f-score(0.996), and recall (0.993), so the system is best done with a brightness of 50 Lux. Face detection is affected by the camera that used, in addition, rotation of face is important for the system to detect faces, despite video recording in high light intensity environment
Machine Learning Model for Language Classification: Bag-of-words and Multilayer Perceptron
Get PDFThe availability of data today has become a great asset for research that is used for various purposes such as for machine learning. One of the basic machine learning methods for natural language processing is bag-of-words. The problem in this study is the difficulty in classifying texts because texts still have unstructured characteristics, so this study will apply a model to classify the language of texts. Texts will be placed in four categories, English, Indonesian, German and French. Research was conducted using Bag-of-words and Multilayer Perceptron to solve this supervised machine learning problem. The use of Bag-of-words to perform text representation for simple patterns, easy processing and good performance. On the other hand, a multilayer perceptron has the ability to study complex data patterns in the form of images, text or videos. This study will collect data using text mining techniques, namely crawling Twitter social media as many as 4000 data records. This study produces a model with an accuracy of 98 percent with a loss of 0.14 percent which shows good model performance in classifying languages based on text data
Development of Decision Support Applications for Granting Mortgages using the Fuzzy-Weighted Product method
Get PDFKredit Pemilikan Rumah adalah salah satu fasilitas yang diberikan bank atau Lembaga keuangan lainnya. Kredit pemilikan rumah ini biasanya diperuntukkan bagi masyarakat yang kesulitan dalam membeli rumah. Maka dari itu adanya fasilitas ini diharapkan dapat membantu masyarakat dalam membeli rumah secara kredit. Permasalahan yang terjadi adalah seleksi yang dilakukan oleh analis kredit masih dilakukan secara subyektif sehingga terkadang analis kredit masih kesulitan dalam menentukan layak atau tidaknya nasabah tersebut menerima kredit. Tujuan dari penelitian ini adalah membangun sebuah system informasi pendukung keputusan dengan metode Fuzzy-WP guna membantu analis kredit dalam memutuskan pemberian kredit kepemilikan rumah. Hasil implementasi algoritma menunjukkan bahwa Fuzzy-WP yang diterapkan pada sistem telah sesuai dengan perhitungan aslinya serta program aplikasi dapat menampilkan hasil rekomendasi sesuai dengan kriteria yang telah ditetapkan untuk pemberian kredit bagi calon debitur.
Home Ownership Credit (KPR) is one of the facilities provided by banks or other financial institutions. These home ownership loans are usually reserved for people who have difficulty buying a home. Therefore, the existence of this facility is expected to assist people in buying houses on credit. The problem that occurs is that the selection made by credit analysts is still carried out subjectively so that sometimes credit analysts still have difficulty determining whether a customer is eligible to receive credit or not. The purpose of this study is to build a decision support information system using the Fuzzy-WP method to assist credit analysts in deciding whether to grant housing loans. The results of the implementation of the algorithm show that the Fuzzy-WP applied to the system is in accordance with the initial calculations and the application program can display recommendation results according to predetermined criteria for granting credit to prospective borrowers
Implementation of the X-Means Algorithm on Unemployment Data in West Java
Get PDFThe unemployment rate is a serious issue in many countries, including Indonesia, with impacts that encompass psychological pressure, declining living standards, reduced national income, and diminished workforce quality. High unemployment rates can lead to poverty, resource wastage, and have implications for a country's economic growth. Additionally, inadequate workforce quality is a determining factor in unemployment rates. The objective of this study is to classify unemployment data, specifically in West Java Province, using the X-Means algorithm based on educational levels and districts/cities, in order to examine the fluctuations in unemployment. The research findings indicate fluctuations in the number of unemployed individuals in West Java Province from 2011 to 2022. The low unemployment rate in West Java Province experienced a 66.7% increase in 2014-2015, remained stable in 2017-2019, while the high unemployment rate decreased by 80% in 2014-2015, remained stable in 2017-2019, but increased by 20% in 2020-2022. Furthermore, Bogor Regency consistently recorded the highest unemployment rate in each cluster, with a total of 135,000 unemployed individuals in Bogor Regency in 2017-2019, while the education level with the highest unemployment rate was junior high school graduate
Identification of Floods in Palembang Area Using Fuzzy Logic Method of Mamdani and Sugeno
Get PDFFloods are one of the natural events that often occur, especially in areas that have a large population. Palembang is one of the cities in Sumatra Province which often experiences flooding due to its dense population and lack of infiltration areas. One way to improve flood preparedness is to identify floods by using fuzzy logic. This study aims to identify floods in the Palembang area using fuzzy logic mamdani and Sugeno methods. Flood identification is carried out in 6 stages, namely variable determination, fuzzification, inference, application of fuzzy rules, defuzzification, and evaluation. The variables used in this study are rainfall, temperature, area elevation, and the results of flood identification that have a value of flood or not flood. The results of flood identification using the Mamdani method achieved an accuracy of 82% while the Sugeno method achieved an accuracy of 91%. The results of this study indicate that the mamdani and Sugeno fuzzy logic methods are quite good in identifying floods so that they can improve flood preparedness attitudes
Implementation of Link Failover on Metronet Network PT. Telkom Indonesia (Persero) Based on Ipv4 and OSPF
