Research on application of ReliefF and improved RVM in water quality grade evaluation

ReliefF algorithm was used to analyze the weight of each water quality evaluation factor, and then based on the Relevance Vector Machine (RVM), Particle Swarm Optimization (PSO) was used to optimize the kernel width factor and hyperparameters of RVM to build a water quality evaluation model, and the experimental results of RVM, PSO-RVM, ReliefF-RVM and PSO-ReliefF-RVM were compared. The results show that ReliefF algorithm, combined with threshold value, selects 5 evaluation factors with significant weight from 8 evaluation factors, which reduces the amount of data used in the model, CSI index is used to calculate the separability of each evaluation factor combination. The results show that the overall separability of the combination is best when the evaluation factor with significant weight is reserved. When different water quality evaluation factors were included, the evaluation accuracy of PSO-ReliefF-RVM model reached 95.74%, 14.23% higher than that of RVM model, which verified the effectiveness of PSO algorithm and ReliefF algorithm, and had a higher guiding significance for the study of water quality grade evaluation. It has good practical application value.

Modeling and optimization of EDM machined AZ-91 Mg alloy using ANFIS-VIKOR method

In this paper, machining of AZ-91 magnesium alloy was performed in EDM using different tool electrode (Cu, CuW and graphite). To perform experiments, Taguchi L18 design of experiments was used to reduce experimental runs. EDM process parameters viz. polarity, current (Ip), pulse-on-time (Ton), pulse-off-time (Toff), and tool electrode material were considered in experimental design to measure the responses (MRR, Ra). Multi-input-single-output adaptive neuro fuzzy inference system (ANFIS) model was developed to predict responses, and predicted results were found in good agreement with the experimental results. Maximum MRR (0.089 g/min) was found at positive polarity, Ip-5 A, Ton-50 μs, Toff - 30 μs, and Cu tool, and minimum Ra (0.08 μm) was at parameters positive polarity, Ip-4 A, Ton-30 μs, Toff -20 μs, and Cu tool. Relative errors between experiential results and ANFIS predicted results were 1.17 % for MRR, and 2.20 % for Ra. Multi response optimization ANFIS-VIKOR method was successfully developed and gave compromise solution for MRR and Ra corresponding to minimum ANFIS-VIKOR index (Qi). A factor level analysis was performed to evaluate optimal factor combination for ANFIS-VIKOR index, and it shows that current (Ip) have a significant effect.

MINIMUM DETECTABLE DIFFERENCE OF CT ANGIOGRAPHY SCANS AT VARIOUS CARDIAC BEATS: EVALUATION VIA A CUSTOMIZED OBLIQUE V-SHAPED LINE GAUGE AND PMMA PHANTOM

The minimum detectable difference (MDD) at various beats/min (BPM) of CT angiography (CTA) was evaluated using an oblique V-shape line gauge and poly methyl methacrylate (PMMA) phantom in this study. The customized phantom with the size of [Formula: see text][Formula: see text]cm3 was made from a 1[Formula: see text]cm-thick PMMA. The reciprocating mechanism in the phantom was run by a step motor with an eccentric gear connected to a crank rod to provide a stable harmonic motion, simulating the cardiac beats. The MDD has a unique feature in defining the quality characteristic of CT-scanned images and provides more information than simple line pair/cm in the previous studies. The derived MDD was quantified according to various BPM, and the CTA factor combination was preset following either the conventional recommendation or the optimal one. In doing so, the performance was substantiated by the Taguchi-based signal-to-noise ratio and integrated by another index, namely, figure of merit (FOM). The MDD and corresponding [Formula: see text] (dB) changed from [Formula: see text][Formula: see text]mm to [Formula: see text][Formula: see text]mm and from 16.7[Formula: see text]dB to 14.2[Formula: see text]dB, respectively, for conventional settings; while those obtained for the optimal preset changed from [Formula: see text][Formula: see text]mm to [Formula: see text][Formula: see text]mm and from 12.2 dB to 16.4 dB, respectively of CTA at 0–90 BPM. The integrated FOM values for conventional or optimal cases were 1240 and 1337, respectively. The MDD proved to be a useful technique in justifying the CTA-scanned images. For compliance with previous studies, MDD results can be converted to the line pair/cm results, but it is more informative than the quantized number of line pairs.

