scholarly journals Privacy-Preserving Stacking with Application to Cross-organizational Diabetes Prediction

2019 ◽  
Author(s):  
Quanming Yao ◽  
Xiawei Guo ◽  
James Kwok ◽  
Weiwei Tu ◽  
Yuqiang Chen ◽  
...  
Keyword(s):  
Transfer Learning ◽  
Differential Privacy ◽  
Original Data ◽  
Privacy Preserving ◽  
Data Sets ◽  
Data Set ◽  
Predicting Performance ◽  
Empirical Performance ◽  
Feature Based ◽  
Diabetes Prediction

To meet the standard of differential privacy, noise is usually added into the original data, which inevitably deteriorates the predicting performance of subsequent learning algorithms. In this paper, motivated by the success of improving predicting performance by ensemble learning, we propose to enhance privacy-preserving logistic regression by stacking. We show that this can be done either by sample-based or feature-based partitioning. However, we prove that when privacy-budgets are the same, feature-based partitioning requires fewer samples than sample-based one, and thus likely has better empirical performance. As transfer learning is difficult to be integrated with a differential privacy guarantee, we further combine the proposed method with hypothesis transfer learning to address the problem of learning across different organizations. Finally, we not only demonstrate the effectiveness of our method on two benchmark data sets, i.e., MNIST and NEWS20, but also apply it into a real application of cross-organizational diabetes prediction from RUIJIN data set, where privacy is of a significant concern.

scholarly journals Classification of jujube defects in small data sets based on transfer learning

2021 ◽  
Author(s):  
Jianping Ju ◽  
Hong Zheng ◽  
Xiaohang Xu ◽  
Zhongyuan Guo ◽  
Zhaohui Zheng ◽  
...  
Keyword(s):  
Transfer Learning ◽  
Loss Function ◽  
Training Model ◽  
Parameter Distribution ◽  
Test Accuracy ◽  
Small Data ◽  
Data Sets ◽  
Data Set ◽  
Small Data Sets

AbstractAlthough convolutional neural networks have achieved success in the field of image classification, there are still challenges in the field of agricultural product quality sorting such as machine vision-based jujube defects detection. The performance of jujube defect detection mainly depends on the feature extraction and the classifier used. Due to the diversity of the jujube materials and the variability of the testing environment, the traditional method of manually extracting the features often fails to meet the requirements of practical application. In this paper, a jujube sorting model in small data sets based on convolutional neural network and transfer learning is proposed to meet the actual demand of jujube defects detection. Firstly, the original images collected from the actual jujube sorting production line were pre-processed, and the data were augmented to establish a data set of five categories of jujube defects. The original CNN model is then improved by embedding the SE module and using the triplet loss function and the center loss function to replace the softmax loss function. Finally, the depth pre-training model on the ImageNet image data set was used to conduct training on the jujube defects data set, so that the parameters of the pre-training model could fit the parameter distribution of the jujube defects image, and the parameter distribution was transferred to the jujube defects data set to complete the transfer of the model and realize the detection and classification of the jujube defects. The classification results are visualized by heatmap through the analysis of classification accuracy and confusion matrix compared with the comparison models. The experimental results show that the SE-ResNet50-CL model optimizes the fine-grained classification problem of jujube defect recognition, and the test accuracy reaches 94.15%. The model has good stability and high recognition accuracy in complex environments.

Auto-sharing parameters for transfer learning based on multi-objective optimization

2021 ◽  
pp. 1-13
Author(s):  
Hailin Liu ◽  
Fangqing Gu ◽  
Zixian Lin
Keyword(s):  
Transfer Learning ◽  
Optimization Problem ◽  
Data Sets ◽  
Multi Objective Optimization ◽  
Particle Swarm Optimizer ◽  
Real World Data ◽  
Data Set ◽  
Target Task ◽  
Main Research ◽  
Multi Objective

