Linear Regression Prediction Model Research Articles

Enhanced Model-Based Predictive Control System Based on Fuzzy Logic for Maintaining Thermal Comfort in IoT Smart Space

Researchers have reached a consensus on the thermal discomfort known as the major cause of sick building syndrome, which hurts people’s health and working efficiency greatly. As a result, the thermal environment satisfaction is important and thus many studies have been dedicated to thermal comfort over the past few decades. Predicted Mean Vote (PMV) is one of the globally used standards to express users’ comfort satisfaction with the given thermal moderate environments. It has been widely used in most of the Heating, Ventilation and Air Conditioning (HVAC) systems to maintain this standard of thermal comfort for occupants of buildings. However, the PMV model is developed on indoor experimental data without taking into account conditions of outdoor space, which greatly affects the performance of the existing HVAC systems and varies with the seasons. In this paper, an enhanced Model-based Predictive Control practical system for maintaining the indoor thermal comfort is demonstrated, including a multiple linear regression predictive model and an innovative fuzzy controller considering both the PMV index and the outdoor environment conditions. To verify the usability of the designed system, an Internet of Things (IoT) smart space prototype was chosen and experimentally tested in a building in Jeju, Korea. Moreover, thermal comfort regulation performances using the proposed approach have been compared with the existing one. The results of our work indicate that the proposed solution is capable of optimizing the thermal comfort condition according to seasonality and outperforms the conventional approaches in different performance indexes.

Planar microwave sensor for detection and discrimination of aqueous organic and inorganic solutions

A planar microwave sensor based on a coplanar waveguide (CPW) loaded with a circular split ring resonator (SRR) was developed for the detection and estimation of salts, sugars and organic acids in aqueous solution. The magnitudes of the spectra at resonance frequencies of 2.3–2.6 GHz were shown to provide a rapid and reliable way to measure concentrations of materials such as sucrose, sorbitol, glucose, fructose, CaCl2, NaCl, KCl, MgCl2, Na2CO3 and citric acid. We generated linear and nonlinear regression prediction models built from correlations between the magnitudes of the transmission coefficient (S21) and concentration of materials using coefficient of determination (R2). Principle component analysis (PCA) method was used for discrimination of liquid solution samples. The studies clearly showed sample concentrations in the range from 4 to 20% (w/V) were correlated with the magnitude of S21 and could be detected by using the proposed sensor. Aqueous solution samples could be discriminated into two groups based on magnitude of the spectra of S21 of different solutions at various concentrations. The proposed device can be used to distinguish between organic and inorganic materials in aqueous solution.

Prediction of Maintenance Material Consumption for Aviation Equipment Using Linear Regression Model

It is very important to scientifically predict the consumption of aviation equipment maintenance material and to make scientific decisions on aviation equipment maintenance resources and make full use of existing resources to improve maintenance capability. In this paper, aiming at the main influencing factors of material consumption of aviation equipment maintenance, a linear regression prediction model of equipment maintenance material consumption was constructed by using actual statistical sample data. Based on the analysis of examples, the simple linear regression method is used to predict and test the material consumption of aviation equipment maintenance materials. The research results provide a scientific and effective method for forecasting the consumption of aviation equipment maintenance materials.

Prediction of sweet corn yield depending on cultivation technology parameters by using linear regression and artificial neural network methods

Artificial neural networks and linear regression are widely used in particularly all branches of science for modeling and prediction. Linear regression is an old data processing tool, and artificial neural networks are a comparatively new one. The goal of the study was to determine whether artificial neural networks are more accurate than linear regression in sweet corn yield prediction. In the study we used a dataset obtained from field experiments on the technological improvement of sweet corn cultivation. The field experiments were conducted during the period from 2014 to 2016 on dark-chestnut soil under drip irrigated conditions in the Steppe Zone of Ukraine. We studied the impact of the moldboard plowing depths, mineral fertilizer application rates and plant densities on the crop yield. A significant impact of all the studied factors on the sweet corn productivity was proved by using the analysis of variance. The highest yield of sweet corn ears without husks (10.93 t ha–1) was under the moldboard plowing at the depth of 20–22 cm, mineral fertilizers application rate of N120P120, plant density of 65,000 plants ha–1. Data processing by using the linear regression and artificial neural network methods showed that the latter is a great deal better than linear regression in sweet corn yield prediction. Higher accuracy of the artificial neural network prediction was proved by the higher value of the coefficient of determination (R2) – 0.978, in comparison to 0.897 for the linear regression prediction model. We conclude that artificial neural networks are a much better data processing tool, especially, in the life sciences and for prediction of the non-linear natural processes and phenomena. The main disadvantage of the neural network models is their “black box” nature. However, linear regression will not lose its popularity among scientists in the nearest future. Linear regression is a much simpler data analysis tool, it is easier to perform the prediction, but it still provides a sufficiently high level of accuracy.

