• Title/Summary/Keyword: series model

Search Result 5,383, Processing Time 0.034 seconds

A Study of Anomaly Detection for ICT Infrastructure using Conditional Multimodal Autoencoder (ICT 인프라 이상탐지를 위한 조건부 멀티모달 오토인코더에 관한 연구)

  • Shin, Byungjin;Lee, Jonghoon;Han, Sangjin;Park, Choong-Shik
    • Journal of Intelligence and Information Systems
    • /
    • v.27 no.3
    • /
    • pp.57-73
    • /
    • 2021
  • Maintenance and prevention of failure through anomaly detection of ICT infrastructure is becoming important. System monitoring data is multidimensional time series data. When we deal with multidimensional time series data, we have difficulty in considering both characteristics of multidimensional data and characteristics of time series data. When dealing with multidimensional data, correlation between variables should be considered. Existing methods such as probability and linear base, distance base, etc. are degraded due to limitations called the curse of dimensions. In addition, time series data is preprocessed by applying sliding window technique and time series decomposition for self-correlation analysis. These techniques are the cause of increasing the dimension of data, so it is necessary to supplement them. The anomaly detection field is an old research field, and statistical methods and regression analysis were used in the early days. Currently, there are active studies to apply machine learning and artificial neural network technology to this field. Statistically based methods are difficult to apply when data is non-homogeneous, and do not detect local outliers well. The regression analysis method compares the predictive value and the actual value after learning the regression formula based on the parametric statistics and it detects abnormality. Anomaly detection using regression analysis has the disadvantage that the performance is lowered when the model is not solid and the noise or outliers of the data are included. There is a restriction that learning data with noise or outliers should be used. The autoencoder using artificial neural networks is learned to output as similar as possible to input data. It has many advantages compared to existing probability and linear model, cluster analysis, and map learning. It can be applied to data that does not satisfy probability distribution or linear assumption. In addition, it is possible to learn non-mapping without label data for teaching. However, there is a limitation of local outlier identification of multidimensional data in anomaly detection, and there is a problem that the dimension of data is greatly increased due to the characteristics of time series data. In this study, we propose a CMAE (Conditional Multimodal Autoencoder) that enhances the performance of anomaly detection by considering local outliers and time series characteristics. First, we applied Multimodal Autoencoder (MAE) to improve the limitations of local outlier identification of multidimensional data. Multimodals are commonly used to learn different types of inputs, such as voice and image. The different modal shares the bottleneck effect of Autoencoder and it learns correlation. In addition, CAE (Conditional Autoencoder) was used to learn the characteristics of time series data effectively without increasing the dimension of data. In general, conditional input mainly uses category variables, but in this study, time was used as a condition to learn periodicity. The CMAE model proposed in this paper was verified by comparing with the Unimodal Autoencoder (UAE) and Multi-modal Autoencoder (MAE). The restoration performance of Autoencoder for 41 variables was confirmed in the proposed model and the comparison model. The restoration performance is different by variables, and the restoration is normally well operated because the loss value is small for Memory, Disk, and Network modals in all three Autoencoder models. The process modal did not show a significant difference in all three models, and the CPU modal showed excellent performance in CMAE. ROC curve was prepared for the evaluation of anomaly detection performance in the proposed model and the comparison model, and AUC, accuracy, precision, recall, and F1-score were compared. In all indicators, the performance was shown in the order of CMAE, MAE, and AE. Especially, the reproduction rate was 0.9828 for CMAE, which can be confirmed to detect almost most of the abnormalities. The accuracy of the model was also improved and 87.12%, and the F1-score was 0.8883, which is considered to be suitable for anomaly detection. In practical aspect, the proposed model has an additional advantage in addition to performance improvement. The use of techniques such as time series decomposition and sliding windows has the disadvantage of managing unnecessary procedures; and their dimensional increase can cause a decrease in the computational speed in inference.The proposed model has characteristics that are easy to apply to practical tasks such as inference speed and model management.

