A Suppression Method of Concentration Background Noise by Transductive Transfer Learning for a Metal Oxide Semiconductor-Based Electronic Nose

Abstract

Signal drift caused by sensors or environmental changes, which can be regarded as data distribution changes over time, is related to transductive transfer learning, and the data in the target domain is not labeled. We propose a method that learns a subspace with maximum independence of the concentration features (MICF) according to the Hilbert-Schmidt Independence Criterion (HSIC), which reduces the inter-concentration discrepancy of distributions. Then, we use Iterative Fisher Linear Discriminant (IFLD) to extract the signal features by reducing the divergence within classes and increasing the divergence among classes, which helps to prevent inconsistent ratios of different types of samples among the domains. The effectiveness of MICF and IFLD was verified by three sets of experiments using sensors in real world conditions, along with experiments conducted in the authors’ laboratory. The proposed method achieved an accuracy of 76.17%, which was better than any of the existing methods that publish their data on a publicly available dataset (the Gas Sensor Drift Dataset). It was found that the MICF-IFLD was simple and effective, reduced interferences, and deftly managed tasks of transfer classification.

Review–Modern Data Analysis in Gas Sensors

“…To minimize the maximum mean discrepancy among the features, many researchers employed joint distribution adaptationbased transfer learning where Principal Component Analysis (PCA) was used to measure the maximum variance of the dataset. Liu et al 75 extracted features using Fisher linear discriminant approach for drift compensation which decreases the inter-concentration discrepancy of the feature distribution, resulting in concentration independent features. Balanced distribution adaptation algorithm showed better performance over joint distribution adaptation according to Jiang et al 76 where the marginal and conditional distributions are considered.…”

“…102 Maximum independence of concentration features was calculated to reduce the inter-concentration discrepancy of distribution thereby improving the noise influence on the dataset. 75 Chao et al employed systematic and measurement noise inside a Kalman filter equation, which was successful in reducing overall noise, hence increasing the signal to noise ratio. 103 A CNN based denoising autoencoder technique showed significant efficiency in noise removal.…”

“…Liu et al 75 proposed a concentration independent drift compensation approach utilizing transfer learning technique with an average accuracy of 76.17%. A short-term memory and SVM incorporated ensemble technique was mentioned in an earlier article.…”

Machine Learning Enhances the Performance of Bioreceptor-Free Biosensors

“…Transfer learning has been extensively employed to counter Enose sensor drift and reduce the need for complete retraining [ 35 , 36 , 37 , 38 ]. It has also been used to reduce the deleterious effect of background interference [ 39 , 114 ]. Although several of the above papers [ 35 , 36 , 38 , 39 ] demonstrate the efficacy of their approach on a shared sensor drift dataset shown in Figure 7 [ 115 ], ranking of the methods is difficult due to inconsistent benchmarking metrics.…”

A Machine Learning Method for the Fine-Grained Classification of Green Tea with Geographical Indication Using a MOS-Based Electronic Nose

“…Therefore, the evaluation of the discriminative characteristics that distinguish different teas is not straightforward. Many studies focused on improving the performance of the E-nose system by selecting sensitive material [ 26 ], optimizing the sensor array [ 27 ], or enhancing the feature extraction methods and algorithms [ 28 , 29 ]. The feature extraction method is a crucial aspect of the optimization methods, and it is important to extract useful information from the sensor signals [ 30 ] To the best of our knowledge, most researchers employing E-noses used steady-state or transient-state features (maximum values [ 31 , 32 ], stable values [ 33 , 34 ], integrals [ 35 , 36 ], and derivatives [ 36 , 37 ]).…”