Claim Missing Document

Found 2 Documents

Analisa Web untuk Memahami Perilaku Konsumen Online Studi Kasus “Store Steam Powered” Supriyadi, Endang
Transparansi : Jurnal Ilmiah Ilmu Administrasi Vol 7, No 2: September 2015
Publisher : Institut Ilmu Sosial dan Manajemen STIAMI

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (608.942 KB) | DOI: 10.31334/transparansi.v7i2.849


The development of internet users from year to year to date continues to increase, it can be known through a survey conducted Internet Worldstats. According to a survey conducted by Internet Worldstats, United Stated ranks 2nd of internet users worldwide after China. From the results of the survey shows that almost all activities / business activities undertaken by the world population and in particular the citizens of United State / America using internet facilities in all activities, especially transactions online. And one of the tools that can help to analyze online consumer behavior is by using web analytics. The purpose of this paper is to analyze how data analysis can be effectively used to understand current consumer behavior online. Expected with the use of web analytics, companies can predict future consumer behavior online and special messages on promotions that have been sent to each individual. Understanding the patterns and things on the web analytics is expected to help to develop business processes and the contents and design that need to be made especially "store steam powered". Store steam powered is one of Valve's web-based digital distributor. The main service that is done is steam game sales through an e-commerce web that aims to avoid piracy is very rampant at the moment.
Metode SVM Berbasis PSO untuk Meningkatkan Prediksi Ketepatan Waktu Kelulusan Mahasiswa Supriyadi, Endang
Jurnal Sistem Informasi Vol 6 No 2 (2017)

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (485.652 KB)


Abstract—This scientific article discusses how the swarm optimization particle method can improve accuracy in predicting student graduation accuracy. This is done in order to anticipate the decrease in the number of students in their lectures. The experimental results have shown that particle swarm optimization (PSO) can improve the accuracy of the support vector machine method from 80.14% accuracy to 82.05% so that there is an increase of accuracy of 1.91%. Intisari— Artikel ilmiah ini membahas bagaimana metode partikel swarm optimization dapat meningkatkan akurasi dalam memprediksi ketepatan kelulusan mahasiswa. Hal tersebut dilakukan agar dapat mengantisipasi penurunan jumlah mahasiswa dalam kegiatan perkuliahannya. Hasil eksperimen yang telah dilakukan terbukti bahwa particle swarm optimization ( PSO ) dapat meningkatkan akurasi metode support vector machine dari  akurasi sebesar 80.14% menjadi 82,05% sehingga terbukti ada peningkatan akurasi sebesar  1.91 %. Kata Kunci— SVM, PSO, akurasi, prediksi, data mining.