Articles
<![CDATA[Segmentasi Loyalitas Pelanggan Berbasis RFM (Recency, Frequency, Monetary) Menggunakan K-Means pada PD. Persada Ikan ]]>
<![CDATA[Yosia Oktavian Pailan]]>;
<![CDATA[Yulison Herry Chrisnanto]]>;
<![CDATA[Asep Id Hadianna]]>
<![CDATA[ Prosiding SISFOTEK Vol 4 No 1 (2020): Vol 4 No 1 (2020): SISFOTEK 2020 ]]>
Publisher : <![CDATA[Ikatan Ahli Informatika Indonesia ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (457.406 KB)
<![CDATA[Customer loyalty for the company is very important if the competition between similar companies is high enough which results in a threat to the company. Customer loyalty is very useful to determine the level of customer loyalty to the company. Customer segmentation is also needed to group customers who have the same characteristics into one so as to simplify the management process. The analysis used is the RFM (Recency, Frequency, Monetary) model to analyze customer buying behavior in terms of Recency (last transaction time span), Frequency (number of transactions), and Monetary (rupiah issued). The grouping method used is K-Means. The data used in this study are historical data on fish feed purchases from 2015 to 2017. The application of RFM analysis and the K-Means method resulted in 4 clusters based on Elbow calculations. The results in this study obtained the number of objects in cluster 1 as many as 142 customers, cluster 2 as many as 28 customers, cluster 3 as many as 41 customers and cluster 4 as many as 41 customers. The accuracy level of the cluster is measured using Silhoutte Coeffisien with results close to 1 which means the clustering is quite good. The interpretation of the RFM shows that 16.27% of customers have a high potential for loyalty, while 11.11% of customers have the potential as loyal customers, and the remaining 56.35% have a low level of loyalty. It can be concluded that this research can classify the level of customer loyalty using RFM analysis and the K-Means algorithm.]]>
<![CDATA[Sistem Prediksi Mutu Air Di Perusahaan Daerah Air Minum Tirta Raharja Menggunakan K – Nearest Neighbors (K – NN) ]]>
<![CDATA[Rahandanu Rachmat]]>;
<![CDATA[Yulison Herry Chrisnanto]]>;
<![CDATA[Fajri Rakhmat Umbara]]>
<![CDATA[ Prosiding SISFOTEK Vol 4 No 1 (2020): Vol 4 No 1 (2020): SISFOTEK 2020 ]]>
Publisher : <![CDATA[Ikatan Ahli Informatika Indonesia ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (408.793 KB)
<![CDATA[PDAM (Perusahaan Daerah Air Minum) Tirta Raharja is the only Regional Business Entity (BUMD) that has the task of providing clean water services to the people of Cimahi City. Clean water is the main requirement that must be consumed by the community and managed in the smooth running of community activities. The development of the city of Cimahi is currently quite fast, with plans to build smart cities, causing the need for clean air as needed. K - Nearest Neighbor (KNN) is a classification algorithm that considers several supporting parameters to carry out a classification process that results in ease of calculation and power. KNN can be considered as one of the most famous non-parametric models. In the research and implementation process of data mining in the regulation of water quality feasibility in PDAM Tirta Raharja using K - the nearest neighbor can be denied as the K - the nearest neighbor implemented in the process of testing the drinking water feasibility in PDAM Tirta Raharja, can be used 93% to be used with the Eligible label Drunk, and 98% for accuracy testing with the label Not Eligible to drink with a K value of 14 where the K value is the most ideal amount that must go through K - Fold Validation Validation of a total of 1,818 data.]]>
<![CDATA[Pembangunan Sistem Informasi Manajemen Aset pada PT. Kraft Ultrajaya Indonesia ]]>
<![CDATA[Mukti Kinani]]>;
<![CDATA[Yulison Herry Chrisnanto]]>;
<![CDATA[Irma Santikarama]]>
<![CDATA[ Prosiding SISFOTEK Vol 4 No 1 (2020): Vol 4 No 1 (2020): SISFOTEK 2020 ]]>
Publisher : <![CDATA[Ikatan Ahli Informatika Indonesia ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (1076.236 KB)
