Claim Missing Document
Check
Articles

Found 12 Documents
Search

Identifikasi Karakter Siswa Menggunakan Metode K-Means (Studi Kasus Sdn 156 Pekanbaru) Kasini Kasini
Jurnal Inovasi Teknik Informatika Vol. 1 No. 1 (2018): Maret 2018
Publisher : Universitas Pahlawan Tuanku Tambusai

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (634.063 KB)

Abstract

Good character education can have a characteristic impact on students. each student has a different character. Various ways done by the school in character education based on kemendiknas, including State Elementary School 156 Pekanbaru. Problems that arise in the field is there is no method that can determine the character of the students so that the school's special teachers can not understand precisely the characters in the students. The lack of understanding of the character of the students makes the vision of the school mission has not been seen so that character education in SDN 156 Pekanbaru has not been right target. Therefore, it needs to be done grouping student character in SDN 156 Pekanbaru with the aim of school know character owned by students in school. The K-Means algorithm is used to classify the character of the students with the number of clusters as much as 2 using the six attributes of characters studied: Honest, disciplined, confident, caring, creative and responsible with 130 student data. The results of K-Means manual calculation with sample data 10 data from 130 data that is weak character (C1) amounted to 1 student and weak character of 9 students, this result is same with calculation executed by RapidMiner application. Test results with 130 data using RapidMiner resulted in the number of students with weak character 26 students with the average centroid (0.665) with caring and creative characters. While students who have strong character 104 students with average value of centroid (0.900) with honest character, discipline, confidence, and responsibility. The result of character grouping based on class cluster position in RapidMiner is grade 3 which has weak character (C1) 8 students from 35 students, grade 4 is 8 out of 24 students, 5th grade is 1 of 17 students and grade 6 is 9 of 46 students. While clusters with strong characters (C2) class 3 amounted to 27 students, grade 4 amounted to 24 students, class 5 amounted to 16 students, and grade 6 amounted to 37 students. From the results of this study is expected Strong characters can be developed by school continue to perform habits which involves the students so that the characters in the students can be seen while for the caring and creative characters so as not to be weak then the school always provide guidance to the students and give examples of good habits and activities that can be followed by students in school .
Clasterization Of Zeeida Product Sales Using K-Means Method In Medan Distributors Nani Hidayati; Kasini; Sabrina Aulia Rahmah
Jurnal Mantik Vol. 6 No. 2 (2022): August: Manajemen, Teknologi Informatika dan Komunikasi (Mantik)
Publisher : Institute of Computer Science (IOCS)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35335/mantik.v6i2.2545

Abstract

Product clustering is one of the determinants of product development in sales activities. Zeeida Herbal products are engaged in health and beauty, of all the products sold, not all of them are sold, some are less well sold. Sales at the distributor of zeeida Medan products are still not computerized, namely by using manual recording. Every buyer who purchases either an agent, sub-agent, reseller or general customer who makes purchases through social media such as WhatsApp, Facebook, marketplace, and other E-Commerce is recorded in the manual bookkeeping, so there is often stock accumulation and even stock shortages at distributors. In this study, the authors apply the k-means clustering algorithm to classify products that do not sell (C0), sell very well (C1) and sell (C2). Clustering is a technique of one of the data mining functionality, the Clustering Algorithm is an algorithm for grouping a number of data into a certain data group (cluster). From this study, the output generated from the last 4 months, namely January-April 2022, shows that from 47 Zeeida products, sales of Zeeida products did not sell well in cluster 0, there were 39 products, while sales were very good in cluster (C1), there were 4 products and sales were sold in cluster. (c2) there are 4 products.
Perbandingan Algoritma K-Means dan K-Medoids untuk Pengelompokkan Daerah Rawan Tanah Longsor Pada Provinsi Jawa Barat: Comparison of K-Means and K-Medoids Algorithms for Grouping Landslide Prone Areas in West Java Province Mufidah Herviany; Saleha Putri Delima; Triyana Nurhidayah; Kasini Kasini
MALCOM: Indonesian Journal of Machine Learning and Computer Science Vol. 1 No. 1 (2021): MALCOM April 2021
Publisher : Institut Riset dan Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (616.781 KB) | DOI: 10.57152/malcom.v1i1.60