Sufficient Conditions for Stationarity of Smooth Transition ARMA/GARCH Models

  • Lee, Oe-Sook
    • Journal of the Korean Data and Information Science Society
    • /
    • v.18 no.1
    • /
    • pp.237-245
    • /
    • 2007
  • Nonlinear asymmetric time series models have the growing interest in econometrics and finance. Threshold model is one of the successful asymmetric model. We consider a smooth transition ARMA model which converges a.s. to a threshold ARMA model and show that the smooth transition ARMA model admits a stationary measure, provided a suitable condition on the coefficients of the autoregressive parts of the different regimes is satisfied. Stationarity of a smooth transition GARCH model is also obtained.

  • PDF

Studies on the Variation Pattern of Water Resources and their Generation Models by Simulation Technique (Simulation Technique에 의한 수자원의 변동양상 및 그 모의발생모델에 관한 연구)

  • Lee, Sun-Tak;An, Gyeong-Su;Lee, Ui-Rak
    • Water for future
    • /
    • v.9 no.2
    • /
    • pp.87-100
    • /
    • 1976
  • These studies are aimed at the analysis of systematic variation pattern of water resources in Korean river catchments and the development of their simulation models from the stochastic analysis of monthly and annual hydrologic data as main elements of water resources, i.e. rainfall and streamflow. In the analysis, monthly & annual rainfall records in Soul, Taegu, Pusan and Kwangju and streamflow records at the main gauging stations in Han, Nakdong and Geum river were used. Firstly, the systematic variation pattern of annual streamflow was found by the exponential function relationship between their standard deviations and mean values of log-annual runoff. Secondly, stochastic characteristics of annual rainfall & streamflow series were studied by the correlogram Monte Carlo method and a single season model of 1st-order Markov type were applied and compared in the simulation of annual hydrologic series. In the simulation, single season model of Markov type showed better results than LN-model and the simulated data were fit well with historical data. But it was noticed that LN-model gave quite better results in the simulation of annual rainfall. Thirdly, stochastic characteristics of monthly rainfall & streamflow series were also studied by the correlogram and spectrum analysis, and then the Model-C, which was developed and applied for the synthesis of monthly perennial streamflow by lst author and is a Markov type model with transformed skewed random number, was used in the simulation of monthly hydrologic series. In the simulation, it was proved that Model-C was fit well for extended area in Korea and also applicable for menthly rainfall as well as monthly streamflow.

  • PDF

Development of the ice resistance series chart for icebreaking ships

  • Lee, Chun-Ju;Joung, Tae-Hwan;Lew, Jae-Moon
    • International Journal of Naval Architecture and Ocean Engineering
    • /
    • v.10 no.6
    • /
    • pp.794-802
    • /
    • 2018
  • The ice resistance series charts for icebreaking ships were developed through a series of systematic model tests in the ice tank of the Korean Research Institute of Ship and Ocean Engineering (KRISO). Spencer's (1992) component-based scaling system for ship-ice model tests was applied to extend the model ship correlations. Beam to draft ratio (B/T), length to beam ratio (L/B), block coefficient ($C_B$) and stem angle (${\alpha}$) were selected as geometric parameters for hull form development. The basic hull form (S1) of twin pod type with B/T of 3.0, L/B of 6.0, $C_B$ of 0.75 and stem angle of $25^{\circ}$ was generated with a modern hull design concept. A total of 13 hulls were designed varying the geometric parameters; B/T of 2.5 and 3.5, L/B of 5.0 and 7.0, $C_B$ from 0.65 to 0.85 in intervals of 0.05, and 5 stem angles from $15^{\circ}$ to $35^{\circ}$. Ice resistance tests were first carried out with the basic hull form in level ice with suitable speed. Four more tests for $C_B$ variations from 0.65 to 0.85 were conducted and two more for beam to draft and length to beam ratios were also performed to study the effect of the geometric parameters on ice resistance. Ice resistance tests were summarized using the volumetric coefficient, $C_V$ ($={\nabla}/L^3$), instead of L/B and $C_B$ variations. Additional model tests were also carried out to account for the effect of the stem angle, ice thickness and ice strength on ice resistance. In order to develop the ice resistance series charts with a minimum number of experiments, the trends of the ice resistance obtained from the experiments were assumed to be similar for other model ship with different geometric parameters. A total of 18 sheets composed of combinations of three different beam to draft ratios and six block coefficients were developed as a parameter of $C_V$ in the low speed regions. Three correction charts were also developed for stem angles, ice thickness and ice strength respectively. The charts were applied to estimate ice resistance for existing icebreaking ships including ARAON, and the results were satisfactory with reasonable accuracy.

