RESUMO
Time series analysis and prediction have attained significant attention from the research community in the past few decades. However, the prediction accuracy of the models highly depends on the models' learning process. In order to optimize resource usage, a better learning methodology, in terms of accuracy and learning time, is needed. In this context, the current research work proposes EvoLearn, a novel method to improve and optimize the learning process of neural-based models. The presented technique integrates the genetic algorithm with back-propagation to train model weights during the learning process. The fundamental idea behind the proposed work is to select the best components from multiple models during the training process to obtain an adequate model. To demonstrate the applicability of EvoLearn, the method is tested on the state-of-the-art neural models (namely MLP, DNN, CNN, RNN, and GRU), and performances are compared. Furthermore, the presented study aims to forecast two types of time series, i.e. air pollution and energy consumption time series, using the developed framework. In addition, the considered neural models are tested on two datasets of each time series type. From the performance comparison and evaluation of EvoLearn using a one-tailed paired T-test against the conventional back-propagation-based learning approach, it was found that the proposed method significantly improves the prediction accuracy.
RESUMO
Pearlmillet-chickpea cropping system (PCCS) is emerging as an important sequence in semi-arid regions of south-Asia owing to less water-requirement. However, chickpea (dry-season crop) faces comparatively acute soil moisture-deficit over pearlmillet (wet-season crop), limiting overall sustainability of PCCS. Hence, moisture-management (specifically in chickpea) and system intensification is highly essential for sustaining the PCCS in holistic manner. Since, conservation agriculture (CA) has emerged is an important climate-smart strategy to combat moisture-stress alongwith other production-vulnerabilities. Hence, current study comprised of three tillage systems in main-plots viz., Complete-CA with residue retention (CAc), Partial-CA without residue-retention (CAp), and Conventional-tillage (ConvTill) under three cropping systems in sub-plots viz., conventionally grown pearlmillet-chickpea cropping system (PCCS) alongwith two intensified systems i.e. pearlmillet-chickpea-fodder pearlmillet cropping system (PCFCS) and pearlmillet-chickpea-mungbean cropping system (PCMCS) in split-plot design. The investigation outcomes mainly focused on chickpea (dry-season crop) revealed that, on an average, there was a significant increase in chickpea grain yield under CAc to the tune of 27, 23.5 and 28.5% under PCCS, PCFCS and PCMCS, respectively over ConvTill. NPK uptake and micronutrient (Fe and Zn) biofortification in chickpea grains were again significantly higher under triple zero-tilled CAc plots with residue-retention; which was followed by triple zero-tilled CAp plots without residue-retention and the ConvTill plots. Likewise, CAc under PCMCS led to an increase in relative leaf water (RLW) content in chickpea by ~ 20.8% over ConvTill under PCCS, hence, ameliorating the moisture-stress effects. Interestingly, CA-management and system-intensification significantly enhanced the plant biochemical properties in chickpea viz., super-oxide dismutase, ascorbate peroxidase, catalase and glutathione reductase; thus, indicating their prime role in inducing moisture-stress tolerance ability in moisture-starved chickpea. Triple zero-tilled CAc plots also reduced the N2O fluxes in chickpea but with slightly higher CO2 emissions, however, curtailed the net GHG-emissions. Triple zero-tilled cropping systems (PCFCS and PCMCS) both under CAc and Cap led to a significant improvement in soil microbial population and soil enzymes activities (alkaline phosphatase, fluorescein diacetate, dehydrogenase). Overall, the PCCS system-intensification with mungbean (PCMCS) alongwith triple zero-tillage with residue-retention (CAc) may amply enhance the productivity, micronutrient biofortification and moisture-stress tolerance ability in chickpea besides propelling the ecological benefits under semi-arid agro-ecologies. However, the farmers should preserve a balance while adopting CAc or CAp where livestock equally competes for quality fodder.
