Identifying Critical Physicochemical Properties to Predict Microbial Community in a River Using Optimized Multi-Output Machine Learning
Abstract
Microorganisms play an important role in mediating global biochemical cycling in rivers. Understanding the relationship between earth's microbial community and the environment is the main goal of microbial ecology analysis. This study determines critical physiochemical parameters related to microbial community in a river through statistical analysis method. Sixty data points, which collected directly from one of the rivers from Southeast China, were pre-processed. Feature reduction, feature subset selection, and expert engineering judgment method were applied as a dimensional reduction technique. Then, a metaheuristic-optimized multiple-input multiple-output machine learner, which supports the adjustment to the multiple-output prediction form, was used in bioclimatic modeling. The accuracy of prediction and applicability of the model can help microbiologists and ecologists in quantifying the predicted microbial species for further experimental planning with minimal expenditure. This study provides a systematic approach for determining significant physicochemical properties by dimensionality reduction and the applicability of the approach can help microbiologists and ecologists predict future functions of microbial ecology when facing dramatic changes of environmental conditions caused by global warming scenario.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2022
- Bibcode:
- 2022AGUFM.B22F1504L