Search Articles & Publications

Showing 105 articles found for "Processed"

PREDIKSI JUMLAH PENERIMAAN MAHASISWA BARU DENGAN METODE SINGLE EXPONENTIAL SMOOTHING (STUDI KASUS: AMIK ROYAL KISARAN)

Handoko, Wiwin
Abstract: Abstract: A problem requires a solution to solve it. One of them is by using Prediction (Forcasting). Prediction is used to assess the prediction of conditions in the future. at AMIK Royal Kisaran, when it comes to making&#8230; g lecture schedules often hampered because there is no estimated number of students. The data used in this study is the data history of the last 15 Academic Years, from 2003/2004 to 2017/2018. Then the data is processed with the Single Exponential Smoothing Method. Alpha value 0 <α <1. Single Exponential Smoothing makes a comparison with the alpha value until alpha is found which has the minimum error. To find the value of the error, the MSE (Mean Square Error) method is used. The results of the testing of this method are in the academic year 2018/2019 prediction of the number of students for the Informatics Management Study Program as many as 89 people and for Students for the Computer Engineering Study Program as many as 30 people. The Single Exponential Smoothing method can predict the number of students in the next period. Keywords: Prediction; Number of Students; Single Exponential Smoothing; Alpha Value; MSE   Abstrak: Suatu masalah memerlukan sebuah solusi untuk menyelesaikannya. Salah satunya dengan menggunakan Prediksi (Forcasting). Prediksi digunakan untuk menilai prakiraan keadaan dimasa. di AMIK Royal Kisaran, ketika akan membuat jadwal kuliah sering terhambat karena tidak adanya perkiraan jumlah mahasiswa. Data yang digunakan pada penelitian ini adalah histori data 15 Tahun Akademik terakhir, mulai 2003/2004 sampai dengan 2017/2018. Kemudian data diolah dengan Metode Single Exponential Smoothing. Nilai alpha 0<α<1. Single Exponential Smoothing melakukan perbandingan dengan nilai alpha tersebut sampai ditemukan alpha yang memiliki error paling minimum. Untuk mencari nilai Error digunakan Metode MSE (Mean Square Error). Hasil dari pengujian terhadap metode ini adalah pada Tahun akademik 2018/2019 prediksi jumlah Mahasiswa untuk Program Studi Manajemen Informatika sebanyak 89 orang  dan untuk Mahasiswa untuk Program Studi Teknik Komputer sebanyak 30 orang. Metode Single Exponential Smoothing dapat membantu prediksi jumlah mahasiwa pada satu periode kedepan   Kata kunci: Prediksi; Jumlah Mahasiswa; Single Exponential Smoothing; Nilai Alpha; MSE  

Penerapan Metode K-Means Clustering Dalam Menentukan Predikat Kelulusan Mahasiswa Untuk Menganalisa Kualitas Lulusan

Novita Sari, Venny, Yupianti, Yupianti, Maharani, Dewi
Abstract: Abstract: The increasing number of students who graduated each year causes a lot of student data that need to be processed, causing difficulties in grouping the data. In this research apply Data Mining by using Clustering&#8230; g method to classify the quality of graduate students of Faculty of Computer Science Dehasen University of Bengkulu based on GPA and Study Program. The algorithm used is K-Means Clustering, where the data are grouped based on the same characteristics will be entered into the same group and the data set entered into the group does not overlap. Information displayed in the form of group ?? a group of graduate students who dominate the Study Program, so it is known to the group that has the best graduate quality. The results of this study will assist the University in analyzing the quality of graduated students and the most potential study programs. Software used to help this grouping is Rapid Miner. Keywords: K-Means Clustering, Study Program, Graduate Quality, Rapid Miner   Abstrak: Semakin meningkatnya jumlah mahasiswa yang diluluskan setiap tahunnya menyebabkan banyaknya data mahasiswa yang perlu diolah sehingga menyebabkan kesulitan dalam pengelompokan data tersebut. Pada penelitian ini menerapkan Data Mining dengan menggunakan metode Clustering untuk mengelompokkan kualitas lulusan mahasiswa Fakultas Ilmu Komputer Universitas Dehasen Bengkulu berdasarkan IPK dan Program Studi. Algoritma yang digunakan yaitu K-Means Clustering, dimana data dikelompokkan berdasarkan karakteristik yang sama akan dimasukkan ke dalam kelompok yang sama dan set data yang dimasukkan ke dalam kelompok tidak tumpang tindih. Informasi yang ditampilkan berupa kelompok – kelompok lulusan mahasiswa yang mendominasi Program Studi, sehingga diketahui kelompok yang memiliki kualitas lulusan terbaik. Hasil penelitian ini akan membantu pihak Universitas dalam menganalisa kualitas mahasiswa yang diluluskan dan program studi yang paling berpotensi diminati. Software yang digunakan untuk membantu pengelompokan ini adalah Rapid Miner.   Keyword:  K-Means Clustering, Program Studi, Kualitas Lulusan, Rapid Miner    

