Parrot: A Progressive Analysis System on Large Text Collections

Zhang, Yazhong; Zhang, Hanbing; He, Zhenying; Jing, Yinan; Zhang, Kai; Wang, X. Sean

doi:10.1007/s41019-020-00144-y

Cited by 6 publications

(2 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In contrast, the machine learning imputation approaches are to train parametric models in machine learning [23], [24] to estimate the missing values, including decision tree models like XGBoost imputation [25], MissFI (MissForest imputation) [11], and Baran [26], and regression models like MICE (multivariate imputation by chained equations) [8] and imputation via individual model [27]. These imputation methods employ the batch gradient descent techniques [28] to calculate the model gradient over the entire dataset.…”

Section: A Existing Imputation Methodsmentioning

confidence: 99%

Differentiable and Scalable Generative Adversarial Models for Data Imputation

Wu¹,

Wang²,

Miao³

et al. 2022

Preprint

View full text Add to dashboard Cite

Data imputation has been extensively explored to solve the missing data problem. The dramatically increasing volume of incomplete data makes the imputation models computationally infeasible in many real-life applications. In this paper, we propose an effective scalable imputation system named SCIS to significantly speed up the training of the differentiable generative adversarial imputation models under accuracyguarantees for large-scale incomplete data. SCIS consists of two modules, differentiable imputation modeling (DIM) and sample size estimation (SSE). DIM leverages a new masking Sinkhorn divergence function to make an arbitrary generative adversarial imputation model differentiable, while for such a differentiable imputation model, SSE can estimate an appropriate sample size to ensure the user-specified imputation accuracy of the final model. Extensive experiments upon several real-life large-scale datasets demonstrate that, our proposed system can accelerate the generative adversarial model training by 7.1x. Using around 7.6% samples, SCIS yields competitive accuracy with the stateof-the-art imputation methods in much shorter computation time.Index Terms-Data imputation, generative adversarial network, differentiable model, large-scale incomplete data 1 The support of a distribution is the complement of the largest open set of the random variables having zero-probability.

show abstract

Section: A Existing Imputation Methodsmentioning

confidence: 99%

Differentiable and Scalable Generative Adversarial Models for Data Imputation

Wu¹,

Wang²,

Miao³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…In contrast, the parametric imputation solution is to train a parametric model [44,47] (i.e., machine learning one or deep learning one) to learn the data distribution from the observed data. This solution employs the gradient descent techniques to estimate missing values.…”

Section: Background 21 Existing Imputation Methodsmentioning

confidence: 99%

Efficient and effective data imputation with influence functions

Miao

Chen

et al. 2021

Proc. VLDB Endow.

View full text Add to dashboard Cite

Data imputation has been extensively explored to solve the missing data problem. The dramatically rising volume of missing data makes the training of imputation models computationally infeasible in real-life scenarios. In this paper, we propose an efficient and effective data imputation system with influence functions , named EDIT, which quickly trains a parametric imputation model with representative samples under imputation accuracy guarantees. EDIT mainly consists of two modules, i.e., an imputation influence evaluation (IIE) module and a representative sample selection (RSS) module. IIE leverages the influence functions to estimate the effect of (in)complete samples on the prediction result of parametric imputation models. RSS builds a minimum set of the high-effect samples to satisfy a user-specified imputation accuracy. Moreover, we introduce a weighted loss function that drives the parametric imputation model to pay more attention on the high-effect samples. Extensive experiments upon ten state-of-the-art imputation methods demonstrate that, EDIT adopts only about 5% samples to speed up the model training by 4x in average with more than 11% accuracy gain.

show abstract