<![CDATA[Asset management information systems are widely used by private and government companies because they can help companies and governments in managing assets that aim to manage, maintain, and supervise important assets. In this study, researchers tried to implement a system to manage asset management at PT. Kraft Ultrajaya Indonesia because there are still problems where asset management that does not have inaccurate information about the assets owned causes the managers not to have specific references used in determining needs in the procurement process. In addition, there is an inadequate monitoring process for the realization of assets received that are not in accordance with the purchase order form, resulting in a lack of information on the status of realization of assets received. In other cases there is also insufficient depreciation of asset maintenance information, resulting in inconsistencies in data and information regarding the condition of assets that are inconsistent with the condition of assets that are depreciated in the field, which may result in uncontrolling of the available assets. From this problem, researchers tried to implement an asset management information system in order to make it easier to help access historical data that is easy in the process of asset data processing and asset control starting from the asset planning process to revenue so that it can help the company evaluate each asset owned and simplify the decision-making process, thereby minimizing the purchase of excess assets that can harm the company and can also control every asset management that exists in the company. In addition, the asset management information system that is implemented will provide information on asset status management, especially the depreciation of each asset, so as to know the quality status of existing assets and the condition of asset depreciation.]]>
<![CDATA[Sistem Segmentasi Program Talk Show Berdasarkan Media Sosial Twitter Menggunakan Metode K-Medoids Clustering ]]>
<![CDATA[Kharisma Jevi Shafira Sepyanto]]>;
<![CDATA[Yulison Herry Chrisnanto]]>;
<![CDATA[Fajri Rakhmat Umbara]]>
<![CDATA[ Prosiding SISFOTEK Vol 4 No 1 (2020): Vol 4 No 1 (2020): SISFOTEK 2020 ]]>
Publisher : <![CDATA[Ikatan Ahli Informatika Indonesia ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
Full PDF (718.428 KB)
<![CDATA[Innovations on a talk show on television can be a threat. Audience will be divided into groups so that it can make a downgrade rating program. Program ratings affect companies that will use advertising services. Television companies will go bankrupt. The biggest source of income is sales of advertising services. One way to overcome them can be analyzed in public opinion. The results of the analysis can provide information about the attractiveness of the community towards the program. But the analysis process takes a long time and can be done only by a competent person so another process is needed to get the results of the analysis that is fast and can be done by anyone. In this study using K-Medoids Clustering in the process of identifying public opinion. The clustering process known as unsupervised learning will be combined with the labeling process. The previous episode's tweet data will be labeled and then used to obtain the predicted labels from other cluster members. Labels consist of three types, namely 1) theme, 2) resource persons, and 3) programs. Before going through the clustering stage, the tweet data will go through the text preprocessing stage then transformed into a numeric form based on the appearance of the word. Transformation data will be clustered by calculating proximity using Cosine Similarity. Labels from the Medoids cluster will be used on unlabeled tweet data. The cluster results were tested using the Silhouette Coefficient method to get 0.19 results. However, this method successfully predicted public opinion and achieved an accuracy of 80%. Innovations on a talk