Modelling and Residual Analysis for Water Level Series of Upo Wetland (우포늪 수위 자료의 시계열 모형화 및 잔차 분석)

  • Kim, Kyunghun;Han, Daegun;Kim, Jungwook;Lim, Jonghun;Lee, Jongso;Kim, Hung Soo
    • Journal of Wetlands Research
    • /
    • v.21 no.1
    • /
    • pp.66-76
    • /
    • 2019
  • Recently, natural disasters such as floods and droughts are frequently occurred due to climate change and the damage is also increasing. Wetland is known to play an important role in reducing and minimizing the damage. In particular, water level variability needs to be analyzed in order to understand the various functions of wetland as well as the reduction of damage caused by natural disaster. Therefore, in this study, we fitted water level series of Upo wetland in Changnyeong, Gyeongnam province to a proper time series model and residual test was performed to confirm the appropriateness of the model. In other words, ARIMA model was constructed and its residual tests were performed using existing nonparametric statistics, BDS statistic, and Close Returns Histogram(CRH). The results of residual tests were compared and especially, we showed the applicability of CRH to analyze the residuals of time series model. As a result, CRH produced not only accurate randomness test result, but also produced result in a simple calculation process compared to the other methods. Therefore, we have shown that CRH and BDS statistic can be effective tools for analyzing residual in time series model.

The Prediction of Cryptocurrency on Using Text Mining and Deep Learning Techniques : Comparison of Korean and USA Market (텍스트 마이닝과 딥러닝을 활용한 암호화폐 가격 예측 : 한국과 미국시장 비교)

  • Won, Jonggwan;Hong, Taeho
    • Knowledge Management Research
    • /
    • v.22 no.2
    • /
    • pp.1-17
    • /
    • 2021
  • In this study, we predicted the bitcoin prices of Bithum and Coinbase, a leading exchange in Korea and USA, using ARIMA and Recurrent Neural Networks(RNNs). And we used news articles from each country to suggest a separated RNN model. The suggested model identifies the datasets based on the changing trend of prices in the training data, and then applies time series prediction technique(RNNs) to create multiple models. Then we used daily news data to create a term-based dictionary for each trend change point. We explored trend change points in the test data using the daily news keyword data of testset and term-based dictionary, and apply a matching model to produce prediction results. With this approach we obtained higher accuracy than the model which predicted price by applying just time series prediction technique. This study presents that the limitations of the time series prediction techniques could be overcome by exploring trend change points using news data and various time series prediction techniques with text mining techniques could be applied to improve the performance of the model in the further research.

Development of Fault Detector for Series Arc Fault in Low Voltage DC Distribution System using Wavelet Singular Value Decomposition and State Diagram

  • Oh, Yun-Sik;Han, Joon;Gwon, Gi-Hyeon;Kim, Doo-Ung;Kim, Chul-Hwan
    • Journal of Electrical Engineering and Technology
    • /
    • v.10 no.3
    • /
    • pp.766-776
    • /
    • 2015
  • It is well known that series arc faults in Low Voltage DC (LVDC) distribution system occur at unintended points of discontinuity within an electrical circuit. These faults can make circuit breakers not respond timely due to low fault current. It, therefore, is needed to detect the series fault for protecting circuits from electrical fires. This paper proposes a novel scheme to detect the series arc fault using Wavelet Singular Value Decomposition (WSVD) and state diagram. In this paper, the fault detector developed is designed by using three criterion factors based on the RMS value of Singular value of Approximation (SA), Sum of the absolute value of Detail (SD), and state diagram. LVDC distribution system including AC/DC and DC/DC converter is modeled to verify the proposed scheme using ElectroMagnetic Transient Program (EMTP) software. EMTP/MODELS is also utilized to implement the series arc model and WSVD. Simulation results according to various conditions clearly show the effectiveness of the proposed scheme.

