Articles

Found 4 Documents
Search

Pengelompokkan Data Pembelian Tinta Dengan Menggunakan Metode K-Means Susliansyah, Susliansyah; Sumarno, Heny; Priyono, Hendro; Hikmah, Noer
J-SAKTI (Jurnal Sains Komputer dan Informatika) Vol 3, No 2 (2019): EDISI SEPTEMBER
Publisher : STIKOM Tunas Bangsa Pematangsiantar

Show Abstract | Original Source | Check in Google Scholar | Full PDF (641.686 KB)

Abstract

PT. Mayer Indah Indonesia is engaged in the production of goods, where the most important part to prepare the needs for production needs is the purchasing department, but in the purchasing section it is difficult to determine which items must be bought a lot, are and few in meeting the demand requirements of each part because of the needs goods for production are very unpredictable, eventually causing some goods demand not to be fulfilled because the goods are out of stock. To solve the problems experienced by the purchasing part, datamining using clustering algorithm is k-means method, where the initial stages determine the centroid randomly and do the first iteration calculation and determine the new centroid from the first iteration, then the second iteration calculation is done, because the results of the first and second iterations in the smallest layout of the three groups, the calculation stops. The results obtained by using the ink purchase data seen from the three attributes of incoming goods, items purchased and stock of goods, making it easier and help the purchasing department in classifying items that must be purchased a lot, medium and little.
Komparasi Algoritma Klasifikasi Machine Learning Pada Analisis Sentimen Film Berbahasa Indonesia Sumarno, Heny
Bina Insani ICT Journal (OLD) Vol 4 No 2 (2017): Bina Insani ICT Journal
Publisher : Penelitian dan Pengabdian Masyarakat STMIK Bina Insani

Show Abstract | Original Source | Check in Google Scholar | Full PDF (465.128 KB)

Abstract

Abstrak: Analisa Sentimen adalah proses yang bertujuan membedakan antara polarita di antara tiga harga yaitu positif, negatif dan netral. Opini publik adalah sumber informasi penting yang dibutuhkan dalam pengambilan keputusan sesorang terhadap suatu produk. Saat ini, opini konsumen terhadap pengalaman suatu produk semakin meningkat melalui media online. Untuk membaca opini-opini ini membutuhkan waktu yang banyak, tetapi jika hanya mengambil opini dalam jumlah yang sedikit dapat menimbulkan bias informasi. Algoritma Klasifikasi seperti Naïve Bayes (NB), Support Vector Machine (SVM), dan C.45 dapat digunakan peneliti untuk tujuan melakukan analisa sentimen dari opini suatu produk film. Berdasarkan hal ini, dalam penelitian ini dilakukan perbandingan dari tiga algoritma tersebut untuk mendapatkan tingkat pengetesan data yang paling tinggi. Dari penelitian ini didapat kesimpulan bahwa algoritma Naïves Bayeslah yang mendapatkan tingkat yang paling tinggi. Setelah dilakukan kombinasi antara algoritma Naïve Bayes dan Algoritma Genetika dengan seleksi fitur untuk meningkatkan tingkat akurasi dari Naïve Bayes classifier. Evaluasi selesai dilakukan dengan menggunakan metode 10 fold cross validation. Akurasi dari tingkat pengukuran diukur dengan menggunakan confussion matrix dan kurva ROC. Hasil akhir yang didapat dari klasifikasi text yang merupakan penggabungan dari opini positif dan negatif menunjukan terjadi peningkatan dalam hal akurasi sebesar 73 sampai dengan 80 persen pada algoritma Naïve Bayes.   Kata Kunci: Algoritma Genetika, Analisa Sentimen, Machine, C4.5, Naïve Bayes, Opini, Support Vector   Abstract: Sentiment analysis is the process aiming to determine whether the polarity of a towards the positive, negative or neutral. Public opinion is an important source in the decision-making person to a product. Nowadays consumers are increasingly making their opinions and experiences online. Reading those opinions are time-consuming, but, if only few opinions were read, the evaluation would be biased. Classification algorithms such as Naive Bayes (NB), Support Vector Machine (SVM), and C4.5 were proposed by many researchers to be used in sentiment analysis of movie opinions. Therefore, in this study will be to compare the third is to get agorima agoritma where most superior in the test data. So Naive Bayes algorithm generated the most superior. After the Naive Bayes algorithm will be combined with genetic algorithm feature selection in order to improve the accuracy of Naive Bayes classifier. The evaluation was done using 10 fold cross validation. While the measurement accuracy is measured by the confusion matrix and ROC curves. This research resulted in text classification in the form of a positive or negative opinions Indonesian language film. The results showed an increase in the accuracy of Naive Bayes 73.00% to 80.50%.   Keywords: C4.5, Genetic Algorith,.Sentimetn Analysis, Naive Bayes, Opinion, Support Vector Machine.
Sistem Informasi Evaluasi Dosen Pada Asian Banking Finance And Informatics Institute (ABFI) Perbanas Christo, Petrus; Sumarno, Heny; Atmaja, Widi
INFORMATION SYSTEM FOR EDUCATORS AND PROFESSIONALS Vol 2 No 1 (2017): INFORMATION SYSTEM FOR EDUCATORS AND PROFESSIONALS (Desember 2017)
Publisher : Penelitian dan Pengabdian Masyarakat STMIK Bina Insani