show on television can be a threat. Audience will be divided into groups so that it can make a downgrade rating program. Program ratings affect companies that will use advertising services. Television companies will go bankrupt. The biggest source of income is sales of advertising services. One way to overcome them can be analyzed in public opinion. The results of the analysis can provide information about the attractiveness of the community towards the program. But the analysis process takes a long time and can be done only by a competent person so another process is needed to get the results of the analysis that is fast and can be done by anyone. In this study using K-Medoids Clustering in the process of identifying public opinion. The clustering process known as unsupervised learning will be combined with the labeling process. The previous episode's tweet data will be labeled and then used to obtain the predicted labels from other cluster members. Labels consist of three types, namely 1) theme, 2) resource persons, and 3) programs. Before going through the clustering stage, the tweet data will go through the text preprocessing stage then transformed into a numeric form based on the appearance of the word. Transformation data will be clustered by calculating proximity using Cosine Similarity. Labels from the Medoids cluster will be used on unlabeled tweet data. The cluster results were tested using the Silhouette Coefficient method to get 0.19 results. However, this method successfully predicted public opinion and achieved an accuracy of 80%.]]>
<![CDATA[KLASIFIKASI SENTIMEN PERGELARAN MOTOGP DI INDONESIA MENGGUNAKAN ALGORITMA CORRELATED NAÏVE BAYES CLASIFIER ]]>
<![CDATA[RIDWAN INDRANSYAH]]>;
<![CDATA[Yulison Herry Chrisnanto]]>;
<![CDATA[Puspita Nurul Sabrina]]>
<![CDATA[ INFOTECH journal Vol. 8 No. 2 (2022) ]]>
Publisher : <![CDATA[Universitas Majalengka ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.31949/infotech.v8i2.3103
<![CDATA[Knowing the public's sentiment towards the international MotoGP event which has been held in Indonesia in 2022 is very necessary because the role of the community is very influential in the implementation and public interest in visiting an international event is still few and difficult because the information is still limited. Tweets, comments, reviews, and opinions of people using social media play an important role in determining whether a particular population is satisfied with products, performances, and services. The method used in this study is the Correlated Naïve Bayes Classifier (CNBC). The Correlated Naive Bayes Classifier (CNBC) method recalculates the correlation value for each attribute of the dataset to that class. There are several processes carried out in this study including data acquisition, data labeling, data preprocessing, feature extraction, classifying data using the Correlated Naive Bayes Classifier (CNBC) method, visualizing data, and finally evaluating the results. This study resulted in an accuracy of 82%.]]>
<![CDATA[KLASIFIKASI PENENTUAN KELAYAKAN PINJAMAN KOPERASI DENGAN ALGORITMA CART MENGGUNAKAN ALGORITMA ADABOOST ]]>
<![CDATA[Muhammad Rendy Raihan]]>;
<![CDATA[Yulison Herry Chrisnanto]]>;
<![CDATA[Ade Kania Ningsih]]>
<![CDATA[ INFOTECH journal Vol. 8 No. 2 (2022) ]]>
Publisher : <![CDATA[Universitas Majalengka ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.31949/infotech.v8i2.3247
<![CDATA[According to the Cooperative Bureau, cooperatives became a mainstay for the lower middle class to revive and stabilize their respective economies when the Covid-19 Pandemic broke out in Indonesia. Through savings and loan cooperatives, people can provide loans to cooperatives. In this case, cooperatives provide money lending services to their members, and certain conditions apply to determine which loans are eligible. In connection with this, the officer will analyze the loan by filling out a loan application form accompanied by certain requirements in each loan application. In a mechanism that is