A Subspace Method Based on Data Generation Model with Class Information

2008 ◽  
pp. 547-555
Author(s):  
Minkook Cho ◽  
Dongwoo Yoon ◽  
Hyeyoung Park
Keyword(s):  
Generation Model ◽  
Data Generation ◽  
Subspace Method ◽  
Class Information

Intrusion detection of railway clearance from infrared images using generative adversarial networks

2020 ◽  
pp. 1-13
Author(s):  
Yundong Li ◽  
Yi Liu ◽  
Han Dong ◽  
Wei Hu ◽  
Chen Lin
Keyword(s):  
Intrusion Detection ◽  
Synthetic Data ◽  
Generative Adversarial Networks ◽  
Generation Model ◽  
Single Shot ◽  
Data Generation ◽  
Infrared Images ◽  
Adversarial Networks ◽  
Training Samples ◽  
Rgb Images

The intrusion detection of railway clearance is crucial for avoiding railway accidents caused by the invasion of abnormal objects, such as pedestrians, falling rocks, and animals. However, detecting intrusions using deep learning methods from infrared images captured at night remains a challenging task because of the lack of sufficient training samples. To address this issue, a transfer strategy that migrates daytime RGB images to the nighttime style of infrared images is proposed in this study. The proposed method consists of two stages. In the first stage, a data generation model is trained on the basis of generative adversarial networks using RGB images and a small number of infrared images, and then, synthetic samples are generated using a well-trained model. In the second stage, a single shot multibox detector (SSD) model is trained using synthetic data and utilized to detect abnormal objects from infrared images at nighttime. To validate the effectiveness of the proposed method, two groups of experiments, namely, railway and non-railway scenes, are conducted. Experimental results demonstrate the effectiveness of the proposed method, and an improvement of 17.8% is achieved for object detection at nighttime.

scholarly journals A Time-Series Data Generation Method to Predict Remaining Useful Life

Processes ◽  
2021 ◽  
Vol 9 (7) ◽  
pp. 1115
Author(s):  
Gilseung Ahn ◽  
Hyungseok Yun ◽  
Sun Hur ◽  
Si-Yeong Lim
Keyword(s):  
Time Series ◽  
Time Series Data ◽  
Remaining Useful Life ◽  
Series Data ◽  
Generation Model ◽  
Data Generation ◽  
Training Time ◽  
Symbolic Aggregate Approximation ◽  
Useful Life ◽  
Occurrence Patterns

Accurate predictions of remaining useful life (RUL) of equipment using machine learning (ML) or deep learning (DL) models that collect data until the equipment fails are crucial for maintenance scheduling. Because the data are unavailable until the equipment fails, collecting sufficient data to train a model without overfitting can be challenging. Here, we propose a method of generating time-series data for RUL models to resolve the problems posed by insufficient data. The proposed method converts every training time series into a sequence of alphabetical strings by symbolic aggregate approximation and identifies occurrence patterns in the converted sequences. The method then generates a new sequence and inversely transforms it to a new time series. Experiments with various RUL prediction datasets and ML/DL models verified that the proposed data-generation model can help avoid overfitting in RUL prediction model.

scholarly journals Optimal Design of Manufacturing Cells Considering Machine Usage Percentage

2020 ◽  
Vol 19 (03) ◽  
pp. 411-423
Author(s):  
Tamal Ghosh
Keyword(s):  
Mathematical Model ◽  
Design Methodology ◽  
Cellular Manufacturing ◽  
Mathematical Formulation ◽  
Total Cell ◽  
Generation Model ◽  
Data Generation ◽  
Generation Technique ◽  
Ratio Data ◽  
Objective Model

This paper demonstrates an exclusive design methodology in Cellular Manufacturing (CM) considering machine usage percentage as ratio data. This research correctly emphasized the fundamental of ratio data and proposed a novel and precise mathematical formulation of the design problem. This multi-objective model carefully optimizes the total exceptional utilization (TEU), number of voids and total cell utilization (TCU). Due to the novelty in the model, a new data generation technique is proposed. The test datasets are obtained and tested using IBM CPLEX tool successfully. The contribution of this research is twofold. First, the ratio data concept is correctly emphasized and a precise mathematical model is developed. Second, since the model is new and datasets are not readily available, therefore a dedicated data generation model is proposed.

Using Bifactor Models to Examine the Predictive Validity of Hierarchical Constructs: Pros, Cons, and Solutions

2020 ◽  
pp. 109442812091552
Author(s):  
Bo Zhang ◽  
Tianjun Sun ◽  
Mengyang Cao ◽  
Fritz Drasgow
Keyword(s):  
Monte Carlo ◽  
Predictive Validity ◽  
Research Community ◽  
Incremental Validity ◽  
Generation Model ◽  
Data Generation ◽  
The Past ◽  
Bifactor Models ◽  
Augmentation Strategy ◽  
Practical Recommendations

The use of bifactor models has increased substantially in the past decade. However, bifactor models are prone to a nonidentification problem in the context of prediction that is not well recognized in the general research community. Moreover, the practical consequences of adopting different conceptualizations of hierarchical constructs when examining their predictive validity has received little attention. Therefore, Study 1 examined the statistical performance of bifactor models and investigated the effectiveness of an augmentation strategy to remedy the nonidentification problem. Monte Carlo simulations showed that the augmentation strategy is effective. The second simulation study demonstrated that researchers may arrive at different conclusions regarding the predictive validity of hierarchical constructs depending on their choice of models. In general, augmented bifactor models, which are restricted variants of the more general bifactor-(S·I-1) model, reasonably recovered the overall predictive validity ( R2) of hierarchical constructs and led to correct substantive conclusions regarding the incremental validity of facets regardless of the true data-generation model given a sufficiently large sample ( n ≥ 600). The authors discussed implications of those findings and made practical recommendations for further users of bifactor models.

