Eko Nugroho

Parallel implementation of genetic algorithm for searching optimal parameters of artificial neural networks

Genetic algorithm is a population-based evolutionary optimization approach, and this has been used to train artificial neural networks successfully. When a large number of individuals make up the population, however, the running time of the algorithm often becomes very long. Parallel computation is a technique that can potentially be used to address this issue. This research investigates the adoption of parallel genetic algorithm for searching optimal parameters of artificial neural networks (multilayer perceptrons). Paralellization is conducted using Message Passing Interface, where sub-populations (and their fitness values) are exchanged between processors while selection, crossover, and mutation processes, which are necessary to generate new sub-populations, are performed independently between processors. Experiments of this research show that parallelization has indeed reduced the running time of genetic algorithm. The proposed method produces an average running time of 2.066, 2.656, and 47.788 seconds/generation for the game of Tic-Tac-Toe, car evaluation, and game of chess (King-Rook vs. King-Pawn) datasets, respectively. For comparison, the average running time of the serial version of the algorithm is 14.397, 12.961, and 350.963 seconds/generation for the same three datasets, respectively.

Pengembangan Website Multi-Screen dan Analisis Perbandingan User Behaviournya

Abstract. The wide availability of affordable smartphone and tablet in the market leads to the increase of internet access using mobile devices. Some users have started to leave desktop devices and are now using mobile devices. Some decide to use mobile device, some are still using desktop device, while some others are using both. This study aims to analyze the differences in the users’ behaviours when visiting a website in a desktop and mobile version. The aspects of the user behaviour studied are page view/visit, time on site, bounce rate, and registration ratio. The data are taken and saved by Google Analytics and database website which are analyzed by using descriptive statistic and ANOVA. The result shows that the desktop version has high value in those previously mentioned aspects compared to the mobile version. Keywords: User Behaviour, desktop, mobile, analytics. Abstrak. Membanjirnya smartphone dan tablet di pasaran dengan harga terjangkau, mendorong naiknya pengaksesan internet menggunakan mobile device. Sebagian user telah berpindah dari pengguna desktop device menjadi pengguna mobile device. Sebagian masih tetap menggunakan desktop device, dan sebagian lagi menggunakan keduanya. Tujuan dari penelitian ini adalah mengembangkan website multi-screen dan menganalisis perbedaan user behaviour ketika mengunjungi website versi desktop dan versi mobile. Adapun user behaviour yang diteliti adalah: pageview/visit (banyaknya halaman yang dibuka user tiap berkunjung), time on site (durasi kunjungan user), bounce rate (rasio user yang hanya membuka satu halaman kemudian meninggalkan website), dan registration ratio (rasio user yang bersedia mendaftar). Data diambil dan disimpan oleh Google Analytic dan database internal website, yang kemudian dianalisis menggunakan statistik deskriptif dan ANOVA. Hasilnya menunjukkan bahwa semua jenis user behaviour website versi desktop dan versi mobile yang diteliti mempunyai perbedaan yang signifikan. Kata Kunci: User behaviour, desktop, mobile, analytics.

SISTEM PENDUKUNG KEPUTUSAN UANG KULIAH TUNGGAL DENGAN METODE ACTIVITY BASED COSTING

Higher education is one of education institution with tight competition. University have to increase their quality to captivate the public. In addition to consideration of the quality of higher education, the cost of education is also a consideration for choosing a university. In 2012, Indonesia has 33 universities in a form of Badan Layanan Umum (BLU). According to PMK no 76 2008, BLU must have a Cost Accounting System which generates unit cost. It’s used by university to define their tuition fee. In 2013 many of BLU universities don’t have a cost structure analysis. The research goal is designing, creating application that can define unit cost in a BLU university with Activity Based Costing (ABC) method. The research use PHP as programming language and MySQL as database engine. The result of this research is design the system, prototype and testing the application has already done. The System contains define cost structure template, input cost structure and simulate the student tuition fee. The system can help university to define their unit cost. The system has been tested using black box method is done at the level of departments, faculties, bureaus, and divisions in Sebelas Maret University.