Random Forest Models and Ensemble Learning-Based dynamic Sowing Methodology

Optimizing the use of marginal saline soils to improve food security requires solutions that involve exploiting local potential resources, including the use of biological agents. This research aims to determine the ability of Trichoderma sp. Tc-31 and Aspergillus sp. As-27 to overcome salinity stress in rice and soybean seedlings. Both isolates were tested for their ability to grow in vitro in saline soil. Next, the ability to overcome saline soil stress during germination and initial growth of rice and soybean seedlings was tested. The experiment consisted of three types of treatment, namely the use of As-27, Tc-31, and the use of husks treated with complete basic fertilizer. The average inhibitory power of saline soil against As-27 and Tc-31 was (-)62.50±16.68% and (-) 52±3.54% respectively at 48 days of incubation. The application of biological agents had an effect on germination in rice 94.10-94.46% and soybeans 74.49-77.04%. Biological agents also influence the height growth of rice and soybean seedlings. These two fungal isolates have the potential to be used as biological agents to help rice and soybean seeds overcome salinity stress.

Predictive analytics in emergency care has mostly been limited to the use of clinical decision rules (CDRs) in the form of simple heuristics and scoring systems. In the development of CDRs, limitations in analytic methods and concerns with usability have generally constrained models to a preselected small set of variables judged to be clinically relevant and to rules that are easily calculated. Furthermore, CDRs frequently suffer from questions of generalizability, take years to develop, and lack the ability to be updated as new information becomes available. Newer analytic and machine learning techniques capable of harnessing the large number of variables that are already available through electronic health records (EHRs) may better predict patient outcomes and facilitate automation and deployment within clinical decision support systems. In this proof-of-concept study, a local, big data-driven, machine learning approach is compared to existing CDRs and traditional analytic methods using the prediction of sepsis in-hospital mortality as the use case. This was a retrospective study of adult ED visits admitted to the hospital meeting criteria for sepsis from October 2013 to October 2014. Sepsis was defined as meeting criteria for systemic inflammatory response syndrome with an infectious admitting diagnosis in the ED. ED visits were randomly partitioned into an 80%/20% split for training and validation. A random forest model (machine learning approach) was constructed using over 500 clinical variables from data available within the EHRs of four hospitals to predict in-hospital mortality. The machine learning prediction model was then compared to a classification and regression tree (CART) model, logistic regression model, and previously developed prediction tools on the validation data set using area under the receiver operating characteristic curve (AUC) and chi-square statistics. There were 5,278 visits among 4,676 unique patients who met criteria for sepsis. Of the 4,222 patients in the training group, 210 (5.0%) died during hospitalization, and of the 1,056 patients in the validation group, 50 (4.7%) died during hospitalization. The AUCs with 95% confidence intervals (CIs) for the different models were as follows: random forest model, 0.86 (95% CI = 0.82 to 0.90); CART model, 0.69 (95% CI = 0.62 to 0.77); logistic regression model, 0.76 (95% CI = 0.69 to 0.82); CURB-65, 0.73 (95% CI = 0.67 to 0.80); MEDS, 0.71 (95% CI = 0.63 to 0.77); and mREMS, 0.72 (95% CI = 0.65 to 0.79). The random forest model AUC was statistically different from all other models (p ≤ 0.003 for all comparisons). In this proof-of-concept study, a local big data-driven, machine learning approach outperformed existing CDRs as well as traditional analytic techniques for predicting in-hospital mortality of ED patients with sepsis. Future research should prospectively evaluate the effectiveness of this approach and whether it translates into improved clinical outcomes for high-risk sepsis patients. The methods developed serve as an example of a new model for predictive analytics in emergency care that can be automated, applied to other clinical outcomes of interest, and deployed in EHRs to enable locally relevant clinical predictions.

Effects of hydrocarbons, such as methane, ethane, ethylene, propane and propylene, on the root growth of soybean (Glycine max Merr. cv. Enrei) and rice (Oryza sativa L. cv. Nihonbare) seedlings were examined. The results obtained are summarized as follows: 1. When roots of soybean were exposed to 1, 000 ppm of hydrocarbons, ethylene was most effective to inhibit the root growth with promortion of root-hair formation and increase of stem-diameter. The activity was decreased in the order of propyelene, propane, ethan and methane. Inhibitory activity of propylene on root growth was one hundredth of ethylene, and 50% inhibition of soybean root growth was obtained with 1 ppm ethylene. 2. In rice seedling, seminal root growth was stimulated with both of 1, 000 ppm ethylene and propylene, whereas the growth of first crown roots was inhibited greatly as well as the soybean roots. 3. Effect of ethylene on the growth of seminal root of rice seedling was examined in the range of concentrations from 0.1 to 10, 000 ppm. The most effectivc promotion was obtained at 0.1 to 1 ppm. On the other hand, the growth of first crown roots was promoted with 0.1 ppm ethylene, whereas concentrations higher than 10 ppm inhibited the growth greatly. 4. Effect of ethylene on rooting of rice seedlings was examined and promotion of new roots elongation was observed at 0.1 ppm of ethylene, whereas increase in number of new roots was observed at 10 ppm of ethylene. These results strongly suggest the view that ethylene in the rhizosphere may have direct effect on the growth and development of roots.

