A Novel End-To-End Feature Selection and Diagnosis Method for Rotating Machinery.

Feature selection is to obtain effective features from data, also known as feature engineering. Traditional feature selection and predictive model learning are separated, and there is a problem of inconsistency of criteria. This paper presents an end-to-end feature selection and diagnosis method that organically unifies feature expression learning and machine prediction learning into one model. The algorithm first combines the prediction model to calculate the mean impact value (MIVs) of the feature and realizes primary feature selection for the prediction model by selecting the feature with a larger MIV. In order to take into account the performance of the feature itself, the within-class and between-class discriminant analysis (WBDA) method is proposed, and combined with the feature diversity strategy, the feature-oriented secondary selection is realized. Eventually, feature vectors obtained by two selections are classified using a multi-class support vector machine (SVM). Compared with the modified network variable selection algorithm (MIVs), the principal component analysis dimensionality reduction algorithm (PCA), variable selection based on compensative distance evaluation technology (CDET), and other algorithms, the proposed method MIVs-WBDA exhibits excellent classification accuracy owing to the fusion of feature selection and predictive model learning. According to the results of classification accuracy testing after dimensionality reduction on rotating machinery status, the MIVs-WBDA method has a 3% classification accuracy improvement under the low-dimensional feature set. The typical running time of this classification learning algorithm is less than 10 s, while using deep learning, its running time will be more than a few hours.

Long Noncoding RNA Plasmacytoma Variant Translocation 1 Regulates Cisplatin Resistance via miR-3619-5p/TBL1XR1 Axis in Gastric Cancer.

Background: Chemoresistance greatly hinders the treatment of gastric cancer (GC). Long noncoding RNA (lncRNA) plasmacytoma variant translocation 1 (PVT1) has been corroborated to be involved in chemoresistance in diverse cancers, including GC. The authors' aim was to investigate the underlying molecular mechanism of PVT1 in cisplatin (DPP) resistance in GC. Methods: Quantitative real-time polymerase chain reaction was conducted to detect the expression levels of PVT1, microRNA (miR)-3619-5p, and transducin beta like 1 x-linked receptor 1 (TBL1XR1) in DDP-resistant GC tissues and cells. 3-(4, 5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay and flow cytometry assay were used to check cell viability, half inhibition concentration (IC50), and apoptosis, respectively. The abilities of cell migration and invasion were evaluated by transwell assay. The protein levels of drug resistance-related proteins permeability glycoprotein (P-gp), glutathione s-transferase pi (GST-π), multidrug resistance-associated protein, and TBL1XR1 in samples were measured by Western blot. A xenograft tumor model was established to investigate the biological function of PVT1 in vivo. The starBase site was utilized to predict binding sites between miR-3619-5p and PVT1 or TBL1XR1, and the dual-luciferase reporter assay was performed to verify the interaction. Results: The levels of PVT1 and TBL1XR1 were significantly upregulated in DPP-resistant GC tissues and cells, while miR-3619-5p was notably declined. Knockdown of PVT1 enhanced DPP sensitivity of DPP-resistant GC cells. Also, knockdown of PVT1 enhanced the sensitivity of DPP-resistant GC cells to DPP and inhibited tumor growth in vivo. Meanwhile, PVT1 silencing decreased the expression of drug-resistant proteins. Moreover, PVT1 interacted with miR-3619-5p, and TBL1XR1 was a target of miR-3619-5p. Further studies indicated that downregulation of miR-3619-5p transposed PVT1 silencing- or TBL1XR1 silencing-mediated effects on viability, apoptosis, migration, and invasion of DPP-resistant GC cells. Conclusions: PVT1 silencing attenuated the DPP resistance in GC by downregulating TBL1XR1 via sponging miR-3619-5p.

Design of Compressed Sensing Algorithm for Coal Mine IoT Moving Measurement Data Based on a Multi-Hop Network and Total Variation.

As the application of a coal mine Internet of Things (IoT), mobile measurement devices, such as intelligent mine lamps, cause moving measurement data to be increased. How to transmit these large amounts of mobile measurement data effectively has become an urgent problem. This paper presents a compressed sensing algorithm for the large amount of coal mine IoT moving measurement data based on a multi-hop network and total variation. By taking gas data in mobile measurement data as an example, two network models for the transmission of gas data flow, namely single-hop and multi-hop transmission modes, are investigated in depth, and a gas data compressed sensing collection model is built based on a multi-hop network. To utilize the sparse characteristics of gas data, the concept of total variation is introduced and a high-efficiency gas data compression and reconstruction method based on Total Variation Sparsity based on Multi-Hop (TVS-MH) is proposed. According to the simulation results, by using the proposed method, the moving measurement data flow from an underground distributed mobile network can be acquired and transmitted efficiently.