Analisis Dengan Metode Klasifikasi Menggunakan Decission Tree Untuk Memprediksi Penentuan Resiko kredit Bank

Syafnur, Afdhal
Abstract: Abstract: There are several facilities in distributing funds to the customer which is owned by Bank Syariah Bukopin. One of them is Kredit Pemilikan Rumah / Housing Loan (mortgage), so far the bank when provides mortgages&#8230; s to customers still uses risk prediction manually in giving credit to customers which is taking up a lot of time and energy especially when the customer reports is further analyzed by the Bank. One technique that can help in predicting the Bank's credit risk determination is Decision Tree which is a technique that is a part of Data Mining techniques to take a decision in the form of a tree. With Decision Tree techniques, it is expected to help the bank to allow faster and easier in predicting the data and getting a conclusion from existing data. One of the ways to predict the data is using Dtreg software. This software only uses data that is in the format of "csv (comma delimited)”, if it is not using the format" csv (comma delimited)", so that the data can not be processed by Dtreg software. When the excel format has been converted to the "csv (comma delimited)" format, the analysis process can be done. Dtreg can generate decision tree, one of them is the result of risk decision from the number of mortgages based on the number of customers.             Keywords: data mining, decision tree     Abstrak: Ada beberapa fasilitas dalam penyaluran dana ke nasabah yang di miliki Bank Syariah Bukopin. Salah satunya Kredit Pemilikan Rumah (KPR), selama ini pihak Bank memberikan KPR ke nasabah masih menggunakan prediksi resiko secara manual dalam meberikan kredit kepada nasabah yang banyak menyita waktu dan tenaga apalagi pada saat laporan nasabah  dianalisa lebih lanjut oleh pihak Bank. Salah satu teknik yang dapat membantu pihak Bank dalam memprediksi Penentuan resiko kredit  adalah teknik Decision Tree yang merupakan bagian dari teknik Data Mining untuk mengambil suatu keputusan dalam bentuk pohon. Dengan teknik Decision Tree diharapkan dapat membantu pihak bank agar  lebih cepat dan mudah dalam memprediksi  data dan  menarik suatu kesimpulan dari data yang ada.Salah satu cara memprediksi data tersebut dengan menggunakan software Dtreg. Pada software  ini data yang digunakan hanya bisa dalam bentuk format “csv (comma delimited), jika tidak menggunakan format “csv (comma delimited)“ maka data tersebut tidak bisa diproses oleh software Dtreg dan selanjutnya jika format excel yang telah dirubah ke format “csv (comma delimited)”, maka akan dapat dilakukan proses analisa. Dtreg dapat menghasilkan pohon keputusan, salah satu nya yaitu hasil keputusan  resiko dari jumlah kredit pemilikan rumah berdasarkan jumlah nasabah.     Kata kunci: data mining, decision tree

Analysis of Liquidity, Solvency, and Working Capital Turnover: Implications for Company Profitability in the Digital Era

Fuadi, Agus, Sulistyorini Wulandari, Dian, Nurhasan, Astrya
Abstract: This study investigates the impact of liquidity, solvency, and working capital turnover on profitability among manufacturing companies in the consumer goods sector for the 2019-2022 period. The research utilized a sample&#8230; of 50 companies based on financial reports from IDX. Using purposive sampling with specific criteria, the final sample included 108 companies. The data was analyzed using multiple linear regression, processed with SPSS 25, following classic assumption tests for normality, multicollinearity, autocorrelation, and heteroscedasticity. The analysis reveals that liquidity and working capital turnover do not significantly affect profitability, while solvency significantly impacts profitability. Overall, liquidity, solvency, and working capital turnover together significantly affect profitability.