Abstract

Bencana alam ialah insiden yang tidak bisa dihindari. Tetapi akibat dari bencana bisa dikurangi dengan mengidintifikasi pemicu terjadinya bencana serta mengkaji peristiwa bencana yang sudah pernah terjadi melalui analisa data bencana yang ada. Indonesia sering mengalami bencana yang disebabkan oleh kerusakan alam akibat perbuatan manusia seperti bencana banjir dan tanah longsor. Berdasarkan data dari Jabar Open Data pada periode 2019, provinsi jawa barat mengalami 609 peristiwa tanah longsor. Badan Penanggulangan Bencana Daerah (BPBD) belum dapat mengoptimalkan pelayanan terhadap korban bencana, misalnya lamanya datang bantuan karena terbatasnya peralatan dan makanan pada daerah bencana. Sedangkan dengan adanya pemetaan resiko bencana menjadi sangat penting dalam penataan penanggulangan bencana yang terarah dan tepat. Maka diperlukannya pengolahan data untuk mengetahui daerah kabupaten/kota yang sering terjadi bencana tanah longsor. Peneliti menggunakan pengolahan data dengan metode perbandingan algoritma K-Means dan K-Medoids. Metode yang didapatkan dari pengelompokkan dengan method K-Means lebih optimal daripada mengguanakan method K-Medoids pada data kejadian tanah longsor Provinsi Jawa Barat pada tahun 2019 dengan jumlah k paling optimal adalah k = 6. Perolehan cluster dominan, menunjukkan bahwa kluster 2 merupakan kluster dengan jumlah daerah paling banyak. Dan jumlah kejadian terbanyak terletak pada kluster 5 dengan jumlah 4 daerah dan jumlah kejadian sebanyak 106 kejadian.
Classification of books at SMP YPK Pematang Siantar using the k-means clustering method Nani Hidayati; Kasini Kasini
Jurnal Mantik Vol. 7 No. 2 (2023): Agustus: Manajemen, Teknologi Informatika dan Komunikasi (Mantik)
Publisher : Institute of Computer Science (IOCS)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35335/mantik.v7i2.3846

Abstract

The school library is a very important facility in supporting the process of improving the quality of education to produce quality young people. The YPK Pematang Siantar Private Middle School Library has ± 200 book titles in several categories, so that these books can be used optimally there must be a system that regulates the number of book stocks, the number of book loans each month, so that it can be seen which student's reading interest is the most popular in each category . YPK Pematang Siantar Middle School has not implemented an optimal computerized system or everything is still manual. By applying grouping of students' reading interest using the clustering method at SMP YPK Pematang Siantar, it is hoped that the process in the library will be more effective, fast, and precise. Clustering is the most suitable method for optimizing library services. The purpose of this research is to classify which category of books YPK SMP students are most interested in. After calculating the 20 book categories for 3 months, the final result is C1 or the most popular, namely the 3 book categories most in demand. most interested (C2) with 8 book categories, and finally C3 which is less desirable there are 9 book categories. By creating clusters of books which are the most desirable and not desirable, it can improve library services and students' interest in reading and also prevent accumulation of books that are not of interest every year
Penerapan Data Mining Untuk Clustering Pada Toko Laura Grosir Dan Eceran Menggunakan Algoritma K-Means Kasini; Nani Hidayati
JUSTER : Jurnal Sains dan Terapan Vol. 2 No. 3 (2023): JUSTER: Jurnal Sains dan Terapan
Publisher : Jompa Research and Development