Get PDFMetro ethernet is a data communication network owned by PT. Telkom Indonesia (Persero) through distribution channels, one of which is Open Shortest Path First (OSPF). As technology develops, IPv4 is used as the core addressing protocol for every communicating network device and has the responsibility of finding the route taken by packets to reach their destination. The problem that occurs in the data communication network is the absence of stable connectivity when there are links between cities experiencing downtime, resulting in several link congestion on the data packet path. To overcome the above problems, this study analyzed IPv4-based metro ethernet networks through the OSPF routing protocol using Graphical Network Simulator-3 (GNS-3). This study implements a failover system as a backup link in the event of a Jakarta-Bogor downlink. Based on the simulation and test results, the failover link shows a success status and functions to minimize network connection breakdowns. This is because the use of OSPF routing isOSPF routing canen the main link is down, so the path will be diverted via the backup link. QoS measurements using a single link obtained an average throughput of 195 Kbps, while dual links obtained an average throughput of 258 Kbps. This is because the distance from the source to the destination router is passed by more single links than dual links. Overall, the measurement shows that having a backup link can increase the stability of the data communication network connection when it is down and tested according to ETSI standard
Business Process Design With BPR Life Cycle in Fertilizer Provision
Get PDFFertilizer is a supporter of soil growth and enrichment so as to increase agricultural production while at the same time improving environmental quality. The use of fertilizer is adjusted to the amount of land needed and the number of seeds purchased. In reality, fertilizer for shallots is still difficult to obtain, and prices are still high above retail prices, making shallot farmers want fertilizer subsidies from the government even though the subsidies provided are still inadequate and uneven. The purpose of this study was to analyze the effect of providing fertilizer on land area, seed needs, and prediction of fertilizer needs for shallots. The results of this analysis resulted in the design of business processes in the supply of fertilizers using the Business Process Reengineering Life Cycle method. Primary data collection was carried out using questionnaires and interviews with shallot farmers. The results showed that the production of fertilizer needs in accordance with the area of land and the number of seeds used. Therefore, efforts to increase the competitiveness of shallots can be carried out by proposing the BPR Life Cycle, so that the proposed framework is expected to be realized in order to increase productivity and build farmer institutions to realize further agricultural development
Analysis of Vsat Gyro performance based on the power level of the antenna receiver on the Pelni ship
Get PDFThe satellite communication system that is used for ship communication in Indonesian waters is expected to transmit real time data at any time under any conditions. However, currently, the Vsat Gyrsoscope antenna that is installed on the Pelni ship still uses the Ku band frequency. This frequency band is very susceptible to rain attenuation. Vsat Gyro will automatically search for satellites that cover it. Only when it is raining and the ship is moving towards the port, is there a decrease in the receiving power level. SNR and power and C/N will drop. Therefore, it is necessary to do an analysis to overcome this problem so that it is not disconnected during the transmission. This study conducted an analytical study to validate the analytical predictions of phasing the vsat power signal level on the pelni ship to maintain communication during sea operations, which are included in TKT 3 (three). The tools used in the measurement are vsat modem, PRTG application, laptop, RJ 45 cable, and Spectrum Analyzer. By optimizing the calculation of the margin link budget, the calculation of Qos and the parameters vsat Eb/No, C/N and SNR when adjusted to the operating conditions of the ship with high rain intensity so that communication can continue to run even when the worst conditions can be maintained properly.Â
Development of Engineering Vocational Choice Systems in Higher Education
Get PDFThe high unemployment rate is a challenge for the world of education because not all graduates succeed in entering the world of work. Higher education is one of the largest contributors to unemployment in Indonesia. Likewise with the objectives of vocational education which are expected to be able to immediately work and create jobs. The problem is that not all majors are according to personality type. Therefore, researchers will develop a system aimed at recommending majors in higher education based on personality types. The method used is the Research and Development (R&D) method using the Four D (4 D) model a. Defining stage (Define); b) Design Stage (Design); c) Development Stage (Develop); d) Stage of Spread (Disseminate). This research was only carried out to the extent of development and resulted in a valid system to be implemented based on the validator's assessment. The data collection instrument consisted of product and linguistic expert validation sheets. This study shows that the validation results have an average percentage of 92.5% with a valid category. Based on the results of the research and discussion, it can be concluded that the system is valid to be used as a reference in selecting major
Enhancing Unbalanced Data Classification with Cross-Validation and Extreme Gradient Boosting: A Comprehensive Analysis
Get PDFAs a novel and efficient ensemble learning algorithm, XGBoost has been widely applied due to its multiple advantages, but its classification effect in cases of data imbalance is often not ideal. Aiming at this problem, efforts were made to optimize XGBoost and the Cross Validation algorithm. The main idea is to combine cross validation and XGBoost on unbalanced data for data processing, and then get the final model based on XGBoost through training. At the same time, optimal parameters are searched and adjusted automatically through optimization algorithms to realize more accurate classification predictions. In the testing phase, the area under the curve (AUC) is used as an evaluation indicator to compare and analyze the classification performance of various sampling methods and algorithm models. The results of the model analysis using AUC are expected to verify the feasibility and effectiveness of the proposed algorithm