Understanding the beneficial plant-microbe interactions is becoming extremely critical for deploying microbes imparting plant fitness and achieving sustainability in agriculture. Diazotrophic bacteria have the unique ability to survive without external sources of nitrogen and simultaneously promote host plant growth, but the mechanisms of endophytic interaction in cereals and legumes have not been studied extensively. We have studied the early interaction of two diazotrophic bacteria, Gluconacetobacter diazotrophicus (GAB) and Bradyrhizobium japonicum (BRH), in 15-day-old seedlings of rice and soybean up to 120 h after inoculation (hai) under low-nitrogen medium. Root colonization of GAB in rice was higher than that of BRH, and BRH colonization was higher in soybean roots as observed from the scanning electron microscopy at 120 hai. Peroxidase enzyme was significantly higher at 24 hai but thereafter was reduced sharply in soybean and gradually in rice. The roots of rice and soybean inoculated with GAB and BRH harvested from five time points were pooled, and transcriptome analysis was executed along with control. Two pathways, "Plant pathogen interaction" and "MAPK signaling," were specific to Rice-Gluconacetobacter (RG), whereas the pathways related to nitrogen metabolism and plant hormone signaling were specific to Rice-Bradyrhizobium (RB) in rice. Comparative transcriptome analysis of the root tissues revealed that several plant-diazotroph-specific differentially expressed genes (DEGs) and metabolic pathways of plant-diazotroph-specific transcripts, viz., chitinase, brassinosteroid, auxin, Myeloblastosis (MYB), nodulin, and nitrate transporter (NRT), were common in all plant-diazotroph combinations; three transcripts, viz., nitrate transport accessory protein (NAR), thaumatin, and thionin, were exclusive in rice and another three transcripts, viz., NAC (NAM: no apical meristem, ATAF: Arabidopsis thaliana activating factor, and CUC: cup-shaped cotyledon), ABA (abscisic acid), and ammonium transporter, were exclusive in soybean. Differential expression of these transcripts and reduction in pathogenesis-related (PR) protein expression show the early interaction. Based on the interaction, it can be inferred that the compatibility of rice and soybean is more with GAB and BRH, respectively. We propose that rice is unable to identify the diazotroph as a beneficial microorganism or a pathogen from an early response. So, it expressed the hypersensitivity-related transcripts along with PR proteins. The molecular mechanism of diazotrophic associations of GAB and BRH with rice vis-à-vis soybean will shed light on the basic understanding of host responses to beneficial microorganisms.

Novel design and fabrication of manual rice transplantation machine based on linkage dimension from Thomas to suit the varying terrains of Nepal and to initiate mechanization from the small land holders farmers are proposed in this study. The two row manual rice planter from Chandragiri Machiniery Industry, Satungal was taken as reference to modify transplanting mechanisms and driving systems to ease the rice plantation process. The optimum adjustable spacing of the rice seedlings is 20 to 25 cm hill to hill spacing and 10 cm row to row spacing. The depth of penetration of rice seedlings is also adjustable in this design. The power transmission for the driving mechanism is achieved through the traction power of the rice planting wheel. The capacity of rice transplanter is 270 sq. meters per hour which is equivalent to 5 number of manual hand rice transplanter workers. By using the proposed rice transplanter machine the total cost of hand rice transplantation can be reduced by approximately 75%.

As a large power consumer, the iron and steel industry urgently needs to improve productivity, reduce energy consumption, and save costs by revolutionizing energy management. In this study, we design a power demand management system for the iron and steel industry, and around the load management module, we propose a data-driven strategy based on daily load situational awareness and introduce the random forest model into daily load forecasting in the iron and steel industry. At the same time, the projection principle is applied to improve the traditional gray correlation similar day selection algorithm, and a combination method of daily load forecasting based on the gray projection improved random forest algorithm is proposed. The electric load data of the iron and steel industry in a specific region is utilized as an experimental sample to investigate the model’s forecasting performance and the impact of the data-driven strategy. The model for daily load forecasting in this paper has an average relative error of 1.18%, which is better than other models. The application of the data-driven strategy brought about 6.99% and 6.69% reductions in load demand and basic electricity cost. The data-driven strategy for the steel industry based on the Random Forest model can predict electric loads more accurately and reduce energy costs, as shown by the results.