Transfer learning methods exploit similarities between different datasets to improve the performance of the target task by transferring knowledge from source tasks to the target task. “What to transfer” is a main research issue in transfer learning. The existing transfer learning method generally needs to acquire the shared parameters by integrating human knowledge. However, in many real applications, an understanding of which parameters can be shared is unknown beforehand. Transfer learning model is essentially a special multi-objective optimization problem. Consequently, this paper proposes a novel auto-sharing parameter technique for transfer learning based on multi-objective optimization and solves the optimization problem by using a multi-swarm particle swarm optimizer. Each task objective is simultaneously optimized by a sub-swarm. The current best particle from the sub-swarm of the target task is used to guide the search of particles of the source tasks and vice versa. The target task and source task are jointly solved by sharing the information of the best particle, which works as an inductive bias. Experiments are carried out to evaluate the proposed algorithm on several synthetic data sets and two real-world data sets of a school data set and a landmine data set, which show that the proposed algorithm is effective.

Bayesian Classifier for Sparsity-Promoting Feature Selection

2015 ◽  
Vol 29 (06) ◽  
pp. 1550022 ◽  
Author(s):  
Danlei Xu ◽  
Lan Du ◽  
Hongwei Liu ◽  
Penghui Wang
Keyword(s):  
Feature Selection ◽  
Synthetic Data ◽  
Original Data ◽  
Radar Data ◽  
Bayesian Classifier ◽  
Classification Model ◽  
Data Sets ◽  
Data Set ◽  
Classification Boundary ◽  
Nonlinear Mappings

A Bayesian classifier for sparsity-promoting feature selection is developed in this paper, where a set of nonlinear mappings for the original data is performed as a pre-processing step. The linear classification model with such mappings from the original input space to a nonlinear transformation space can not only construct the nonlinear classification boundary, but also realize the feature selection for the original data. A zero-mean Gaussian prior with Gamma precision and a finite approximation of Beta process prior are used to promote sparsity in the utilization of features and nonlinear mappings in our model, respectively. We derive the Variational Bayesian (VB) inference algorithm for the proposed linear classifier. Experimental results based on the synthetic data set, measured radar data set, high-dimensional gene expression data set, and several benchmark data sets demonstrate the aggressive and robust feature selection capability and comparable classification accuracy of our method comparing with some other existing classifiers.

scholarly journals Optimization through Bayesian Classification on the k-Anonymized Data

2012 ◽  
pp. 245-249
Author(s):  
L Mohana Tirumala ◽  
S. Srinivasa Rao
Keyword(s):  
Cost Effective ◽  
Privacy Preserving ◽  
Bayesian Classification ◽  
Data Sets ◽  
Data Set ◽  
Equal Importance ◽  
Cost Effective Approach ◽  
Effective Manner ◽  
Anonymized Data ◽  
Refinement Algorithm

Privacy preserving in Data mining & publishing, plays a major role in today networked world. It is important to preserve the privacy of the vital information corresponding to a data set. This process can be achieved by k-anonymization solution for classification. Along with the privacy preserving using anonymization, yielding the optimized data sets is also of equal importance with a cost effective approach. In this paper Top-Down Refinement algorithm has been proposed which yields optimum results in a cost effective manner. Bayesian Classification has been proposed in this paper to predict class membership probabilities for a data tuple for which the associated class label is unknown.

Feature-Based Uncertainty Visualization

Big Data ◽  
2016 ◽  
pp. 261-287
Author(s):  
Keqin Wu ◽  
Song Zhang
Keyword(s):  
Scientific Data ◽  
Data Sets ◽  
Data Set ◽  
Uncertainty Visualization ◽  
Scalar Data ◽  
Critical Issues ◽  
Contour Level ◽  
Feature Based ◽  
2D Data ◽  
The Impact

While uncertainty in scientific data attracts an increasing research interest in the visualization community, two critical issues remain insufficiently studied: (1) visualizing the impact of the uncertainty of a data set on its features and (2) interactively exploring 3D or large 2D data sets with uncertainties. In this chapter, a suite of feature-based techniques is developed to address these issues. First, an interactive visualization tool for exploring scalar data with data-level, contour-level, and topology-level uncertainties is developed. Second, a framework of visualizing feature-level uncertainty is proposed to study the uncertain feature deviations in both scalar and vector data sets. With quantified representation and interactive capability, the proposed feature-based visualizations provide new insights into the uncertainties of both data and their features which otherwise would remain unknown with the visualization of only data uncertainties.