Nondestructive, indirect assessment of the biomechanical properties of the rat intervertebral disc using contrast-enhanced μCT.

Mechanical characterization of the intervertebral disc involves labor-intensive and destructive experimental methodology. Contrast-enhanced micro-computed tomography is a nondestructive imaging modality for high-resolution visualization and glycosaminoglycan quantification of cartilaginous tissues. The purpose of this study was to determine whether anionic and cationic contrast-enhanced micro-computed tomography of the intervertebral disc can be used to indirectly assess disc mechanical properties in an ex vivo model of disc degeneration. L3/L4 motion segments were dissected from female Lewis rats. To deplete glycosaminoglycan, samples were treated with 0 U/ml (Control) or 5 U/ml papain. Contrast-enhanced micro-computed tomography was performed following incubation in 40% Hexabrix (anionic) or 30 mg I/ml CA4+ (cationic) for 24 h (n = 10/contrast agent/digestion group). Motion segments underwent cyclic mechanical testing to determine compressive and tensile modulus, stiffness, and hysteresis. Glycosaminoglycan content was determined using the dimethylmethylene blue assay. Correlations between glycosaminoglycan content, contrast-enhanced micro-computed tomography attenuation, and mechanical properties were assessed via the Pearson correlation. The predictive accuracy of attenuation on compressive properties was assessed via repeated random sub-sampling cross validation. Papain digestion produced significant decreases in glycosaminoglycan content and corresponding differences in attenuation and mechanical properties. Attenuation correlated significantly to glycosaminoglycan content and to all compressive mechanical properties using both Hexabrix and CA4+ . Predictive linear regression models demonstrated a predictive accuracy of attenuation on compressive modulus and stiffness of 79.8-86.0%. Contrast-enhanced micro-computed tomography was highly predictive of compressive mechanical properties in an ex vivo simulation of disc degeneration and may represent an effective modality for indirectly assessing disc compressive properties. © 2018 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 36:2030-2038, 2018.

Application of grey model (1, 1) -linear regression coupling model in the prediction of regional nursing human resource

Objective Taking the prediction of the number of nurses per thousand cases as an example, to discuss the application of grey model GM (1, 1) -linear regression Coupling Model in the prediction of regional nursing human resource, so as to provide methodology reference for health human resource prediction. Methods The grey model GM (1, 1) was built up by EXCEL formula to predict the influencing factors of the number of nurses per thousand people. The predictive value of the influencing factors were substituted into the regression equation for the annual prediction of regional nursing human resource. Results The fitting error between predicted value and actual value was small when using the grey model GM (1, 1) -linear regression coupling model in the prediction of regional nursing human resource. The accuracy level of this coupling model prediction is excellent, and the prediction result can be used as a reference for target year. Conclusions The coupling model not only makes up for the lack of 1inear factors in the grey system model, but also improves the defect that exponential growth cannot be expressed in the linear regression prediction model, so the construction of the coupling model is reasonable and feasible. Key words: Nursing administration research; Grey model (1, 1); Linear regression models; Nursing human resource; Forecasting