not simple, problems often arise when eligibility decisions are not appropriate, namely bad credit. This research aims to solve the problem by designing a data mining application with a function to determine the feasibility of giving loans to customers. The method used is the CART algorithm method and uses the Adaboost algorithm. The results of the application of the CART algorithm method optimized with Adaboost turned out to be able to classify the eligibility of cooperative lending well, simplify the mechanism in credit analysis activities and be able to provide accurate eligibility status, which is guaranteed by the accuracy results of CART and Adaboost.]]>
<![CDATA[KLASIFIKASI KALIMAT PADA BERITA OLAHRAGA SECARA OTOMATIS MENGGUNAKAN METODE ARTIFICIAL NEURAL NETWORK ]]>
<![CDATA[Asep Saepul Ridwan]]>;
<![CDATA[Yulison H Chrisnanto]]>;
<![CDATA[Ridwan Ilyas]]>
<![CDATA[ J-Icon : Jurnal Komputer dan Informatika Vol 9 No 1 (2021): Maret 2021 ]]>
Publisher : <![CDATA[Universitas Nusa Cendana ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.35508/jicon.v9i1.3708
<![CDATA[Sports news is of great interest to today's society. This is because sports have grown into entertainment. There is a lot of sports news today that covers a wide range of sports, from branches that use the ball as objects for games like football, to sports in automotive race like formula 1. Beyond that, the substance of the sports news itself is as diverse as the news of managerial from a sports club, results matches, player injuries, et cetera. Surely such a thing would be difficult. The network wants one in the field of discussion in the news. Overlap data occurs in the sports news document because it mixes one sentence data with the other. Some of the content in the sports news is about managerial, players, schedules, previews, reviews, standings, statistics, champions, etc. Becomes a problem when the reader wants a topic on the news that focuses on one particular discussion. This study has built a book on sentence classification meaning Artificial Neural Network (ANN) with a method of learning Backpropagation. The feature used is the frequency of the occurrence of a term in the corresponding sentence and the calculating result of a distributed term. The testing of our proposed methods shows an accuracy of 99% to best results on training data and 57% on test data.]]>
<![CDATA[Identifikasi Kemampuan Akademik Mahasiswa Menggunakan K-Means Clustering ]]>
<![CDATA[Eka Rahmawati]]>;
<![CDATA[Yulison Herry Chrisnanto]]>;
<![CDATA[Asri Maspupah]]>
<![CDATA[ Prosiding SEMNAS INOTEK (Seminar Nasional Inovasi Teknologi) Vol. 3 No. 1 (2019): PROSIDING SEMNAS INOTEK Ke-III Tahun 2019 ]]>
Publisher : <![CDATA[Universitas Nusantara PGRI Kediri ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.29407/inotek.v3i1.518
<![CDATA[Peningkatan peran dosen dalam penjaminan mutu di perguruan tinggi akan tercapainya kinerja akademik yang sesuai dengan kemampuan yang dimiliki oleh mahasiswa. Untuk menganalisa kinerja akademik sendiri berarti mengidentifikasi keunikan-keunikan yang ada pada mahasiswa sehingga dalam mengidentifikasi keunikan- keunikan tersebut akan terdapat variabel-variabel yang berpengaruh terhadap kemampuan akademik mahasiswa. Dalam mengidentifikasi kemampuan akademik mahasiswa tidak semua variabel yang digunakan akan berpengaruh terhadap kemampuan akademik mahasiswa. Salah satu cara untuk mengetahui variabel yang berpengaruh terhadap kemampuan akademik mahasiswa yaitu dengan algoritma K-Means Clustering. Penelitian ini menghasilkan sistem yang dapat mencari kombinasi variabel untuk dapat mencerminkan suatu kemampuan akademik dari mahasiswa sehingga terdapat variabel yang tidak berpengaruh terhadap kemampuan akademik mahasiswa dengan menggunakan K-Means Clustering. Hasil pengujian menunjukkan bahwa terdapat variabel yang tidak berpengaruh terhadap kemampuan akademik sehingga terdapat 23 data kemampuan akademik mahasiswa yang tidak sesuai dengan kategori cluster sebelumnya. Hasil penelitian ini juga menunjukkan bahwa terdapat 73 data kemampuan akademik mahasiswa yang sesuai dengan kategori cluster sebelumnya sehingga memiliki nilai akurasi 73% dan dari semua jumlah cluster yang dimasukkan, untuk cluster yang berjumlah 3 memiliki nilai silhouette coefficient yang paling mendekati nilai Si=1 dengan nilai 0.835409226.]]>