A Study of Air Freight Forecasting Using the ARIMA Model (ARIMA 모델을 이용한 항공운임예측에 관한 연구)

  • Suh, Sang-Sok;Park, Jong-Woo;Song, Gwangsuk;Cho, Seung-Gyun
    • Journal of Distribution Science
    • /
    • v.12 no.2
    • /
    • pp.59-71
    • /
    • 2014
  • Purpose - In recent years, many firms have attempted various approaches to cope with the continual increase of aviation transportation. The previous research into freight charge forecasting models has focused on regression analyses using a few influence factors to calculate the future price. However, these approaches have limitations that make them difficult to apply into practice: They cannot respond promptly to small price changes and their predictive power is relatively low. Therefore, the current study proposes a freight charge-forecasting model using time series data instead a regression approach. The main purposes of this study can thus be summarized as follows. First, a proper model for freight charge using the autoregressive integrated moving average (ARIMA) model, which is mainly used for time series forecast, is presented. Second, a modified ARIMA model for freight charge prediction and the standard process of determining freight charge based on the model is presented. Third, a straightforward freight charge prediction model for practitioners to apply and utilize is presented. Research design, data, and methodology - To develop a new freight charge model, this study proposes the ARIMAC(p,q) model, which applies time difference constantly to address the correlation coefficient (autocorrelation function and partial autocorrelation function) problem as it appears in the ARIMA(p,q) model and materialize an error-adjusted ARIMAC(p,q). Cargo Account Settlement Systems (CASS) data from the International Air Transport Association (IATA) are used to predict the air freight charge. In the modeling, freight charge data for 72 months (from January 2006 to December 2011) are used for the training set, and a prediction interval of 23 months (from January 2012 to November 2013) is used for the validation set. The freight charge from November 2012 to November 2013 is predicted for three routes - Los Angeles, Miami, and Vienna - and the accuracy of the prediction interval is analyzed using mean absolute percentage error (MAPE). Results - The result of the proposed model shows better accuracy of prediction because the MAPE of the error-adjusted ARIMAC model is 10% and the MAPE of ARIMAC is 11.2% for the L.A. route. For the Miami route, the proposed model also shows slightly better accuracy in that the MAPE of the error-adjusted ARIMAC model is 3.5%, while that of ARIMAC is 3.7%. However, for the Vienna route, the accuracy of ARIMAC is better because the MAPE of ARIMAC is 14.5% and the MAPE of the error-adjusted ARIMAC model is 15.7%. Conclusions - The accuracy of the error-adjusted ARIMAC model appears better when a route's freight charge variance is large, and the accuracy of ARIMA is better when the freight charge variance is small or has a trend of ascent or descent. From the results, it can be concluded that the ARIMAC model, which uses moving averages, has less predictive power for small price changes, while the error-adjusted ARIMAC model, which uses error correction, has the advantage of being able to respond to price changes quickly.

A Hybrid System of Joint Time-Frequency Filtering Methods and Neural Network Techniques for Foreign Exchange Rate Forecasting (환율예측을 위한 신호처리분석 및 인공신경망기법의 통합시스템 구축)

  • 신택수;한인구
    • Journal of Intelligence and Information Systems
    • /
    • v.5 no.1
    • /
    • pp.103-123
    • /
    • 1999
  • Input filtering as a preprocessing method is so much crucial to get good performance in time series forecasting. There are a few preprocessing methods (i.e. ARMA outputs as time domain filters, and Fourier transform or wavelet transform as time-frequency domain filters) for handling time series. Specially, the time-frequency domain filters describe the fractal structure of financial markets better than the time domain filters due to theoretically additional frequency information. Therefore, we, first of all, try to describe and analyze specially some issues on the effectiveness of different filtering methods from viewpoint of the performance of a neural network based forecasting. And then we discuss about neural network model architecture issues, for example, what type of neural network learning architecture is selected for our time series forecasting, and what input size should be applied to a model. In this study an input selection problem is limited to a size selection of the lagged input variables. To solve this problem, we simulate on analyzing and comparing a few neural networks having different model architecture and also use an embedding dimension measure as chaotic time series analysis or nonlinear dynamic analysis to reduce the dimensionality (i.e. the size of time delayed input variables) of the models. Throughout our study, experiments for integration methods of joint time-frequency analysis and neural network techniques are applied to a case study of daily Korean won / U. S dollar exchange returns and finally we suggest an integration framework for future research from our experimental results.

  • PDF