Quantitative Indices of the Oxidizability of Fatty Acid Compositions

The correlation between a wide range of fatty acid composition indices, resulted from different groups of purified edible oils and their binary blends, and two quantitative criteria of primary (induction period of hydroperoxide formation at 60 °C) and secondary (oxidative stability index, OSI, at 100 °C) oxidations was evaluated. On the basis of highly correlated linear regression models, a corrected form of calculated oxidizability index and the content of saturated fatty acids, SFA, were jointly found to be appropriate indices of primary oxidative stability of all vegetable oils whereas the ratio between palmitic and linoleic acids was mainly considered to show their secondary oxidative stability. As for the oils containing fish oil (FO), a combination of SFA/(C18:4 + C20:4 + C20:5 + C22:6), and monounsaturated fatty acids/(C18:2 + C18:3) ratios was used to represent the primary and secondary oxidative stabilities. Among the oils containing vegetable oils and even fish oil, canola oil, and its blends had the best fatty acid composition from a health point of view. Practical Application: Since edible oils have a vast diversity in fatty acid compositions of very different oxidative stability, prediction of oxidizability based on fatty acid compositions has become too complicated. Considering primary and secondary oxidation products over the course of lipid oxidation, the present study simplifies the prediction of the oxidative stability of fatty acid compositions by introducing four predictive linear regression models for vegetable and fish oils.Oxidative stability of five different edible oils and their binary blends are determined through the induction period (IP) of hydroperoxide formation and the oxidative stability index (OSI) at 60 and 100 °C, respectively. Considering a range of linear regression models between fatty acid composition indices and the oxidative stability quantities, IP and OSI, four linear regression models of highest correlation were obtained to simplify the prediction of oxidizability of diverse fatty acid compositions.

Linear regression models for prediction of annual heating and cooling demand in representative Australian residential dwellings

Abstract: This paper presents the development methodology of linear regression models that were developed for the prediction of annual thermal loads in representative residential buildings across three major climates in New South Wales, Australia, and the assessment of the impact of building envelope upgrades. A differential sensitivity analysis was undertaken for sixteen building envelope parameters, with six parameters being identified as significant. These six parameters were then explored using EnergyPlus simulation, and a number of linear regression models developed from the simulation outputs. Random values for design parameters were generated, and the results of EnergyPlus simulations using these parameters were used to verify the outputs of the regression models. The differences between regression-predicted and EnergyPlus-simulated annual thermal energy requirements were of order 10%-15%. The coefficient of determination (R2) was over 0.90, indicating a good agreement between simulation and the regression models, and suggesting that the annual heating and cooling energy requirements can be forecasted with an acceptable accuracy using the regression models. It is envisaged that the regression models developed can be used as a quick alternative to building simulation for residential buildings of the area and the climate covered by our study, and can serve to rapidly estimate the likely energy savings/penalty during the retrofitting design stage when different building schemes and design concepts are being considered.

Platelet reactivity in sepsis syndrome: results from the PRESS study.

Platelet activation mediates systemic inflammatory response during infection. However, data on platelet reactivity (PR) varies among different settings. We assessed PR along different stages of sepsis and tried to predict for determinants of its variance. In parallel, we evaluated it as an early bedside diagnostic biomarker. This was an observational prospective cohort study. Incoming patients were assorted to distinct groups of uncomplicated infection, sepsis, and severe sepsis/septic shock. A control group of healthy volunteers was used as comparison. PR was assessed using the bedside point-of-care VerifyNow assay, in P2Y12 reaction units (PRU) alongside with levels of major inflammatory markers and whole blood parameters. A total of 101 patients and 27 healthy volunteers were enrolled. PR significantly and reversibly increases during sepsis compared to uncomplicated infection and healthy controls (244±66.7 vs 187.33±60.98, p<0.001 and 192.17±47.51, p<0.001, respectively). In severe sepsis, PR did not significantly differ compared to other groups. Sepsis stage uniquely accounts for 15.5% of PR in a linear regression prediction model accounting for 30% of the variance of PR (F=8.836, p<0.001). PRU >253 had specificity of 91.2% and sensitivity of 40.8% in discriminating septic from non-septic patients. The addition of PRU to SOFA and qSOFA scores significantly increased their c-statistic (AUC SOFA+PRU, 0.867 vs SOFA, 0.824, p<0.003 and AUC qSOFA+PRU, 0.842 vs qSOFA, 0.739, p<0.001), making them comparable (AUC SOFA+PRU vs qSOFA+PRU, p=0.4). PR significantly and reversibly increases early in sepsis, but seems to exhaust while disease progresses. Bedside assessment of PR can provide robust discriminative accuracy in the early diagnosis of septic patients.