QUALITATIVE ANALYSIS OF PORKCINE CONTENT ON NUGGET AND MEATBALLS CIRCULATING IN THE CITY OF MAMUJU

tikirik, wita oyleri
Abstract: Halal food means food that is permitted under Islamic law and meets the requirements, namely that it does not contain any ingredients that are not permitted under Islamic law. The absence of information regarding the halal&#8230; al food contained in food sold in Mamuju City means that it is necessary to carry out research that can provide information regarding the halalness of food products sold by traders in Mamuju City, especially instant food or ready-to-eat food that does not have a halal brand, logo or permit. BPOM (Food and Drug Supervisory Agency). This research aims to detect pork contamination in unbranded nuggets and meatballs sold by producers and traders in Mamuju. The method used was to carry out tests using a Porkcine detection kit on unbranded nugget and meatball food samples. The positive control used was a sample of processed meat containing pork. The test results showed that all food samples of nuggets (3 types) and meatballs (3 types) did not contain pork protein (negative results). This way the food is safe for consumption. This way the food is safe for consumption. Even though in this study the food tested negative contained non-halal food ingredients, consumers should still be careful in choosing the food they want to consume, especially products that do not have the product name, list of ingredients used, net weight or net content, name and address of the party. Who produces or imports, halal for those required, production date and code, expiration information, distribution permit number, and the origin of certain food ingredients

DATA PRIVACY AND SECURITY PROTECTION STRATEGIES IN LIBRARY ELECTRONIC RESOURCES MANAGEMENT

Dahlian Persadha, Pratama, Judijanto, Loso, Susanti, Melly, Kreshna Reza, Heru
Abstract: Security is a crucial aspect in the digital age, especially in the management and protection of information. As the volume of information processed increases, the need to organize knowledge and provide adequate security&#8230; becomes more pressing. This research emphasizes the importance of cybersecurity in the context of digital libraries, which must comply with certain technological and regulatory standards to protect user data and guarantee privacy when accessing electronic resources. Libraries face various challenges in protecting personal data on their electronic resources. This research addresses topics such as user privacy, data encryption, access management, and compliance with privacy laws. By addressing these issues comprehensively, libraries can ensure the protection of user privacy while optimizing the benefits of digital resources in today's information environment. The October 2023 cyberattack by a hacker group known as Rhysida on the British Library's internet information system emphasizes the importance of cybersecurity and data privacy for digital libraries. This research aims to provide insights and solutions to address these challenges, ensuring digital libraries can operate securely and efficiently.

The Influence of Leadership Style on Public Service at the Teluk Ambon District Office, Ambon City

Matakena, Julent Audri, Sahetapy, Petronela, Patty, Julia Theresia
Abstract: This study examines the influence of leadership style on public service at the Teluk Ambon District Office, Ambon City. The research was motivated by practical service problems at the district level, including uneven implementation&#8230; lementation of frontliner service procedures, varying employee discipline, and weak coordination among organizational units. A quantitative survey design was applied to test the causal relationship between leadership style and public service quality. Data were collected from 50 respondents consisting of 24 district employees and 26 community service users through structured questionnaires supported by observation, interviews, and documentation. Leadership style was measured through decision-making ability, motivational ability, subordinate control, and emotional control, whereas public service quality was measured using the SERVQUAL dimensions of tangibles, reliability, responsiveness, assurance, and empathy. The data were processed using IBM SPSS with validity, reliability, normality, simple linear regression, t-test, and coefficient of determination procedures. The findings show that all questionnaire items were valid and reliable. The regression model produced a positive coefficient, a standardized beta of 0.906, and an R Square value of 0.820, indicating that leadership style explained 82.0% of the variation in public service quality. The novelty of this study lies in its focus on district-level public service governance in an archipelagic urban context, where leadership is not only administrative but also coordinative and adaptive. The study implies that stronger leadership supervision, staff arrangement, and service discipline are necessary to improve the consistency of public service delivery.