Differential privacy preserving clustering using Daubechies-2 wavelet transform

2015 ◽  
Vol 13 (04) ◽  
pp. 1550028 ◽  
Author(s):  
Mohammad Reza Ebrahimi Dishabi ◽  
Mohammad Abdollahi Azgomi
Keyword(s):  
Wavelet Transform ◽  
Clustering Analysis ◽  
Differential Privacy ◽  
Original Data ◽  
Privacy Preserving ◽  
The Other ◽  
High Dimensionality ◽  
Worst Case ◽  
Strong Notion

Most of the existing privacy preserving clustering (PPC) algorithms do not consider the worst case privacy guarantees and are based on heuristic notions. In addition, these algorithms do not run efficiently in the case of high dimensionality of data. In this paper, to alleviate these challenges, we propose a new PPC algorithm, which is based on Daubechies-2 wavelet transform (D2WT) and preserves the differential privacy notion. Differential privacy is the strong notion of privacy, which provides the worst case privacy guarantees. On the other hand, most of the existing differential-based PPC algorithms generate data with poor utility. If we apply differential privacy properties over the original raw data, the resulting data will offer lower quality of clustering (QOC) during the clustering analysis. Therefore, we use D2WT for the preprocessing of the original data before adding noise to the data. By applying D2WT to the original data, the resulting data not only contains lower dimension compared to the original data, but also can provide differential privacy guarantee with high QOC due to less noise addition. The proposed algorithm has been implemented and experimented over some well-known datasets. We also compare the proposed algorithm with some recently introduced algorithms based on utility and privacy degrees.

Inverse Parametric Modeling From Independently Generated Product Data Sets

2002 ◽  
Author(s):  
Zhengdong Huang ◽  
Derek Yip-Hoi
Keyword(s):  
Product Family ◽  
Parametric Model ◽  
Parametric Modeling ◽  
Data Sets ◽  
Data Set ◽  
Product Families ◽  
Part Family ◽  
Product Data ◽  
Feature Based ◽  
Parameter Ranges

Parametric modeling has become a widely accepted mechanism for generating data set variants for product families. These data sets that include geometric models and feature-based process plans are created by specifying values for parameters within feasible ranges specified as constraints in the definition. The ranges denote the extent or envelope of the product family. Increasingly, with globalization the inverse problem is becoming important. This takes independently generated product data sets that on observation belong to the same product family and creates a parametric model for that family. This problem is also of relevance to large companies where independent design teams may work on product variants without much collaboration only to attempt consolidation later on to optimize the design of manufacturing processes and systems. In this paper we present a methodology for generating a feature-based part family parametric model through merging independently generated product data sets. We assume that these data sets are feature-based with relationships such as precedences captured using graphs. Since there are typically numerous ways in which these data sets can be merged, we formulate this as an optimization problem and solve using the A* algorithm. The parameter ranges generated by this approach will be used to design appropriate Reconfigurable Machine Tools (RMTs) and systems (RMS) for manufacturing the resulting part family.

scholarly journals Dynamic spatio-temporal generation of large-scale synthetic gridded precipitation: with improved spatial coherence of extremes

2019 ◽  
Vol 34 (9) ◽  
pp. 1369-1383 ◽  
Author(s):  
Dirk Diederen ◽  
Ye Liu
Keyword(s):  
Large Scale ◽  
Spatial Coherence ◽  
Original Data ◽  
Return Level ◽  
Data Sets ◽  
Large Set ◽  
Precipitation Data ◽  
Data Set ◽  
Spatio Temporal ◽  
Synthetic Precipitation