Prediction of Student Performance in Academic and Military Learning Environment: Use of Multiple Linear Regression Predictive Model and Hypothesis Testing

The variance in students’ academic performance in a civilian institute and in a military technological institute could be linked to the environment of the competition available to the students. The magnitude of talent, domain of skills and volume of efforts students put are identical in both type of institutes. The significant factor is the physical training, students undergo in a military college. It is important to couple the dominating factor which is academic perceivable effort under a different environment with each students learning capability. This paper determine whether there is a relationship between students’ performance and influencing factors like academic aptitude, military or physical training, and the time spent on training need analysis (TNA) modules. A sample of 242 first year- undergraduate students from four different engineering programs (Marine, System, Civil, and Aeronautical) at Military College was used to explore this relationship. The multiple regression model used for predicting the students’ performance is adequate for independent variables of aptitude test score, time spent in physical training, and time spent in TNA modules. The values of R2 indicate that at least one of the predictor variables contributes to information for the prediction of the students’ performance. The model makes it possible to predict moderately the possibility of attrition in engineering program. This study verifies that military academy has a very defined and directed core engineering course load and TNA course load which every student must take. Therefore, choice of specific discipline have less impact than at civilian institutions. The early detection of students at academic risk is a useful instrument that can help to design mentoring strategies right from the end of admission process.

Early prediction of extreme stratospheric polar vortex states based on causal precursors

AbstractVariability in the stratospheric polar vortex (SPV) can influence the tropospheric circulation and thereby winter weather. Early predictions of extreme SPV states are thus important to improve forecasts of winter weather including cold spells. However, dynamical models are usually restricted in lead time because they poorly capture low‐frequency processes. Empirical models often suffer from overfitting problems as the relevant physical processes and time lags are often not well understood. Here we introduce a novel empirical prediction method by uniting a response‐guided community detection scheme with a causal discovery algorithm. This way, we objectively identify causal precursors of the SPV at subseasonal lead times and find them to be in good agreement with known physical drivers. A linear regression prediction model based on the causal precursors can explain most SPV variability (r2 = 0.58), and our scheme correctly predicts 58% (46%) of extremely weak SPV states for lead times of 1–15 (16–30) days with false‐alarm rates of only approximately 5%. Our method can be applied to any variable relevant for (sub)seasonal weather forecasts and could thus help improving long‐lead predictions.

Bayesian and Phylogenic Approaches for Studying Relationships among Table Olive Cultivars.

To enhance table olive tree authentication, relationship, and productivity, we consider the analysis of 18 worldwide table olive cultivars (Olea europaea L.) based on morphological, biological, and physicochemical markers analyzed by bioinformatic and biostatistic tools. Accordingly, we assess the relationships between the studied varieties, on the one hand, and the potential productivity-quantitative parameter links on the other hand. The bioinformatic analysis based on the graphical representation of the matrix of Euclidean distances, the principal components analysis, unweighted pair group method with arithmetic mean, and principal coordinate analysis (PCoA) revealed three major clusters which were not correlated with the geographic origin. The statistical analysis based on Kendall's and Spearman correlation coefficients suggests two highly significant associations with both fruit color and pollinization and the productivity character. These results are confirmed by the multiple linear regression prediction models. In fact, based on the coefficient of determination (R 2) value, the best model demonstrated the power of the pollinization on the tree productivity (R 2=0.846). Moreover, the derived directed acyclic graph showed that only two direct influences are detected: effect of tolerance on fruit and stone symmetry on side and effect of tolerance on stone form and oil content on the other side. This work provides better understanding of the diversity available in worldwide table olive cultivars and supplies an important contribution for olive breeding and authenticity.