<![CDATA[SISTEM ESTIMASI BIAYA DAN SUMBER DAYA PROYEK PERANGKAT LUNAK BERBASIS COCOMO II MENGGUNAKAN NEURAL NETWORK ]]>
<![CDATA[Joko Irawan]]>;
<![CDATA[Yulison Herry Chrisnanto]]>;
<![CDATA[Asri Maspupah]]>
<![CDATA[ Prosiding SEMNAS INOTEK (Seminar Nasional Inovasi Teknologi) Vol. 3 No. 1 (2019): PROSIDING SEMNAS INOTEK Ke-III Tahun 2019 ]]>
Publisher : <![CDATA[Universitas Nusantara PGRI Kediri ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.29407/inotek.v3i1.519
<![CDATA[Estimasi biaya dan sumber daya perangkat lunak merupakan bagian tak terpisahkan dari proyek perangkat lunak. Saat melakukan estimasi sering menghadapi dua permasalahan yaitu estimasi berlebihan dan estimasi yang kurang. Estimasi berlebihan akan menimbulkan penambahan alokasi biaya dan sumber daya, sedangkan estimasi yang kurang akan mengurangi kualitas produk. Untuk mengantisipasi terjadinya kesalahan estimasi maka dikembangkan suatu metode untuk mengestimasi biaya dan sumber daya proyek perangkat lunak berbasis Constructive Cost Model 1997 (COCOMO II) dengan menggunakan Algoritma Backpropagation Neural Network. COCOMO II menghasilkan nilai estimasi dari data latih, sedangkan Algoritma Backpropagation digunakan untuk menghasilkan nilai estimasi data uji. Hasil dari perhitungan Backpropagation dilakukan penggabungan dengan model COCOMO II sehingga menghasilkan estimasi biaya dan sumber daya proyek perangkat lunak. Jumlah data yang digunakan dalam penelitian ini sebanyak 60 data. Model yang dikembangkan ini dievaluasi menggunakan PRED. Berdasarkan hasil dari penelitian diperoleh bahwa penggunaan model COCOMO II, dan algoritma Backpropagation memiliki kedekatan dalam melakukan estimasi biaya, dan sumber daya proyek perangkat lunak dengan nilai PRED sebesar 75%. Kesimpulan dari penelitian ini model COCOMO II, dan algoritma Backpropagation dapat mengurangi tingkat kegagalan dalam proyek perangkat lunak, berdasarkan estimasi biaya, dan sumber daya.]]>
<![CDATA[Model Sistem Penelurusan Alumni Institusi Pendidikan Tinggi Berbasis Teknologi Informasi ]]>
<![CDATA[Yulison Herry Chrisnanto]]>;
<![CDATA[Wawan Setiawan]]>;
<![CDATA[Puspo Dewi Dirgantari]]>
<![CDATA[ Jurnal Ilmiah Matrik Vol 22 No 2 (2020): Jurnal Ilmiah Matrik ]]>
Publisher : <![CDATA[Direktorat Riset dan Pengabdian Pada Masyarakat (DRPM) Universitas Bina Darma ]]>
Show Abstract
|
Download Original
|
Original Source
|
Check in Google Scholar
|
DOI: 10.33557/jurnalmatrik.v22i2.932
<![CDATA[Abstrak: Pemenuhan kebutuhan tenaga terampil yang siap pakai di dunia kerja akan dipengaruhi oleh kemampuan institusi pendidikan tinggi (IPT) dalam mengelola proses untuk menghasilkan lulusan yang produktif, sehingga upaya konstruktif perlu dibangun dalam mempersiapkan mutu lulusan yang dapat diterima di pasar kerja. Kebutuhan yang diperlukan oleh pasar tenaga kerja dapat diketahui oleh IPT sudah barang tentu dipengaruhi oleh informasi yang didapatkan dari para lulusan yang telah menjadi alumni. Dalam hal ini penelusuran alumni menjadi aspek yang penting bagi setiap IPT, di mana melalui penelusuran alumni ini dapat diperoleh berbagai informasi terkait produktivitas alumni pada dunia kerja. Penelitian ini melakukan kajian terhadap peranan teknologi informasi yang mampu memberikan dukungan bagi terbentuknya sistem penelusuran alumni. Pada penelitian ini diawali dengan berbagai kegiatan observasi terhadap dokumen-dokumen utama dan kondisi eksisting pada IPT, khususnya di Universitas Jenderal Achmad Yani (UNJANI) dan mengusulkan sebuah model system penelusuran alumni di institusi pendidikan tinggi (IPT) dengan berbasis teknologi informasi. Kata kunci: Model, Penelusuran almuni, Teknologi Informasi, institusi pendidikan tinggi]]>