scholarly journals Development of an estimation-experimental method for estimating the preservation indicators of single-use articles

2021 ◽  
Vol 6 (3 (114)) ◽  
pp. 18-35
Author(s):  
Boris Lanetskii ◽  
Vadym Lukianchuk ◽  
Igor Koval ◽  
Hennadii Khudov ◽  
Andrii Hordiienko ◽  
...  
Keyword(s):  
Distribution Function ◽  
Experimental Method ◽  
Continuous Monitoring ◽  
Parametric Estimation ◽  
Generation Model ◽  
Data Generation ◽  
Censored Samples ◽  
Operational Tests ◽  
Single Use ◽  
Non Parametric Estimation

To manage the operation of modern single-use products, it is necessary to evaluate their preservation indicators as uncontrolled, non-repairable, and maintenance-free objects. Data for assessing its parameters are considered as one-time censored samples with continuous monitoring, which does not correspond to the mode of storage of products during operation. Under the conditions of limited volumes of censored samples, it is problematic to identify the parametric model of persistence. To solve this problem, a non-parametric estimation-experimental method has been devised, which is a set of models for data generation, estimation of the function of the distribution of the preservation period and preservation indicators. The data generation model is represented by a scheme of operational tests and analytical relationships between the quantities of tested and failed articles. The model of estimating the distribution function describes the process of its construction on the generated data. Models for estimating preservation indicators are represented by ratios for their point and interval estimates, as functionals from the restored distribution function. Unlike the well-known ones, the developed method implements the assessment of indicators under the conditions of combined censorship. The method can be used to assess the preservation indicators of single-use articles with an error of at least 7 %. At the same time, their lower confidence limits are estimated at 0.9 with an error not worse than 14 % with a censorship degree of not more than 0.23. The restored distribution function agrees well (reliability 0.9, error 0.1) with the actual persistence of articles with censorship degrees of not more than 0.73, which is acceptable for solving the problems of managing their operation.

scholarly journals Data Augmentation Based on Adversarial Autoencoder Handling Imbalance for Learning to Rank

2019 ◽  
Vol 33 ◽  
pp. 411-418
Author(s):  
Qian Yu ◽  
Wai Lam
Keyword(s):  
Data Augmentation ◽  
Learning To Rank ◽  
Ensemble Methods ◽  
Limiting Factor ◽  
Generation Model ◽  
Training Procedure ◽  
Data Generation ◽  
Data Imbalance ◽  
Imbalance Problem ◽  
Latent Representations

Data imbalance is a key limiting factor for Learning to Rank (LTR) models in information retrieval. Resampling methods and ensemble methods cannot handle the imbalance problem well since none of them incorporate more informative data into the training procedure of LTR models. We propose a data generation model based on Adversarial Autoencoder (AAE) for tackling the data imbalance in LTR via informative data augmentation. This model can be utilized for handling two types of data imbalance, namely, imbalance regarding relevance levels for a particular query and imbalance regarding the amount of relevance judgements in different queries. In the proposed model, relevance information is disentangled from the latent representations in this AAE-based model in order to reconstruct data with specific relevance levels. The semantic information of queries, derived from word embeddings, is incorporated in the adversarial training stage for regularizing the distribution of the latent representation. Two informative data augmentation strategies suitable for LTR are designed utilizing the proposed data generation model. Experiments on benchmark LTR datasets demonstrate that our proposed framework can significantly improve the performance of LTR models.

AN EFFECTIVE DISTRIBUTED MODEL FOR XMLISED POWER SYSTEM DATA GENERATION

2012 ◽  
Vol 21 (07) ◽  
pp. 1250053
Author(s):  
GOMATHI VENUGOPAL ◽  
ANBALAGAN PALANIMUTHU ◽  
BALASINGH MOSES
Keyword(s):  
Power Systems ◽  
Power System ◽  
Software Design ◽  
Distributed Model ◽  
Generation Model ◽  
Loose Coupling ◽  
Data Generation ◽  
Text File ◽  
System Data ◽  
And Control

One of the outcomes of the continuous research on the evolution of distributed computing is the Web services. The aim of this paper is to represent Power System data effectively in XML in order to improve the interoperability and to develop an enhanced distributed model for unique XMLised Power System Data generation for solving various Power System applications in heterogeneous environment. Power System industries are now increasingly becoming privatized and hence the system data is becoming increasingly distributed, with more constrained and complex operational and control requirements. Because of the complex physical connectivity of the power systems, all levels of industry like generation, transmission, distribution and market need proper operational and equipmental data. As expected, the data to be shared between different power system applications is huge and hence it is vital to have an efficient and reliable data generation model to reduce more human efforts and to have the data in a secure and compatible form. The developed JAX-RPC-based model has the capability to generate the data dynamically in XML, fetching the power system data from various sources such as database, text file, etc. The standards such as XML and SOAP enable software design based on loose coupling which reduces restriction and eliminates similarity requirement between coordinating applications.

Sign in / Sign up

Export Citation Format

Share Document