A Pareto-optimal moving average multigene genetic programming model for daily streamflow prediction

Genetic programming (GP) is able to systematically explore alternative model structures of different accuracy and complexity from observed input and output data. The effectiveness of GP in hydrological system identification has been recognized in recent studies. However, selecting a parsimonious (accurate and simple) model from such alternatives still remains a question. This paper proposes a Pareto-optimal moving average multigene genetic programming (MA-MGGP) approach to develop a parsimonious model for single-station streamflow prediction. The three main components of the approach that take us from observed data to a validated model are: (1) data pre-processing, (2) system identification and (3) system simplification. The data pre-processing ingredient uses a simple moving average filter to diminish the lagged prediction effect of stand-alone data-driven models. The multigene ingredient of the model tends to identify the underlying nonlinear system with expressions simpler than classical monolithic GP and, eventually simplification component exploits Pareto front plot to select a parsimonious model through an interactive complexity-efficiency trade-off. The approach was tested using the daily streamflow records from a station on Senoz Stream, Turkey. Comparing to the efficiency results of stand-alone GP, MGGP, and conventional multi linear regression prediction models as benchmarks, the proposed Pareto-optimal MA-MGGP model put forward a parsimonious solution, which has a noteworthy importance of being applied in practice. In addition, the approach allows the user to enter human insight into the problem to examine evolved models and pick the best performing programs out for further analysis.

Design, development and method validation of a novel multi-resonance microwave sensor for moisture measurement

Microwave sensor systems using resonance technology at a single resonance in the range of 2–3 GHz have been shown to be a rapid and reliable tool for moisture determination in solid materials including pharmaceutical granules. So far, their application is limited to lower moisture ranges or limitations above certain moisture contents had to be accepted. Aim of the present study was to develop a novel multi-resonance sensor system in order to expand the measurement range. Therefore, a novel sensor using additional resonances over a wide frequency band was designed and used to investigate inherent limitations of first generation sensor systems and material-related limits. Using granule samples with different moisture contents, an experimental protocol for calibration and validation of the method was established. Pursuant to this protocol, a multiple linear regression (MLR) prediction model built by correlating microwave moisture values to the moisture determined by Karl Fischer titration was chosen and rated using conventional criteria such as coefficient of determination (R2) and root mean square error of calibration (RMSEC). Using different operators, different analysis dates and different ambient conditions the method was fully validated following the guidance of ICH Q2(R1).The study clearly showed explanations for measurement uncertainties of first generation sensor systems which confirmed the approach to overcome these by using additional resonances. The established prediction model could be validated in the range of 7.6–19.6%, demonstrating its fit for its future purpose, the moisture content determination during wet granulations.

播期对秋闲田饲用高粱单株生产力与产量的回归分析

【目的】为了探索播期与秋闲田饲用高粱单株生产力和生物产量的线性回归预测模型。【方法】采用田间试验与线性回归分析的方法，对播期与饲用高粱单株生产力和草产量的关系进行模拟和比较。【结果】播期对秋闲田饲用高粱单株生产力和生物产量具有重要影响，其单株生产力和生物产量随播期的推迟而降低。构建了播期与单株生产力和生物产量关系的回归模型，单株鲜重与播期的回归模型为ŷ鲜 = 0.618 − 0.015x，单株干重与播期的回归模型为 ŷ干 = 0.184 − 0.005x，单产与播期的回归模型为 ŷ单 = 29126.461 − 711.448x。在7月23日至8月30日之间的播期每推迟1 d，饲用高粱的单株鲜重平均减少0.015 g，单株干重平均减少0.005 g，单产平均减少711.448 kg/hm2。【结论】本研究构建的3个回归模型，为秋闲田饲用高粱生产提供理论支撑。 [Objective] The aim was to explore the linear regression prediction model of different sowing times and plant productivity, biological yields of forage sorghum in autumn idle land. [Method] To different sowing times and plant productivity, biological yields of forage sorghum for regression model analysis were simulated and compared by employing field plot experiment and linear re-gression analysis. [Results] The results showed that the forage sorghum sowing times on plant productivity and biological yields have important influence, and its plant productivity and biolog-ical yields were decreased with the delay of sowing times. To construct the regression model be-tween sowing times and plant productivity and biological yields, regression model of the single fresh weight and sowing times is ŷfresh = 0.618 − 0.015x; model of the single dry weight and sowing is ŷhay = 0.184 − 0.005x x; model of the per unit area yield and sowing is ŷper = 29126.461 − 711.448x. Between July 23rd and August 30th, the sowing times were delayed 1 d, and forage sorghum plant fresh weight was reduced by 0.015 g on average, and the average plant dry weight was reduced by 0.005 g, and the average grass yield was reduced by 711.448 kg/hm2. [Conclusion] This study has constructed the three regression model, which will provide theoretical support for the production of Forage Sorghum in Autumn Idle Land.

