International Journal of Advances in Intelligent Informatics
Not a member yet
235 research outputs found
Sort by
SLA based cloud service composition using genetic algorithm
Get PDFCloud computing tends to provide high quality on-demand services to the users. Numerous services are evolving today. Functionally similar services are having different non-functional properties such as reliability, availability, accessibility, response time and cost. A single service is inadequate for constructing the business process. Such business process is modeled as composite service. Composite service consists of several atomic services connected by workflow patterns. Selecting services for service composition with the constraints specified in Service Level Agreement is the NP-hard problem. Such a cloud service composition problem is modeled in this paper. Genetic based cloud service composition algorithm (GCSC) is proposed. Proposed algorithm is compared with the existing genetic based cloud service composition algorithm based on average utility rate and convergence time. It is proved that the proposed algorithm provides better performance as compared to the existing cloud service composition algorithm
Overdispersion study of poisson and zero-inflated poisson regression for some characteristics of the data on lamda, n, p
Get PDFPoisson distribution is one of discrete distribution that is often used in modeling of rare events. The data obtained in form of counts with non-negative integers. One of analysis that is used in modeling count data is Poisson regression. Deviation of assumption that often occurs in the Poisson regression is overdispersion. Cause of overdispersion is an excess zero probability on the response variable. Solving model that be used to overcome of overdispersion is zero-inflated Poisson (ZIP) regression. The research aimed to develop a study of overdispersion for Poisson and ZIP regression on some characteristics of the data. Overdispersion on some characteristics of the data that were studied in this research are simulated by combining the parameter of Poisson distribution (λ), zero probability (p), and sample size (n) on the response variable then comparing the Poisson and ZIP regression models. Overdispersion study on data simulation showed that the larger λ, n, and p, the better is the model of ZIP than Poisson regression. The results of this simulation are also strengthened by the exploration of Pearson residual in Poisson and ZIP regression
A method for automatic gamelan music composition
Get PDFAim of this study is designing a method for automatic gamelan music composition using rule-base expert system approach. The program is designed for non-expert user in order to help them composing gamelan music or analyzing their composition to achieve explanation and recommendation of ideal composition. There are 2 essential components in this method, which are knowledge and inference. Knowledge is represented into basic knowledge and melodic knowledge. Basic knowledge contains rules that control the structure of gamelan song, and melodic knowledge supports system in composing or analyzing notations sequence that fit the characteristics of melody in gamelan music. Basic knowledge represents basic rules of gamelan music that have quantitative value, so deterministic approach is used for basic knowledge acquisition. Melodic knowledge consists of dynamic data, so stochastic approach is used to create the melodic knowledge base. The rules of composing and analyzing a composition are defined based on basic knowledge and melodic knowledge. The inference engine is designed to compose and analyze a composition. Automatic composition for gamelan music is proposed using Generate and Test method (GAT) with random technique, and composition analysis is proposed using backward chaining metho
Generated rules for AIDS and e-learning classifier using rough set approach
Get PDFThe emergence and growth of internet usage has accumulated an extensive amount of data. These data contain a wealth of undiscovered valuable information and problems of incomplete data set may lead to observation error. This research explored a technique to analyze data that transforms meaningless data to meaningful information. The work focused on Rough Set (RS) to deal with incomplete data and rules derivation. Rules with high and low left-hand-side (LHS) support value generated by RS were used as query statements to form a cluster of data. The model was tested on AIDS blog data set consisting of 146 bloggers and E-Learning@UTM (EL) log data set comprising 23105 URLs. 5-fold and 10-fold cross validation were used to split the data. Naïve algorithm and Boolean algorithm as discretization techniques and Johnson’s algorithm (Johnson) and Genetic algorithm (GA) as reduction techniques were employed to compare the results. 5-fold cross validation tended to suit AIDS data well while 10-fold cross validation was the best for EL data set. Johnson and GA yielded the same number of rules for both data sets. These findings are significant as evidence in terms of accuracy that was achieved using the proposed mode
RETRACTED: Minimum makespan task scheduling algorithm in cloud computing
Get PDFRETRACTEDFollowing a rigorous, carefully concerns and considered review of the article published in International Journal of Advances in Intelligent Informatics to article entitled “Minimum makespan task scheduling algorithm in cloud computing†Vol 2, No 3, pp. 123-130, November 2016, DOI: http://dx.doi.org/10.26555/ijain.v2i3.59.This paper has been found to be in violation of the International Journal of Advances in Intelligent Informatics Publication principles and has been retracted.The article contained redundant material, the editor investigated and found that the paper published in International Journal of Grid and Distributed Computing, Vol. 9, No. 11, pp. 61-70, 2016, DOI: http://dx.doi.org/10.14257/ijgdc.2016.9.11.05.The document and its content has been removed from International Journal of Advances in Intelligent Informatics, and reasonable effort should be made to remove all references to this article
