Search

Numerous studies emphasize accuracy in machine learning regression models, yet scalability and execution efficiency are often overlooked, critical for large datasets or extensive computations. This paper introduces a scalable, distributed Spark MLlib regression model through the best subset selection approach to predict Covid-19 statistics in India, demonstrating high accuracy, scalability, and execution efficiency. Notably, limited research focuses on tree-based regression, particularly gradient boost regression, in the context of the Covid-19 dataset. The proposed work optimizes regression models for accuracy and execution time on Spark clusters of varying sizes using the best subset selection approach. Evaluation encompasses Root Mean Square Error (RMSE), Mean Absolute Error (MAE), R2 Error for accuracy, and execution time analysis. Results indicate superior prediction accuracy in tree-based regression, with Gradient Boosted Tree Regression (GBTR) leading, and Random Forest Regression (RFR) surpassing Decision Tree Regression (DTR). Accuracy remains consistent across Python library, Spark MLlib on a single machine, and clusters of varying sizes, with Spark MLlib displaying lower execution times than Python's machine learning library on a single machine. Furthermore, execution times decrease substantially within Spark clusters, particularly for the iterative GBTR. This research uncovers scalability and execution efficiency aspects, highlighting tree-based regression's accuracy and advocating for Spark MLlib's efficacy in enhancing execution efficiency, especially across multi-node clusters. [ABSTRACT FROM AUTHOR]

Environmental and anthropogenic factors are known to drive fish community structure in aquatic systems across the world. This study investigates fish assemblages in lower order streams across contrasting landscapes in central and eastern India. We documented the species diversity of these monsoon driven lower order streams in the two regions. We also investigated the potential common environmental drivers of richness and diversity and effect of season in these tropical streams. The study was based on seasonal data on abundance of fishes and environmental parameters collected between 2015–2017 from streams in states of Madhya Pradesh and West Bengal. Species diversity were compared across regions and seasons, based on their richness (SR) as well as diversity (Shannon index H'). Drivers of overall richness and diversity were analyzed using multiple linear regression methods, based on best subset selection. Analysis of data revealed high diversity in these streams in both regions. Cyprinidae, Bagridae and Channidae were the most dominant families in both regions. Despite the geographical and local ecological differences across the regions, common environmental parameters were found to influence richness and diversity across the two regions, indicating these as being key drivers of fish community structure. Water flow was a common factor driving both richness and diversity across both regions. Our study revealed a lack of seasonal effect in structuring fish communities in tropical streams. With stream and river ecosystems facing increasing threats due to habitat alterations and water quality degradation in countries such as India, a clear understanding of regional and local drivers of community structure of aquatic fauna is crucial. These results on the role of common environmental factors across ecoregions provides baseline information for understanding their ecological roles and developing management plans for important river basins and fish conservation in future. [ABSTRACT FROM AUTHOR]

Lakes are functionally integral and biologically complex freshwater ecosystems which provide a vast array of ecosystem goods and services to society. Nowadays aquatic ecosystems are being used, misused and abused by diverse anthropogenic activities at an unprecedented rate. The management of lake water quality is usually directed to resolution of conflicts between maintenance of desirable water quality and human-induced degradation of aquatic environment. Nutrients play a decisive role in determining lake’s environmental state through regulation of its primary production. The present study on Saheb Bundh Lake located in Purulia District, West Bengal, India, was undertaken to assess the status of nutrients (nitrogen and phosphorus) subject to certain anthropogenic activities, and to construct models using best subset method which could be adopted as a nutrient management tool. The water samples were monitored for different physicochemical parameters adopting standard methods. It was found that the set of variables including turbidity, temperature, pH, redox potential and total hardness has been championed as the best subset of water quality explaining the dynamics of total phosphorus concentration of freshwater Saheb Bundh Lake. The suite of factors comprising dissolved oxygen, pH, temperature, turbidity and total hardness has been proved as the best subset for estimating total nitrogen concentration. The models developed have been validated for total phosphorus and total nitrogen concentration. For total phosphorus, the model values were found to be very close to the measured values but the values varied widely for total nitrogen, championing the former as a very potent down-to-earth model. [ABSTRACT FROM AUTHOR]