基于偏相关与逐步回归方法的AQI影响因素分析及预测

空气质量指数(AQI)是定量描述空气质量状况的无量纲指数，可用于评价大气环境质量以及控制污染。本文以北京、长沙、海口AQI及PM2.5等6种大气污染指标的日均数据为研究对象。一方面，分别对三个城市的各个指标进行偏相关分析，结果表明影响北京空气质量的主要因素是PM10、PM2.5、NO2和O3；影响长沙空气质量的主要指标是PM10、PM2.5、NO2和CO；而影响海口空气质量的主要因素是PM10、PM2.5和SO2。北京、长沙对AQI影响最大的是PM2.5，而海口的PM10与AQI的相关关系最为显著。另一方面，应用逐步回归分析得到了影响上述三城市AQI的最显著的空气污染指标，结果与偏相关分析的结果相吻合。此外，通过偏相关分析和逐步回归分析，获得AQI的最优线性回归预测模型，模型的拟合效果显著。模型检验表明建立的预测方程能有效用于AQI的短期预测。最后针对分析结果提出一些解决对策，为空气质量改善提供建议。 Air quality index (AQI) is a dimensionless index which describes air condition quantitatively. It can be used to evaluate the quality of atmospheric environment as well as control contamination. In this paper, to investigate the influence of atmospheric pollution indexes on AQI, three cities of China, namely, Beijing, Changsha and Haikou, are chosen for your consideration. Six kinds of at-mospheric pollution index contains PM2.5 of daily date are used as observation. On the one hand, partial correlation analysis is employed for each index of the three cities respectively to deter-mine the main pollutants affecting AQI. It shows the main factors affecting the air quality in Beijing, Changsha and Haikou are PM10, PM2.5, NO2 and O3; PM10, PM2.5, NO2 and CO; PM10, PM2.5 and SO2, respectively. Moreover, the most significant correlation between the six indicators and AQI is PM2.5 for both Beijing and Changsha, but that is PM10 for Haikou. On the other hand, we also use stepwise regression method to access the primary factors affecting the AQI, which is consistent with the conclusion obtained from partial correlation analysis. In addition, we obtain the AQI optimal linear regression prediction model through partial correlation analysis and stepwise regression analysis. Besides, by the model tests, we find the proposed models for the three cities are workable, which can be used to forecast short-term AQI. Finally, some suggestions are provided for improving the air quality according to the results.