Whole-Body EMS Superimposed Walking and Nordic Walking on a Treadmill—Determination of Exercise Intensity to Conventional Exercise

Electrical muscle stimulation (EMS) is an increasingly popular training method and has become the focus of research in recent years. New EMS devices offer a wide range of mobile applications for whole-body EMS (WB-EMS) training, e.g., the intensification of dynamic low-intensity endurance exercises through WB-EMS. The present study aimed to determine the differences in exercise intensity between WB-EMS-superimposed and conventional walking (EMS-CW), and CON and WB-EMS-superimposed Nordic walking (WB-EMS-NW) during a treadmill test. Eleven participants (52.0 ± years; 85.9 ± 7.4 kg, 182 ± 6 cm, BMI 25.9 ± 2.2 kg/m2) performed a 10 min treadmill test at a given velocity (6.5 km/h) in four different test situations, walking (W) and Nordic walking (NW) in both conventional and WB-EMS superimposed. Oxygen uptake in absolute (VO2) and relative to body weight (rel. VO2), lactate, and the rate of perceived exertion (RPE) were measured before and after the test. WB-EMS intensity was adjusted individually according to the feedback of the participant. The descriptive statistics were given in mean ± SD. For the statistical analyses, one-factorial ANOVA for repeated measures and two-factorial ANOVA [factors include EMS, W/NW, and factor combination (EMS*W/NW)] were performed (α = 0.05). Significant effects were found for EMS and W/NW factors for the outcome variables VO2 (EMS: p = 0.006, r = 0.736; W/NW: p < 0.001, r = 0.870), relative VO2 (EMS: p < 0.001, r = 0.850; W/NW: p < 0.001, r = 0.937), and lactate (EMS: p = 0.003, r = 0.771; w/NW: p = 0.003, r = 0.764) and both the factors produced higher results. However, the difference in VO2 and relative VO2 is within the range of biological variability of ± 12%. The factor combination EMS*W/NW is statistically non-significant for all three variables. WB-EMS resulted in the higher RPE values (p = 0.035, r = 0.613), RPE differences for W/NW and EMS*W/NW were not significant. The current study results indicate that WB-EMS influences the parameters of exercise intensity. The impact on exercise intensity and the clinical relevance of WB-EMS-superimposed walking (WB-EMS-W) exercise is questionable because of the marginal differences in the outcome variables.

TAGUCHI-BASED OPTIMIZATION OF HEAD AND NECK CT ANGIOGRAPHY: IN-VIVO ENHANCED TRIGGERED TIMING FOR 600 PATIENTS

The Taguchi-based optimization of head and neck CT angiography (CTA) using in-vivo enhanced triggered timing for 600 patients was accomplished in this study. A total of 600 patients were categorized into three (test, verification, and implementation groups with 360, 120, and 120 patients, respectively). The patients in the test group were randomly assigned into 18 sub-groups according to Taguchi L[Formula: see text](2[Formula: see text]) orthogonal array for optimizing factor combinations of head and neck CTA. The signal-to-noise ratio was defined as a high LRA/US ratio (both left and right arterial pressure divided by upper sinuses’ one) and low stdev. The seven factors of CTA were: (A) left- or right-hand injection; (B) tube current, mA; (C) kilovoltage peak, kVp; (D) contrast media concentration; (E) FOV; (F) flow rate of contrast media; and (G) rotation time for one CTA loop. With the Taguchi unique organization, 18 sub-groups (each containing 20 patients’ real LRA/US ratios) could cover 1458 cases and provide the appropriate reliability. The optimal factor combination was identified as follows: (A) left-hand injection, (B) tube current of 280[Formula: see text]mA, (C) 100[Formula: see text]kVp, (D) contrast media concentration of 60%, (E) FOV of 250[Formula: see text]mm, (F) contrast media flow rate of 4.0[Formula: see text]ml/s, and (G) rotation time for one CTA loop of 0.9[Formula: see text]s. The optimal suggestion was validated in the verification group with 120 patients and applied to the implementation group with 120 patients. The successful factor combination is instrumental in defining a standard protocol for medical staff in deriving the maximal LRA/US value ([Formula: see text]) of head and neck CTA.