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Perkembangan teknologi dan informasi di era 5.0 membuat toko-toko grosir dan ecer di tuntut untuk mengikutin perkembangan teknologi agar usaha mereka lebih berkembang dan dikenal tidak hanya dikenal di sekitar toko saja, Toko Laura adalah sebuah toko grosir dan ecer yang menjual kebutuhan primer, kebutuhan sekunder. Banyaknya produk yang dijual oleh toko laura membuat terjadinya penumpukkan stok, karena ada produk yang laris terjual, laris dan kurang laris terjual. Data- data yang terdapat di Toko Laura ini tidak tersusun dengan baik seperti data penjualan, data pembelian dan pengeluaran tak terduga hanya sebagai arsip toko, sehingga tidak dapat dijadikan sebagai pengembangan strategi pemasaran toko. Oleh karena itu, perlu diterapkan data mining menggunakan metode K-Means pada Toko Laura. Penerapan metode K-Means dapat diterapkan pada Toko Laura untuk menentukan produk mana saja yang sangat laris, laris dan kurang laris. Algoritma K-Means metode Clustering diolah menggunakan software RapidMiner. Setelah menghitung 75 data yang ada di toko laura ecer dan grosir selama periode 3 bulan dan kemudian dimasukkan ke dalam aplikasi RapidMiner dan dilakukan pengujiam sesuai dengan pengelompokan 3 cluster sehingga didapatkan hasil pengelompokkan barang sesuai dengan perhitungan yang dilakukan secara manual yaitu kategori laris (C1) 3 barang, Laris (C2) 9 Barang dan kurang laris (C3) 63 barang.
Penerapan Sistem Inferensi Fuzzy untuk Menentukan Jumlah Pembelian Produk Berdasarkan Data Persediaan dan Penjualan dengan Menggunakan Metode Mamdani (Studi: Kasus RM Habibi) Nani Hidayati; Kasini Kasini; Aprilia Permata
Jurnal Teknik Industri Terintegrasi (JUTIN) Vol. 7 No. 3 (2024): July
Publisher : LPPM Universitas Pahlawan Tuanku Tambusai

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/jutin.v7i3.30604

Abstract

RM Habibi is a restaurant business that serves dishes to buyers and provides a place to enjoy the meal, as well as determining food and service costs. Although restaurants generally serve on site, there are also those that provide take away and delivery services as a form of service to buyers. Restaurants usually specialize in the types of food they provide and are their best sellers. Inventory problems are a problem that buyers always face. Inventory is needed because demand patterns are basically irregular. Inventory is carried out to ensure certainty that when needed the product is available. Fuzzy logic is one of the components that make up soft computing.  The application of the fuzzy inference system (FIS) with the mamdani method in the Demand, Supply and Sales prediction system can be concluded that the use of fuzzy logic can be used to predict demand, inventory and totals at the Habibi Restaurant. And test results with real data and surveys show results with a conformity level of up to 90%.
Penerapan Algoritma K-Means untuk Pengelompokan Data Mahasiswa Baru Program Studi Teknik Informatika di Universitas Pahlawan Tuanku Tambusai Kasini, Kasini; Rusnedy, Hidayati; Tanjung, Lailatul Syifa; Munti, Novi Yona Sidratul
Jurnal Teknik Industri Terintegrasi (JUTIN) Vol. 8 No. 1 (2025): January
Publisher : LPPM Universitas Pahlawan Tuanku Tambusai

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/jutin.v8i1.41449

Abstract

Pahlawan Tuanku Tambusai University (UP) in Riau Province has an Informatics Engineering Study Program that accepts new students every year from various regions around Bangkinang. Incoming student data is processed to assist decision making, especially in the field of promotion. This study aims to apply the K-Means algorithm to Informatics Engineering Study Program student data, with attributes of student name and district of origin, to group regions based on promotion potential. The K-Means method is used to group data into three clusters: High Priority, Medium Priority, and Low Priority. The results of the analysis show that there are 22 regions included in the High Priority Cluster, 23 regions in the Medium Priority Cluster, and 43 regions in the Low Priority Cluster. Regions in the High Priority Cluster are the main priority for promotion strategies, while regions in the Medium Priority and Low Priority Clusters require a more focused promotion approach. This study provides an important contribution to the promotion strategy of the Informatics Engineering Study Program at UP by using a data mining approach to increase the visibility of the study program in the community
Sistem Informasi Gereja Berbasis Website di Kecamatan Salo dan Bangkinang Kota Kabupaten Kampar Alfredo Siagian; Safni Marwa; Kasini
Jurnal Inovasi Teknik Informatika Vol. 9 No. 2 (2024): September 2024
Publisher : Universitas Pahlawan Tuanku Tambusai