We conducted a survey on rodent crop damage among farmers in the hinterland of Kisangani (Democratic Republic of Congo). We studied the amount of crop damage, the rodent groups causing crop damage, the growth stages affected and the control techniques used. We conducted this survey in three municipalities using a standard questionnaire form translated into local languages, between November 2005 and June 2006 and during July 2007. We used the Quotas method and interviewed 70 households per municipality. Farmers indicated rodent groups implicated in crop damage on color photographs. Two types of survey techniques were used: individual and focus-group surveys. The sugar cane rat, Thryonomys sp. and Lemniscomys striatus caused most damage to crops, but inside granaries, Rattus rattus was the primary pest species eating stored food supplies and causing damage to stored goods. Cassava and maize were the most cultivated crops, followed by rice. Maize and rice are the most vulnerable during the heading stage. The most usable part of cassava (tubers) was attacked less often. Farmers mostly applied traditional trapping techniques, using local snap traps to decrease rodent abundance and sometimes used a labor-intensive trap-barrier system to trap Thryonomys . Inside buildings (granaries), farmers used chemical control methods.

With the circulation of wild poliovirus (WPV) types 1 and 3 continuing more than a decade after the original goal of eradicating all three types of WPVs by 2000, policymakers consider many immunization options as they strive to stop transmission in the remaining endemic and outbreak areas and prevent reintroductions of live polioviruses into nonendemic areas. While polio vaccination choices may appear simple, our analysis of current options shows remarkable complexity. We offer important context for current and future polio vaccine decisions and policy analyses by developing decision trees that clearly identify potential options currently used by countries as they evaluate national polio vaccine choices. Based on a comprehensive review of the literature we (1) identify the current vaccination options that national health leaders consider for polio vaccination, (2) characterize current practices and factors that appear to influence national and international choices, and (3) assess the evidence of vaccine effectiveness considering sources of variability between countries and uncertainties associated with limitations of the data. With low numbers of cases occurring globally, the management of polio risks might seem like a relatively low priority, but stopping live poliovirus circulation requires making proactive and intentional choices to manage population immunity in the remaining endemic areas and to prevent reestablishment in nonendemic areas. Our analysis shows remarkable variability in the current national polio vaccine product choices and schedules, with combination vaccine options containing inactivated poliovirus vaccine and different formulations of oral poliovirus vaccine making choices increasingly difficult for national health leaders.

The present study discusses the design of a manual rice-transplanting machine. A four bar mechanism is used to achieve the planting action of the machine. The machine is designed to plant multiple rows of seedlings at the same time. It is expected to be much faster and accurate, than the conventional manual transplanting of paddy. Moreover, the machine is operated by pulling it on the paddy field by a person and hence, it does not consume any conventional fossil fuel. A finger like fork is attached as a coupler extension to the four bar mechanism that pushes the rice seedlings from a tray to the planting area. The dimensions of each bar of the four bar linkage are determined from the coupler curve, using the three point coupler curve method. Input power to the mechanism is provided by the rotation of the cogwheels. There is a requirement of definite gear ratio between the driving cogwheel and the crank of the paddy-transplanting machine to sustain an optimum distance between two seedlings. The dimensions of the different links of the four bar mechanism are 406.80 mm (fixed bar), 120 mm (crank), 475.27 mm (coupler) and 278.63 mm (rocker). The present study contains detailed design 148analysis and estimation of different components used in the manual rice transplanting machine. Computer-aided drafting / design of the model for its fabrication is also included.

Rapid transit systems are critical components of urban public transportation networks in their impact, not only on personal mobility but also on the energy and environmental costs associated with network operations. To facilitate effective planning for current and future needs, a framework is required that provides important consumption metrics and also explains the various contributors to energy consumption, along with their interactions. To address this gap, we estimated models that utilized operational and ridership data for the Massachusetts Bay Transportation Authority’s rapid transit system, as well as ambient temperature, to accurately predict system-wide electricity consumption. The models were trained with data from 2019 and tested with data from 2020. The estimated multiple linear regression (MLR) and random forest (RF) models explained 93% and 95% of the variance in the data set, respectively. The MLR model provided predictions with a root mean squared error (RMSE) of 2.7 MWh and mean absolute percentage error (MAPE) of 4.68%, while the RF model resulted in an RMSE of 2.94 MWh and MAPE of 5.01%. We also investigated the impacts of COVID-19 on the transit system by exploring the effects on ridership, energy consumption, cost, and train movement metrics before and during the pandemic. We find that the models are robust and perform well, even with the significant disruptions associated with the COVID-19 pandemic.

