nan Haviluddin

Forecasting network activities using ARIMA method

This paper presents an approach for a network traffic characterization by using an ARIMA (Autoregressive Integrated Moving Average) technique. The dataset used in this study is obtained from the internet network traffic activities of the Mulawarman University for a period of a week. The results are obtained using the Box-Jenkins Methodology. The Box-Jenkins methodology consists of five ARIMA models which include ARIMA (2, 1, 1) (1, 1, 1) ¹², ARIMA (1, 1, 1) (1, 1, 1) ¹², ARIMA (2, 1, 0) (1, 1, 1) ¹², ARIMA (0, 1, 0) (1, 1, 1) ¹², and ARIMA (0, 1, 0) (1, 2, 1) ¹². In this paper, ARIMA (0, 1, 0) (1, 2, 1) ¹² was selected as the best model that can be used to model the internet network traffic.

Comparing performance of Backpropagation and RBF neural network models for predicting daily network traffic

The predicting daily network traffic usage is a very important issue in the service activities of the university. This paper present techniques based on the development of backpropagation (BP) and radial basis function (RBF) neural network models, for modelling and predicting the daily network traffic at Universitas Mulawarman, East Kalimantan, Indonesia. The experiment results indicate that a strong agreement between model predictions and observed values, since MSE is below 0.005. When performance indices are compared, the RBFNN-based model is a more accurate predictor with MSE value is 0.00407999, MAPE is 0.03701870, and MAD is 0.06885187 than the BPNN model. Therefore, the smallest MSE value indicates a good method for accuracy, while RBF finding illustrates proposed best model to analyze daily network traffic.

A Numerical Classification Technique Based on Fuzzy Soft Set Using Hamming Distance

In recent decades, fuzzy soft set techniques and approaches have received a great deal of attention from practitioners and soft computing researchers. This article attempts to introduce a classifier for numerical data using similarity measure fuzzy soft set (FSS) based on Hamming distance, named HDFSSC. Dataset have been taken from UCI Machine Learning Repository and MIAS (Mammographic Image Analysis Society). The proposed modeling consists of four phases: data acquisition, feature fuzzification, training phase and testing phase. Later, head to head comparison between state of the art fuzzy soft set classifiers is provided. Experiment results showed that the proposed classifier provides better accuracy when compared to the baseline fuzzy soft set classifiers.

Modeling Data Containing Outliers using ARIMA Additive Outlier (ARIMA-AO)

The aim this study is discussed on the detection and correction of data containing the additive outlier (AO) on the model ARIMA (p, d, q). The process of detection and correction of data using an iterative procedure popularized by Box, Jenkins, and Reinsel (1994). By using this method we obtained an ARIMA models were fit to the data containing AO, this model is added to the original model of ARIMA coefficients obtained from the iteration process using regression methods. This shows that there is an improvement of forecasting error rate data.

Forecasting Inflow and Outflow of Money Currency in East Java Using a Hybrid Exponential Smoothing and Calendar Variation Model

Money currency availability in Bank Indonesia can be examined by inflow and outflow of money currency. The objective of this research is to forecast the inflow and outflow of money currency in each Representative Office (RO) of BI in East Java by using a hybrid exponential smoothing based on state space approach and calendar variation model. Hybrid model is expected to generate more accurate forecast. There are two studies that will be discussed in this research. The first studies about hybrid model using simulation data that contain pattern of trends, seasonal and calendar variation. The second studies about the application of a hybrid model for forecasting the inflow and outflow of money currency in each RO of BI in East Java. The first of results indicate that exponential smoothing model can not capture the pattern calendar variation. It results RMSE values 10 times standard deviation of error. The second of results indicate that hybrid model can capture the pattern of trends, seasonal and calendar variation. It results RMSE values approaching the standard deviation of error. In the applied study, the hybrid model give more accurate forecast for five variables : the inflow of money currency in Surabaya, Malang, Jember and outflow of money currency in Surabaya and Kediri. Otherwise, the time series regression model yields better for three variables : outflow of money currency in Malang, Jember and inflow of money currency in Kediri. Keywords—Exponential Smoothing, Hybrid, Inflow, Outflow, Time Series Regression, State Space