K-means clustering based filter feature selection on high dimensional data
Get PDFWith hundreds or thousands of features in high dimensional data, computational workload is challenging. In classification process, features which do not contribute significantly to prediction of classes, add to the computational workload. Therefore the aim of this paper is to use feature selection to decrease the computation load by reducing the size of high dimensional data. Selecting subsets of features which represent all features were used. Hence the process is two-fold; discarding irrelevant data and choosing one feature that representing a number of redundant features. There have been many studies regarding feature selection, for example backward feature selection and forward feature selection. In this study, a k-means clustering based feature selection is proposed. It is assumed that redundant features are located in the same cluster, whereas irrelevant features do not belong to any clusters. In this research, two different high dimensional datasets are used: 1) the Human Activity Recognition Using Smartphones (HAR) Dataset, containing 7352 data points each of 561 features and 2) the National Classification of Economic Activities Dataset, which contains 1080 data points each of 857 features. Both datasets provide class label information of each data point. Our experiment shows that k-means clustering based feature selection can be performed to produce subset of features. The latter returns more than 80% accuracy of classification result
Critical analysis of classification techniques for polarimetric synthetic aperture radar data
Get PDFFull polarimetry SAR data known as PolSAR contains information in terms of microwave energy backscattered through different scattering mechanisms (surface-, double- and volume-scattering) by the targets on the surface of land. These scattering mechanisms information is different in different features. Similarly, different classifiers have different capabilities as far as identification of the targets corresponding to these scattering mechanisms. Extraction of different features and the role of classifier are important for the purpose of identifying which feature is the most suitable with which classifier for land cover classification. Selection of suitable features and their combinations have always been an active area of research for the development of advanced classification algorithms. Fully polarimetric data has its own advantages because its different channels give special scattering feature for various land cover. Therefore, first hand statistics HH, HV and VV of PolSAR data along with their ratios and linear combinations should be investigated for exploring their importance vis-Ã -vis relevant classifier for land management at the global scale. It has been observed that individually first hand statistics yield low accuracies. And their ratios are also not improving the results either. However, improved accuracies are achieved when these natural features are stacked together
Forecasting electricity load demand using hybrid exponential smoothing-artificial neural network model
Get PDFShort-term electricity load demand forecast is a vital requirements for power systems. This research considers the combination of exponential smoothing for double seasonal patterns and neural network model. The linear version of Holt-Winter method is extended to accommodate a second seasonal component. In this work, the Fourier with time varying coefficient is presented as a means of seasonal extraction. The methodological contribution of this paper is to demonstrate how these methods can be adapted to model the time series data with multiple seasonal pattern, correlated non stationary error and nonlinearity components together. The proposed hybrid model is started by implementing exponential smoothing state space model to obtain the level, trend, seasonal and irregular components and then use them as inputs of neural network. Forecasts of future values are then can be obtained by using the hybrid model. The forecast performance was characterized by root mean square error and mean absolute percentage error. The proposed hybrid model is applied to two real load series that are energy consumption in Bawen substation and in Java-Bali area. Comparing with other existing models, results show that the proposed hybrid model generate the most accurate forecas
Comparative analysis of multiple target tracking methods
Get PDFMany applications such as intelligent transportation, video surveillance, robotics of computer vision mainly depend on task of multiple object tracking. It includes the process of detection, classifications and tracking. The main focus of the study is to develop an efficient and effective multiple target tracking methods to solve the issues of illumination changes, occlusions and affinity matching. Accordingly, the various multiple target tracking methods are tested and evaluated using the metrics on publicly available datasets from which it is obvious that the outcome of the global energy minimization and optimization techniques is comparatively better than any other existing techniques in all aspects. This comparative study work will also help in better understanding of the problem, knowledge of the methods and experimental evaluation skill for further research works
Signature recognition using neural network probabilistic
No full textThe signature of each person is different and has unique characteristics. Thus, this paper discusses the development of a personal identification system based on it is unique digital signature. The process of preprocessing used gray scale method, while Shannon Entropy and Probabilistic Neural Network are used respectively for feature extraction and identification. This study uses five signature types with five signatures in every type. While the test results compared to actual data compared to real data, the proposed system performance was only 40%