Video Surveillance is an active area of research and provides promising security measures for consumer applications. To ease consumer surveillance investigations, Video Synopsis (VS) serves as a powerful tool to assess hours of video in a shorter retro of time by projecting multiple objects concurrently. The optimization module in VS framework is considered to be a key module, yet to date, only traditional optimization techniques have been addressed for energy minimization. Amongst these, simulated annealing (SA) has been broadly employed to produce global optimal solution without getting trapped in local minima. However, the convergence time of SA is quite high as the next state is chosen randomly to achieve real-time performance. This article presents an improved energy minimization scheme using hybridization of SA and JAYA algorithm to achieve global optimal solution with faster convergence rate. The weights associated with the energy function are computed using analytic hierarchy process (AHP) instead of heuristic selection. From experimental evaluations and analysis, it is seen that the proposed scheme exhibits superior performance to minimize the overall energy cost with lesser computational time. The proposed scheme has a potential to quickly review consumer surveillance video data in a smart and efficient way.

This study offers an analysis of the UK's current economic policy options and a plan for improving life for ordinary citizens via a sensible and realistic understanding of governments' limited ability

The objective of this study was developed a hand held olive harvester suitable for small farms. The portable machine was designed, fabricated and evaluated for its performance and techno-economic feasibility. Some physical and mechanical properties (dimensions, mass, fruit detachment force / mass ratio) of olive fruits that are related to the mechanical harvesting were measured and considered in the machine design. The developed machine was evaluated at three levels of head speed (700, 1100 and 1500 rpm) and two types of head length (H1 has stick length of 6 cm and H2 has stick length of 17 cm). The evaluation criteria were: productivity (Pm), fruit removal percentage (FR) and fruit damage percentage (FD), consumed energy (CE) and harvesting cost (HC). From the obtained results, it is deduced an equation and contour chart for the harvesting machine to predict the suitability of the machine to removal the fruit with different olive variety. The most suitable working parameters of the machine at 1100 rpm and 1500 rpm with (H2) where the maximum value of evaluation parameter were Pm of 91.5 kg h-1, FR of 97.7 %, FD of 6.23 %, CE of 1.31 W h kg-1and HC of 0.26 L.E kg-1.

This paper introduced the design process of shell-breaking machine for hand-stripping hickory-nut. Hand-stripping hickory-nut is an instant snack food. Like peanut, it’s easy to separate kernel from shell. The paper compared methods of shell-breaking nuts which is in common use, got inspiration from manual process and absorbs useful experience of traditional processing techniques combining with impact method. Then the mechanism of striking shelling in the way of mechanical knocking is designed. The design of robot percussion arm mechanism, feed mechanism, force regulating mechanism and principle prototype were proposed. The hemispherical grooves are approximately half spherical, which match the shape of hickory-nut. And then the component strike force can be reasonably reduced. The experiment results showed that the machine have good effect in shell-breaking for the Hickory-nut of different specification. Hickory-nut is not convenient to eat because the nuts are deeply embedded in the spaces formed from the hard shell. Hence, special tools are usually needed such as Hickory-nut clamp. Recently years, the hand-stripping hickory-nut is popular in the market. It is an instant snack food. As peanuts, it needs to separate kernel from shell while eating. The traditional processing method is to put it in the special mold and then strike it with the special hammer. But the productivity is low and it will produce secondary pollution easily. Besides, shortage of labor will lead to short supply. Thus mechanized production has become our urgent demand.

During software development and maintenance, predicting software bugs becomes essential. An essential activity of the quality assurance process, defect prediction at the beginning of the software development life cycle has received extensive research over the past two decades. Early detection of defective modules in software development can assist the development team in making efficient and effective use of the resources at hand to produce high-quality software in a short amount of time. Using a machine learning approach, which finds hidden patterns in software attributes, it is possible to recognize the problematic modules. In the NASA data set JM1, the suggested work is contrasted with various machine learning classification procedures. The limit of random forest speculation is higher than that by other multi-class classifiers because of the effect of bagging and feature selection. Since ensemble learning of random forest requires a ton of decision trees to acquire elite execution, the situation is not appropriate for carrying out the design on the limited-scale equipment like an embedded system. In this paper, we propose a boosted random forest, experimental outcomes show that the proposed technique, which comprises of less decision trees, has higher speculation capacity contrasting with the traditional technique. The experimental findings demonstrated that our proposed boosted random forest model results in greater levels of defect prediction accuracy, improving software quality.