Modeling of time series data for forecasting the number of foreign tourists in east Kalimantan using fuzzy inference system based on ARX model

The government agencies require accurate tourism demand forecasts to plan the required tourism infrastructure, such as accommodation location planning and transportation development. Tourism demand forecasts can be viewed from various factors, one of which is the number of tourists per period. Without ignoring the number of domestic tourists, the increasing number of foreign tourists is prioritized by the government to increase the countrys foreign exchange. Usually, a tourist destination is to visit some tour objects and need some accommodation and hotel sites to rest. With this consideration, the forecasting number of foreign tourists can be done by using data on the number of tour objects, accommodation and hotel sites, foreign and domestic tourists from the previous period. Data on the number of domestic tourists used to measure the tendency of foreign tourists compared with domestic tourists to all existing tour objects. All data history can be viewed as time series data. Conventionally, many researchers have employed traditional methods of time series analysis, modeling, and forecasting such as ARX (Autoregressive with exogenous input) model. FIS (Fuzzy Inference System) is a system that processes the input mapping formulation provided to produce output using Fuzzy Logic. The aim of this study is to forecast the number of foreign tourists by using FIS through a training process conducted by adapting the number of linguistic variables. All the training data are modeled by using the ARX model.

Water Level Prediction of Lake Cascade Mahakam Using Adaptive Neural Network Backpropagation (ANNBP)

A natural hazard information and flood events are indispensable as a form of prevention and improvement. One of the causes is flooding in the areas around the lake. Therefore, forecasting the surface of Lake water level to anticipate flooding is required. The purpose of this paper is implemented computational intelligence method namely Adaptive Neural Network Backpropagation (ANNBP) to forecasting the Lake Cascade Mahakam. Based on experiment, performance of ANNBP indicated that Lake water level prediction have been accurate by using mean square error (MSE) and mean absolute percentage error (MAPE). In other words, computational intelligence method can produce good accuracy. A hybrid and optimization of computational intelligence are focus in the future work.

Comparison Between K-Means and Fuzzy C-Means Clustering in Network Traffic Activities

A network traffic utilization in order to support teaching and learning activities are an essential part. Therefore, the network traffic management usage is requirements. In this study, analysis and clustering network traffic usage by using K-Means and Fuzzy C-Means (FCM) methods have been implemented. Then, both of method were used Euclidean Distance (ED) in order to get better results clusters. The results showed that the FCM method has been able to perform clustering in network traffic.

Performance of Decision Tree C4.5 Algorithm in Student Academic Evaluation

Student academic evaluation is part of academic information system (AIS) performance, in order to control student learning progress is necessary. Furthermore, the evaluation showing whether the student will pass or fail would benefit the student/instructor and act as a guide for future recommendations/evaluations on performance. An in depth study on the student academic evaluation technique by using Decision Tree C4.5 has been conducted. Specific parameters including age, place of birth, gender, high school status (public or private), department in high school, organization activeness, age at the start of high school level, and progress GPA (pGPA) and Total GPA (tGPA) from semester 1–4 with three times graduation criteria (i.e., fast, on, and delay) times have been defined and tested. The scope of the paper has been set for undergraduate programs. The experimental results show that accuracy algorithm (AC) of 78.57% with true positive rate (TP) of 76.72% by using quality training data of 90% have best performance accuracy value.

Scilab software as an alternative low-cost computing in solving the linear equations problem

Numerical computation packages are widely used both in teaching and research. These packages consist of license (proprietary) and open source software (non-proprietary). One of the reasons to use the package is a complexity of mathematics function (i.e., linear problems). Also, number of variables in a linear or non-linear function has been increased. The aim of this paper was to reflect on key aspects related to the method, didactics and creative praxis in the teaching of linear equations in higher education. If implemented, it could be contribute to a better learning in mathematics area (i.e., solving simultaneous linear equations) that essential for future engineers. The focus of this study was to introduce an additional numerical computation package of Scilab as an alternative low-cost computing programming. In this paper, Scilab software was proposed some activities that related to the mathematical models. In this experiment, four numerical methods such as Gaussian Elimination, Gauss-Jordan, Inverse ...