Claim Missing Document
Check
Articles

Found 3 Documents
Search
Journal : INFOKUM

Application Of Fuzzy Logic In The Measurement System Of Student Satisfaction Level Towards Lecturers Based On The Fuzzy Infrence Analysis Of The Mamdani, Sugeno And Tsukatomo Method System Yuda Perwira; Risa Kartika Lubis
INFOKUM Vol. 10 No. 1 (2021): Desember, Data Mining, Image Processing, and artificial intelligence
Publisher : Sean Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (640.482 KB)

Abstract

Service to students is an important aspect for the comfort of students at universities in undergoing study process, one of the services that need to be considered is student satisfaction with lecturers, especially in private universities, student satisfaction with lecturers is very influential because it can have an impact on the absorption of knowledge provided to achievement. for students, it is necessary for the right method to measure the level of student satisfaction with lecturers, the purpose of this study is to perform a comparative analysis of the fuzzy Inference system of the Mamdani, Sugeno, and Tsukamoto methods to be able to measure the level of student satisfaction with lecturers accurately and with the right method, The stages of this research method are data collection from STMIK Pelita Nusantara, data identification as the basis for fuzzification formation, fuzzy inference system process with Mamdani, Sugeno, and Tsukamoto methods, then analysis of the three t-methods is carried out. This is to select which method is the most accurate, then carry out system development and system implementation. The results obtained in this study are the Fuzzy Inference System has been successfully created and can be applied to this research, and from the comparison of the 3 methods, it is known for level measurement research. student satisfaction with lecturers, the best method is the Sugeno method, in addition to its high accuracy, determining the constants to be values that match the criteria for the assessment range and also the calculations are not so complicated.
CLUSTERIZATION OF DIABETES HEALTH INDICATORS WITH K-MEANS CLUSTER ALGORITHM Yuda Perwira; Wira Apriani; Ahmad Zein; Sinta Lia Alfaris
INFOKUM Vol. 10 No. 03 (2022): August, Data Mining, Image Processing, and artificial intelligence
Publisher : Sean Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (481.224 KB)

Abstract

Diabetes is one of the most common chronic diseases in the world, affecting millions of people every year and placing a significant financial burden on the economy. Diabetes is a serious chronic disease in which individuals lose the ability to effectively regulate blood glucose levels, and it can lead to decreased quality of life and life expectancy.The purpose of this study is to cluster diabetes health to be able to cluster quickly whether a person has diabetes, or prediabetes or is free from diabetes so that diabetes can be anticipated as early as possible. The data used in this study is the result of a survey from the US Behavioral Risk Factor Surveillance System (BRFSS) in 2015 which contains a net data collection of 253,680 survey responses to the CDC's 2015 BRFSS. The target variable Diabetes_012 has 3 classes. 0 for no diabetes or only during pregnancy, 1 for prediabetes, and 2 for diabetes.The method used in this study is the K-Means Clustering method where this method has been quite successful and is widely used by many researchers to cluster and predict, indicators of a person's diabetes health can be grouped into 3 groups, namely the health of people without diabetes, the health of people with prediabetes. and the health of people with type 2 diabetes, as for the results of the clustering of 2349 data, there are 235 people with health without diabetes, 1816 people with prediabetes health conditions and 298 people with type 2 diabetes.
PREDICTION OF STUDENTS WORKING ACCORDING TO COMPETENCY WITH THE C4.5 ALGORITMA ALGORITHM wira apriani; Yuda Perwira
INFOKUM Vol. 10 No. 4 (2022): October, computer, information and engineering
Publisher : Sean Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (390.043 KB)

Abstract

In this digital era, data is very important to be extracted in order to mine knowledge in it, the resulting data is then used to make preparations, or decisions for the future in order to get the best planning, in universities there are a lot of data that can be mined, one of which is alumni tracer data, alumni tracer data can be used to determine the characteristics of students who work according to their competencies or not by using data mining techniques, the purpose of this study is to determine the characteristics of students who after graduation can work according to their competencies , if these characteristics or criteria can be known well, then the student department can direct students who are actively learning to be able to pursue these criteria so that later these students can work according to their competencies and according to their interests. commitment to produce graduates as superior human resources who work according to their competencies. The stages of this method are collecting data from the alumni tracer questionnaire from STMIK Pelita Nusantara then processing it with data mining techniques, namely data selection, data cleaning, data transformation, then the data is processed by data mining technique classification decision tree algorithm C4.5 then evaluation and simulation with rapid miner to validate manual calculations then build the system and implement the system.