Research on influencing factors and multiple driving paths of intelligent transformation in China’s manufacturing industry

As the internal requirement of the development of China’s traditional manufacturing industry, intelligent transformation has gradually become the main goal of China’s regional manufacturing transformation. From the perspective of manufacturing value chain, this paper proposes a conceptual model of intelligent transformation of manufacturing industry. Taking 22 sample cities as cases, the paper uses fuzzy set qualitative comparative analysis method to identify the necessary factors of intelligent transformation of manufacturing industry and the configuration analysis of each factor combination, and then obtains three realization paths to promote intelligent transformation. The results show that intelligent technology innovation and intelligent industry investment are the necessary conditions for the intelligent transformation of manufacturing industry, and there are three configuration paths: collaborative R&D and processing, external factor driven and value chain climbing, which promote the intelligent transformation of manufacturing industry. This paper enriches the research on intelligent transformation of manufacturing industry from the perspective of value chain, and uses the qualitative comparative analysis method of fuzzy set to provide countermeasures and suggestions for the direction and path of intelligent transformation of manufacturing industry.

Simple Baseline Machine Learning Text Classifiers for Small Datasets

Text classification is important to better understand online media. A major problem for creating accurate text classifiers using machine learning is small training sets due to the cost of annotating them. On this basis, we investigated how SVM and NBSVM text classifiers should be designed to achieve high accuracy and how the training sets should be sized to efficiently use annotation labor. We used a four-way repeated-measures full-factorial design of 32 design factor combinations. For each design factor combination 22 training set sizes were examined. These training sets were subsets of seven public text datasets. We study the statistical variance of accuracy estimates by randomly drawing new training sets, resulting in accuracy estimates for 98,560 different experimental runs. Our major contribution is a set of empirically evaluated guidelines for creating online media text classifiers using small training sets. We recommend uni- and bi-gram features as text representation, btc term weighting and a linear-kernel NBSVM. Our results suggest that high classification accuracy can be achieved using a manually annotated dataset of only 300 examples.

The Transfer Mechanism of Pollution Industry in China Under Multi-factor Combination Model--Based on Industry, Location and Environment Perspectives

In the process of industry transfer, how to ensure the sustainable development of ecological environment is a very important problem for every local government. Therefore, this paper introduces environmental factors into the transfer mechanism of pollution industry, establishes a Triangle Model, and empirically studies the transfer mechanism of China’s pollution industry under various combination scenarios based on the panel data of 30 provinces in 2000–2015. The results show that: (1) The pollution industry transfer is the result of the change of regional competitive advantages; the industrial advantage is the primary factor of pollution industry transfer in China and is positive role in attracting industrial transfer. (2) With the continuous promotion of regional coordination strategy and ecological civilization construction, the attraction of regional factors to industry transfer has been reduced, and the inhibition of environmental factors to pollution industry is increasing. In addition, it is worth noting that China’s pollution industry had not gone to “innovation highland” but “environmental depression” during the research period, that is, the phenomenon of PHH more likely happened during the process of industry transfer among Chinese provinces. The paper suggests that all provinces should adhere to the concept of green development and promote industrial transformation and upgrading through industry transfer.