Skip to main content

Table 5 Methods combining error detection and correction (RQ2 and RQ3), along with its addressed errors, respective papers, and the total number of papers that proposed that method

From: Sensor data quality: a systematic review

MethodErrors addressedPapersTotal
Principal component analysisOutliers, bias, drift, constant values, noise, stuck-at-zero[46, 55]2
Artificial neural networkOutliers, bias[41, 43]2
Bayesian networkOutliers, missing data[7, 38]2
Grey prediction modelOutliers, bias, constant values, stuck-at-zero[30]1
Dempster–Shafer theoryUncertainty[80]1
Calibration-based methodBias, drift, noise, stuck-at-zero[73]1
Hybrid methods
 Principal component analysis-based methodsOutliers, bias, drift, noise, constant values, stuck-at-zero[60, 72, 74]3
 Kalman filter-based methodsOutliers, bias, drift, missing data[31, 51]2
 Dempster–Shafer theory & OntologyUncertainty (inaccurate data), missing data (incomplete data)[68]1