Performance Evaluation of Missing-Value Imputation Clustering Based on a Multivariate Gaussian Mixture Model

Xiao, Jing; Xu, Qi; Wu, Chuanli; Gao, Yuexia; Hua, Tianqi; Xu, Chenwu

doi:10.1371/journal.pone.0161112

Cited by 10 publications

(7 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These parameters are then applied in computing the probability of each observation. The best number of distributions to fit the data is also determined by minimizing the Akaike information criterion (AIC) (Xiao et al, 2016).…”

Section: Decision-making Algorithmmentioning

confidence: 99%

Dynamic modelling of lettuce transpiration for water status monitoring

Adeyemi

Peets

Domun

et al. 2018

Computers and Electronics in Agriculture

View full text Add to dashboard Cite

Real-time information on the plant water status is an important prerequisite for the precision irrigation management of crops. The plant transpiration has been shown to provide a good indication of its water status. In this paper, a novel plant water status monitoring framework based on the transpiration dynamics of greenhouse grown lettuce plants is presented. Experimental results indicated that lettuce plants experiencing adequate water supply transpired at a higher rate compared to plants experiencing a shortage in water supply. A data-driven model for predicting the transpiration dynamics of the plants was developed using a system identification approach. Results indicated that a second order discrete-time transfer function model with incoming radiation, vapour pressure deficit, and leaf area index as inputs sufficiently explained the dynamics with an average coefficient of determination of 2 = 0.93 ± 0.04. The parameters of the model were updated online and then applied in predicting the transpiration dynamics of the plants in real-time. The model predicted dynamics closely matched the measured values when the plants were in a predefined water status state. The reverse was the case when there was a significant change in the water status state. The information contained in the model residuals (measured transpirationmodel predicted transpiration) was then exploited as a means of inferring the plant water status. This framework provides a simple and intuitive means of monitoring the plant water status in real-time while achieving a sensitivity similar to that of stomatal conductance measurements. It can be applied in regulating the water deficit of greenhouse grown crops, with specific advantages over other available techniques.

show abstract

Section: Decision-making Algorithmmentioning

confidence: 99%

Dynamic modelling of lettuce transpiration for water status monitoring

Adeyemi

Peets

Domun

et al. 2018

Computers and Electronics in Agriculture

View full text Add to dashboard Cite

show abstract

“…Therefore, methods have been proposed to perform cluster analysis in the presence of missing data (Plaehn, 2019). First, likelihood‐based clustering procedures appear particularly well suited to be adapted to the presence of missing data (Hunt & Jorgensen, 2003; Xiao et al., 2016). Several variations of the widely used K‐means clustering algorithm have also been developed (e.g., Chi, Chi, & Baraniuk, 2016's K‐POD algorithm).…”

Section: Introductionmentioning

confidence: 99%

Clustering with missing and left‐censored data: A simulation study comparing multiple‐imputation‐based procedures

et al. 2020

View full text Add to dashboard Cite

Cluster analysis, commonly used to explore large biomedical datasets, can be challenging, notably due to missing data or left‐censored data induced by the sensitivity limits of the biochemical measurement method. Usually, complete‐case analysis, simple imputation, or stochastic simple imputation are applied before clustering. More recently, consensus methods following multiple imputation have been proposed. However, they ignore left‐censoring and do not allow the number of clusters to vary across the partitions of each imputed dataset. Here, we developed a consensus‐based clustering algorithm in which left‐censored data are taken into account using a modified multiple imputation method and the number of clusters is estimated for each imputed dataset. A simulation study was conducted to assess the performance in terms of the number of clusters, the percentage of unclassified observations, and the adjusted Rand index. The simulation results showed that the investigated method works well compared to several alternative approaches. A real‐world application in breast cancer patients showed that the proposed method may reveal novel clusters of patients.

show abstract

“…The algorithm is very subjective and requires the number of clusters which are specified in advance. Many clustering algorithms have been developed, such as grid-based [ 12 ], hierarchy-based [ 13 ], model-based [ 14 ], and density-based [ 15 ] clustering algorithms. The processing time of the grid clustering algorithm is related to the number of cells divided by each dimensional space, which reduces the quality and accuracy of clustering.…”

Section: Methodsmentioning

confidence: 99%

A Bayesian Failure Prediction Network Based on Text Sequence Mining and Clustering

Chang

Meng

et al. 2018

Entropy

View full text Add to dashboard Cite

The purpose of this paper is to predict failures based on textual sequence data. The current failure prediction is mainly based on structured data. However, there are many unstructured data in aircraft maintenance. The failure mentioned here refers to failure types, such as transmitter failure and signal failure, which are classified by the clustering algorithm based on the failure text. For the failure text, this paper uses the natural language processing technology. Firstly, segmentation and the removal of stop words for Chinese failure text data is performed. The study applies the word2vec moving distance model to obtain the failure occurrence sequence for failure texts collected in a fixed period of time. According to the distance, a clustering algorithm is used to obtain a typical number of fault types. Secondly, the failure occurrence sequence is mined using sequence mining algorithms, such as-PrefixSpan. Finally, the above failure sequence is used to train the Bayesian failure network model. The final experimental results show that the Bayesian failure network has higher accuracy for failure prediction.

show abstract

Performance Evaluation of Missing-Value Imputation Clustering Based on a Multivariate Gaussian Mixture Model

Cited by 10 publications

References 29 publications

Dynamic modelling of lettuce transpiration for water status monitoring

Dynamic modelling of lettuce transpiration for water status monitoring

Clustering with missing and left‐censored data: A simulation study comparing multiple‐imputation‐based procedures

A Bayesian Failure Prediction Network Based on Text Sequence Mining and Clustering

Contact Info

Product

Resources

About