COMPARATIVE ANALYSIS OF EXPLAINABLE AI USING LIME AND SHAP FOR DIABETES PREDICTION BASED ON LIFESTYLE FACTORS

Ricky Salim, Agung Mulyo Widodo
Abstract: The rapid advancement of artificial intelligence (AI) has significantly impacted the healthcare sector, particularly in supporting the early detection of diabetes; however, many AI models still face challenges due to their&#8230; ir black-box nature, where decision-making processes are not easily understood. This study aims to compare two Explainable Artificial Intelligence (XAI) methods, namely Local Interpretable Model-Agnostic Explanations (LIME) and SHapley Additive Explanations (SHAP), in interpreting the prediction results of an Artificial Neural Network (ANN) model using the Diabetes Health Indicator dataset. Prior to modeling, the data were preprocessed through cleaning and normalization to ensure quality and consistency. The trained ANN model was then analyzed using LIME and SHAP to evaluate the contribution of each feature to the prediction outcomes. The results show that both methods are capable of providing meaningful and interpretable explanations, although SHAP demonstrates more consistent and stable interpretations across the dataset. These findings highlight the importance of integrating XAI techniques to enhance model transparency, thereby increasing trust and supporting more reliable decision-making in clinical settings, particularly for diabetes diagnosis.

PEANUT SHELLING MACHINE USING A PETROL ENGINE

William Laia, Nani Sri Rezeki
Abstract: This study discusses a peanut shelling machine driven by a gasoline-engineered motor . In Indonesia, many processed peanut products require peanuts as a component, such as salted peanuts, peanut oil, peanut paste, peanut&#8230; tofu, and others. Manual peanut shelling is usually inefficient and time-consuming. Therefore, with this machine, which can automatically peel peanuts, it will greatly help save time and increase production capacity. Based on planning calculations for the planned components, the following data was obtained: Gasoline engine used = 5.5 Hp , Pad = 6005 , Belt length to peeler = 65 inches , Belt length to blower = 25 inches , Diameter pulley blower = 4 inches , Pulley diameter peeler = 11 inches , Diameter pulley driver = 2 inches . After analyzing the production/manufacturing costs of this machine, the price for manufacturing 1 peanut peeling machine was Rp4,019,372.24 .

THE EFFECT OF MOTIVATION, DISCIPLINE, AND WORK ENVIRONMENT ON EMPLOYEE PERFORMANCE AT PT. ASURANSI ASEI INDONESIA, MEDAN BRANCH

Dewi Anjani
Abstract: This research intends to try the influence of motivation, work discipline and work environment on employee performance at PT. Asei Indonesia Insurance Medan Branch. In this research, there is a conclusion that the problem&#8230; m is whether motivation, work discipline and work environment have a significant effect on the ability of employees at PT. Ace? Therefore this research was attempted to identify and analyze the effects. Encouragement, Activity Discipline and Activity Areas to the Ability of Employees at PT. Asei Indonesia Insurance Medan Branch. In this research using quantitative research procedures. The population in this research is all employees of the Asei Indonesia insurance industry, agents in the research illustration area, totaling 71 people. The information analysis method used is descriptive analysis of respondents, descriptive analysis of variables, classical assumption experiments, multiple linear regression analysis and assumption experiments (t experiments, F experiments and determinant coefficient experiments). The results of the research processed with the SPSS type 23 program, based on the t experiment proved that motivation, activity discipline and the activity environment have a positive and jointly important effect on employee abilities, the adjusted R square number is 0.735 or 73.5%, which means Employee ability is influenced by motivation, activity discipline and activity area and more than 26.5% can be explained by other factors that were not examined in this research. compared to the results of the previous research of the Great God Kresna Valiant. 2017 the effect of communication, encouragement, sports environment on the happiness of employees' activities in the Denpasar city area proves that there is a positive and important effect on employee activity satisfaction. It can be concluded that the results of this research are in line with the results of previous research by AA Besar Oka Pramadita 2015