播期对秋闲田饲用高粱株高、叶茎比与干鲜比的回归分析

[目的]为了探索播期与秋闲田饲用高粱株高、叶茎比、干鲜比的线性回归预测模型。[方法]采用田间试验与线性回归分析的方法，对播期与饲用高粱株高、叶茎比、干鲜比的关系进行模拟和比较。[结果]播期对秋闲田饲用高粱株高、叶茎比、干鲜比具有重要影响，其株高、干鲜比随着播期的推迟而降低，其叶茎比随着播期的推迟而上升。构建了播期与株高、叶茎比、干鲜比关系的回归模型，株高与播期的回归模型为ŷ高 = 234.725 − 5.005x，叶茎比与播期的回归模型为ŷ叶 = 0.096 + 0.019x，干鲜比与播期的回归模型为ŷ干 = 0.305 − 0.002x。在7月23日至8月30日之间的播期每推迟1 d，饲用高粱的株高平均降低5.005 cm，叶茎比平均增加0.019，干鲜比平均减少0.002。[结论]本研究构建的3个回归模型，为秋闲田饲用高粱生产提供了理论支撑。 [Objective] The aim was to explore the linear regression prediction model of different sowing times and plant height, leaf stem ratio and DW/FW ratio of forage sorghum in autumn idle land. [Method] To different sowing times and plant height, leaf stem ratio and DW/FW ratio of forage sorghum for regression model analysis were simulated and compared by employing field plot experiment and linear regression analysis. [Results] The results showed that the forage sorghum sowing times on plant height, leaf stem ratio and DW/FW ratio have important influence, and its plant height and DW/FW ratio were decreased with the delay of sowing times, and its leaf stem ratio were increased with the delay of sowing times. To construct the regression model between sowing times and plant height, leaf stem ratio and DW/FW ratio, regression model of the plant height and sowing times is ŷh = 234.725 − 5.005x, model of the leaf stem ratio and sowing is ŷleaf = 0.096 + 0.019x, model of the DW/FW ratio and sowing is ŷdry = 0.305 − 0.002x. Between July 23rd and August 30th the sowing times were delayed 1 d, forage sorghum plant height was reduced by 5.005 cm on average, and the average leaf stem ratio was increased by 0.019, and the average DW/FW ratio was reduced by 0.002. [Conclusion] This study has constructed the three regression models, and will provide theoretical support for the production of Forage Sorghum in Autumn Idle Land.

A Predictive Linear Regression Model for Affective State Detection of Mobile Touch Screen Users

Emotion, being important human factor, should be considered to improve user experience of interactive systems. For that, we first need to recognize user's emotional state. In this work, the author proposes a model to predict the affective state of a touch screen user. The prediction is done based on the user's finger strokes. The author defined seven features on the basis of the strokes. The proposed predictor is a linear combination of these features, which the author obtained using a linear regression approach. The predictor assumes three affective states in which a user can be: positive, negative and neutral. The existing works on affective touch interaction are few and rely on many features. Some of the feature values require special sensors, which may not be present in many devices. The seven features we propose do not require any special sensor for computation. Hence, the predictor can be implemented on any device. The model is developed and validated with empirical data involving 57 participants performing 7 touch input tasks. The validation study demonstrates a high prediction accuracy of 90.47%.

Modified Stepwise Regression Approach to Streamlining Predictive Analytics for Construction Engineering Applications

AbstractA literature review has identified the absence of a robust framework that guides the development of streamlined and valid multiple linear regression (MLR) predictive models for construction...

Multivariate optimization of the hollow fibre liquid phase microextraction of muscimol in human urine samples.

A liquid phase microextraction based on hollow fibre followed by liquid chromatographic determination was developed for the extraction and quantitation of the hallucinogenic muscimol from urine samples. Method applicability on polar hallucinogens was also tested on two alkaloids, a psychedelic hallucinogen, tryptamine and a polar amino acid, tryptophan which exists in its charged state in the entire pH range. A multivariate design of experiments was used in which a half fractional factorial approach was applied to screen six factors (donor phase pH, acceptor phase HCl concentration, carrier composition, stirring rate, extraction time and salt content) for their extent of vitality in carrier mediated liquid microextractions. Four factors were deemed essential for the effective extraction of each analyte. The vital factors were further optimized for the extraction of single-spiked analyte solutions using a central composite design. When the simultaneous extraction of analytes was performed under universal factor conditions biased towards maximizing the enrichment of muscimol, a good composite desirability value of 0.687 was obtained. The method was finally applied on spiked urine samples with acceptable enrichments of 4.1, 19.7 and 24.1 obtained for muscimol, tryptophan and tryptamine respectively. Matrix-based calibration curves were used to address matrix effects. The r(2) values of the matrix-based linear regression prediction models ranged from 0.9933 to 0.9986. The linearity of the regression line of the matrix-based calibration curves for each analyte was directly linked to the analyte enrichment repeatability which ranged from an RSD value of 8.3-13.1%. Limits of detection for the developed method were 5.12, 3.10 and 0.21ngmL(-1) for muscimol, tryptophan and tryptamine respectively. The developed method has proven to offer a viable alternative for the quantitation of muscimol in human urine samples.