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The web-based church information system in Salo and Bangkinang Kota Districts, Kampar Regency, is designed to improve efficiency and accuracy in disseminating information and managing church data. Churches in this region still use manual methods to share worship schedules and activities, which creates challenges for congregants and visitors in accessing essential information. This study aims to analyze, design, and develop a web-based church information system that can be accessed anytime and anywhere. The research employs the Prototype method, with system testing conducted using Black Box Testing. The system is developed using the PHP programming language, the CodeIgniter framework, and MySQL database. The findings indicate that the system successfully provides real-time worship information and simplifies church data management, including congregation numbers, vision and mission, and church history. The system fulfills functional requirements with a 100% success rate. This system enhances communication, data management, and church services, making worship schedules and church activities more easily accessible to congregants, the local community, and newcomers. Keywords: Church Information System, Website, UI, Efficiency, Prototype.
HUBUNGAN KEPATUHAN KONTROL DENGAN KADAR GULA DARAH PUASA PASIEN DIABETES MELLITUS DI PUSKESMAS NGRAHO Fitria, Mei; Fitria Kurniati, Mei; Zainal Abidin, Ahmad; Kasini
Jurnal Ilmu Kesehatan MAKIA Vol 13 No 1 (2023): Jurnal Ilmu Kesehatan MAKIA
Publisher : LPPM ISTeK ICsada Bojonegoro

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37413/jmakia.v13i1.256

Abstract

ABSTRACT Compliance control is an important aspect for success in controlling blood sugar levels.Difficulties in knowing health developments, causing the risk of complications will arise in patients with diabetes mellitus who do not routinely control blood sugar.The purpose of the study was to determine the relationship between control compliance with fasting blood sugar levels in patients with diabetes mellitus at the Ngraho Public Health Center. The research design uses quantitative methods with a Cross Sectional approach.The independent variable in this study is control compliance and the dependent variable is blood sugar levels.The population in this study were all 32 DM patients who took control in January-December 2020 using total sampling. The results showed that 16 respondents who complied with control had an average fasting blood sugar of 211.45 while 16 respondents who did not comply with control had an average fasting blood sugar of 189.20.Based on the Mann-Whitney test, the p-value was 0.142 with a significance value of <0.05, which means there is no relationship between Control Compliance and Blood Sugar Levels in Diabetes Mellitus Patients at the Ngraho Public Health Center. Increased fasting blood sugar levels occur not solely because of the regularity of the control schedule, but there are other factors such as age, irregular insulin use, foods with high glucose levels, excessive stress levels, and lack of activity. affect the body in controlling blood sugar levels.
Application of TOPSIS and K-Means Clustering Methods in Recommendations and Analysis of Study Program Interests for New Students Rusnedy, Hidayati; Kasini; Tanjung, Lailatul Syifa; Yusmita, Yesi
Journal of Engineering Science and Technology Management (JES-TM) Vol. 5 No. 1 (2025): Maret 2025
Publisher : Journal of Engineering Science and Technology Management

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/jestm.v5i1.251

Abstract

The selection of a study program is one of the crucial initial decisions for prospective students when entering college. This decision is ideally based on a good understanding of their interests, talents, and abilities so that prospective students can study optimally and in accordance with their potential. Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) is a multi-criteria decision-making method where the best alternative has the longest distance from the negative ideal solution and has the shortest distance from the positive ideal solution. The selection of these criteria and alternatives aims to produce relevant and accurate recommendations in helping prospective students determine the choice of study program that best suits their potential and preferences. The results of these recommendations are then further analyzed to group the results of the recommendations based on the category of interest. The K-Means Clustering method using the K-Means method with the results of C1 with 9 Respondentsts of less interested study programs, C2 with 25 Respondentsts of moderately interested study programs, and C3 with 21 Respondentsts of highly interested study programs