Show Abstract | Original Source | Check in Google Scholar | Full PDF (352.089 KB)

Abstract

Abstrak: Dalam dunia pendidikan peran dosen sebagai pengajar sangat penting, karena dosen merupakan kunci utama yang memberikan ilmu kepada mahasiswa dalam proses belajar mengajar. Proses evaluasi dosen yang diselenggarakan ABFI Institute Perbanas masih dilakukan secara manual dengan mengunakan media kertas yang penyebaran kuesionernya dilakukan ke kelas-kelas. Sistem penyebaran ini tidak efektif karena memerlukan banyak waktu, biaya serta dalam perhitungan dan pembuatan laporan masih terjadi kesalahan sehingga hasil yang dicapai tidak cepat dan akurat. Untuk mengatasi masalah tersebut perlu adanya perancangan sistem yang terkomputerisasi agar hasil yang dicapai akan lebih cepat dan akurat.Oleh karena itu penelitian ini membahas mengenai perancangan sistem guna melakukan evaluasi dosen di lingkungan ABFI Perbanas.   Kata Kunci: Evaluasi Dosen, Sistem Informasi, Rapid Aplication Development   Abstract: In the world of education the role of lecturers as a lecturer is very important, because the lecturer is the main key that provides knowledge to students in teaching and learning process. The evaluation process of lecturers held by ABFI Institute Perbanas is still done manually by using paper media which spread the questionnaire done to the classes. This deployment system is ineffective because it requires a lot of time, cost and in calculation and making of the report is still error so that result is not fast and accurate. To overcome these problems need to design a computerized system so that the results achieved will be faster and accurate. Therefore this paper discusses the war system to evaluate the lecturers in the environment ABFI Perbanas.   Keyword: Information System, Lecture Evaluation, Rapid Aplication Development
PENERAPAN METODE TOPSIS DALAM PENILAIAN KINERJA GURU TETAP SD NEGERI KEBALEN 07 Susliansyah, Susliansyah; Rahadjeng, Indra Riyana; Sumarno, Heny; Deleaniara. M, Chyntia Marianna
Jurnal Pilar Nusa Mandiri Vol 15 No 1 (2019): PILAR Periode Maret 2019
Publisher : PPPM Nusa Mandiri

Show Abstract | Original Source | Check in Google Scholar | Full PDF (979.127 KB)

Abstract

To find out the problems faced in the teaching performance assessment process by utilizing the Technique For Order Preference method by Similiarity to Ideal Solution (TOPSIS), to manage the processing of Teacher data is a more optimal consideration. By using the (TOPSIS) method as a basis for processing teacher performance assessment data. This can allow the system to provide an assessment in accordance with the quality of each teacher and is expected to facilitate decision making in the assessment of Teacher's performance. The Technique For Order Preference by similiarity to Ideal Solution has been running well and can result in a weighting of assessment criteria and clear and fast information compared to manual calculations so SD Negeri Kebalen 07 can use it as a tool for making appropriate decisions.