Gradient Descent (GD) is used to find the local minimum value, its purpose is to find variables on the errorfunction so that a function can model the data with minimum error. Therefore, the purpose of this researchis to see how much iteration is needed and how big is the accuracy level in predicting the data when usingGradient Descent (GD) Standard and GD With Momentum and Adaptive Learning Rate (GDMALR)functions. In this study, the data to be processed using the gradient descent function is the data of SchoolParticipation Rate (SPR) in Indonesia aged 19-24 years, which began in 2011 to 2017. The reason forselection This age range is one of the factors that determine success education in a country, especiallyIndonesia. SPR is known as one of the indicators of successful development of education services in an areaof either Province, Regency or City in Indonesia. The higher the value of SPR, then the area is consideredsuccessful in providing access to education services. SPR data are taken from Indonesian Central Bureau ofStatistics. This study uses 3 models of network architecture, namely: 5-5-1, 5-15-1 and 5-25-1. From 3models, the best model is 5-5-1 with epoch 6202 iteration, 94% accuracy and MSE 0.0008658637. Thismodel is then used to predict SPR in Indonesia for the next 3 years (2018-2020). These results will beexpected to help the Indonesian government to further improve the scholarship and improve the quality ofeducation in the future
Pematangsiantar is a developing city in North Sumatra, so it is not surprising that there are a lot of freight forwarding partners that offer goods delivery services to peoples such as POS Indonesia, JNE, JET Express, J & T Express and SiCepat Express. Many factors that influence people, especially online shop users on the selection of freight forwarding companies, such as the length of delivery, price, type of fleet, range, to company experience. Then it is necessary to build a system that helps the community in taking a decision. The model used to build this Decision Support System uses the Multi-Objective Optimization On The Base Of Ratio Analysis Method which is expected to provide convenience for people in the selection of cellular operators that they will use.Keywords: Decision Support System, MOORA, Online Shop, delivery services
The purpose of the research is to classify the concept of understanding students in Mathematics lessons. In the learning process teaching students understanding learning materials is very important. The attainment of student understanding is a function of the being of an educator. Many formulas and concepts to understand make it difficult for students to solve math problems. The data source was obtained from the results of a math comprehension questionnaire of eighth graders at Tamansiswa Tapian Dolok Private Junior High School. The classification method used is the C4.5 Algorithm and assisted with RapidMiner software. Attributes used are student interests, how students learn, student motivation, how to teach teachers, learning media, and infrastructure facilities. The results of the calculation of entropy values and attribute gains obtained 15 rules of mathematical comprehension decisions with 9 rules of understanding status and 6 rules of inconsistency status. Classification modeling with C4.5 Algorithm on RapidMiner obtained 96.00% accuracy Classification with C4.5 Algorithm can be applied and provide new information about the classification of student comprehension concepts in math lessons
Leasing companies are financing equipment or capital goods that are used for the production process of a company, both directly and indirectly. Financing through leasing is a very simple financing in the procedures and implementation and therefore leasing that is used as an alternative payment seems more attractive. As an alternative source of capital financing for companies, especially in the credit analyst section. In credit-worthy consumer decision-making activities, a computer-based system model that can provide data analysis, calculation of criteria for credit applicants, and help processing data into information to make decisions from the semi-structured problem is needed. A decision system (DSS) is the right choice to help select credit applicants. The system is designed with the Elimination and Choice Translating Reality (ELECTRE) method which is one of the multi criteria calculation methods based on the outrangking concept map by using paired comparisons of other alternatives based on each criterion. The data used in this paper are in the form of alternative data samples, criteria data, criteria weight, and suitability rating for each alternative to the criteria.Keywords: ELECTRE, Credit Analyst ,Kredit, Leasing
Teak is one kind of plant that is already widely known and developed by the wider community in the form of plantations and community forests. This is because until now Teak wood is a commodity of luxury, high quality, the price is expensive, and high economic value. Expert systems are a part of the method sciences artificial intelligence to make an application program disease diagnosis teak computerized seek to replace and mimic the reasoning process of an expert or experts in solving the problem specification that can be said to be a duplicate from an expert because science knowledge is stored inside a database Expert System for the diagnosis of disease teak using forward chaining method aims to explore the characteristics shown in the form of questions in order to diagnose the disease teak with web-based software. Device keel expert system can recognize the disease after consulting identity by answering some of the questions presented by the application of expert systems and can infer some kind of disease in plants teak. Data disease known customize rules (rules) are made to match the characteristics of teak disease and provide treatment solutions.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.