Abstract With the ongoing development of distributed hydrological models, flood risk analysis calls for synthetic, gridded precipitation data sets. The availability of large, coherent, gridded re-analysis data sets in combination with the increase in computational power, accommodates the development of new methodology to generate such synthetic data. We tracked moving precipitation fields and classified them using self-organising maps. For each class, we fitted a multivariate mixture model and generated a large set of synthetic, coherent descriptors, which we used to reconstruct moving synthetic precipitation fields. We introduced randomness in the original data set by replacing the observed precipitation fields in the original data set with the synthetic precipitation fields. The output is a continuous, gridded, hourly precipitation data set of a much longer duration, containing physically plausible and spatio-temporally coherent precipitation events. The proposed methodology implicitly provides an important improvement in the spatial coherence of precipitation extremes. We investigate the issue of unrealistic, sudden changes on the grid and demonstrate how a dynamic spatio-temporal generator can provide spatial smoothness in the probability distribution parameters and hence in the return level estimates.

scholarly journals Ambulatory and Laboratory Stress Detection Based on Raw Electrocardiogram Signals Using a Convolutional Neural Network

Sensors ◽  
2019 ◽  
Vol 19 (20) ◽  
pp. 4408 ◽  
Author(s):  
Hyun-Myung Cho ◽  
Heesu Park ◽  
Suh-Yeon Dong ◽  
Inchan Youn
Keyword(s):  
Neural Network ◽  
Transfer Learning ◽  
Mental Arithmetic ◽  
Small Data ◽  
Data Sets ◽  
Learning Method ◽  
Data Set ◽  
Electrocardiogram Signals ◽  
Proposed Model ◽  
Small Data Set

The goals of this study are the suggestion of a better classification method for detecting stressed states based on raw electrocardiogram (ECG) data and a method for training a deep neural network (DNN) with a smaller data set. We suggest an end-to-end architecture to detect stress using raw ECGs. The architecture consists of successive stages that contain convolutional layers. In this study, two kinds of data sets are used to train and validate the model: A driving data set and a mental arithmetic data set, which smaller than the driving data set. We apply a transfer learning method to train a model with a small data set. The proposed model shows better performance, based on receiver operating curves, than conventional methods. Compared with other DNN methods using raw ECGs, the proposed model improves the accuracy from 87.39% to 90.19%. The transfer learning method improves accuracy by 12.01% and 10.06% when 10 s and 60 s of ECG signals, respectively, are used in the model. In conclusion, our model outperforms previous models using raw ECGs from a small data set and, so, we believe that our model can significantly contribute to mobile healthcare for stress management in daily life.

European household’s income, consumption and wealth

2020 ◽  
Vol 36 (4) ◽  
pp. 1175-1188
Author(s):  
Pierre Lamarche ◽  
Friderike Oehler ◽  
Irene Rioboo
Keyword(s):  
Impact Analysis ◽  
Well Being ◽  
Original Data ◽  
Data Sets ◽  
The European Union ◽  
Micro Data ◽  
Data Set ◽  
Statistical Matching ◽  
Full Picture ◽  
Poverty Indicators

Poverty indicators purely based on income statistics do not reflect the full picture of household’s economic well-being. Consumption and wealth are two additional key dimensions that determine the economic opportunities of people or material inequalities. We use non-parametric statistical matching methods to join consumption data from the Household Budget Survey to micro data from the European Union Statistics on Income and Living Conditions. In a second step, micro data from the Household Finance and Consumption Survey are joint to produce a common distribution of income, consumption and wealth variables. A variety of different indicators is then produced based on this joint data set, in particular household saving rates. Care has to be taken when interpreting the indicators, since the statistical matching is based on strong assumptions and a limited number of variables common to all of the three original data sets. We are able to show, however, that the assumptions made are justified by the use of strong proxies as matching variables. Thus, the resulting indicators have the potential to contribute to the analysis of inequality patterns and enhance the possibilities of social, and possibly fiscal, policy impact analysis.

Sign in / Sign up

Export Citation Format

Share Document