Decision trees and random forests *

Casarin, Roberto; Facchinetti, Alessandro; Sorice, Domenico; Tonellato, Stefano Federico

doi:10.4324/9781003037903-2

Cited by 5 publications

(3 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…6A). This type of plot can be used to consolidate decision trees used to train a random forest machine learning model (Casarin et al, 2021). Although much more extensive data are required to generate a such model, the depicted relationships are predictive of sequential mutations that could, for example, be observed in clinical isolates that typically accumulate numerous mutations throughout a chronic infection (Malone, 2015).…”

Section: Resultsmentioning

confidence: 99%

Identification of cyclic-di-GMP-modulating protein residues by bi-directionally evolving a social trait in Pseudomonas fluorescens

Kessler

Kim

2022

Preprint

View full text Add to dashboard Cite

Microbial communities comprise densely packed cells where competition for space and resources are fierce. These communities, generally referred to as biofilms, provide advantages to individual cells against immunological and antimicrobial intervention, dehydration, and predation. High intracellular pools of cyclic diguanylate monophosphate (c-di-GMP) cause individual cells to aggregate during biofilm formation through the production of diverse extracellular polymers. Genes that encode c-di-GMP enzymes or their regulators are commonly mutated during chronic infections due to enhanced resistance to phagocytosis and antibiotics. Genome annotations predict the presence of numerous c-di-GMP catalytic enzymes in most bacterial species, but the functionality and regulatory control of the vast majority remain unconfirmed. Here, we begin to fill this gap by utilizing an experimental evolution system in Pseudomonas fluorescens Pf0-1, which repeatedly produces a unique social trait through bidirectional transitions between two distinct phenotypes converging on c-di-GMP modulation. Parallel evolution of 33 lineages captured 147 unique mutations among 191 evolved isolates in genes that are empirically demonstrated, bioinformatically predicted, or previously unknown to impact the intracellular pool of c-di-GMP. Quantitative chemistry confirmed that each mutation causing the phenotypic shift either amplifies or reduces c-di-GMP production. We integrate our data with current models of known regulatory and catalytic systems, describe a previously unknown relationship between branched-chain amino acids and c-di-GMP production, and predict functions of several new proteins that either regulate or catalyze c-di-GMP production. Sequential mutations that continuously disrupt or recover c-di-GMP production across discrete functional elements suggest a complex and underappreciated interconnectivity within the c-di-GMP regulome.

show abstract

Section: Resultsmentioning

confidence: 99%

Identification of cyclic-di-GMP-modulating protein residues by bi-directionally evolving a social trait in Pseudomonas fluorescens

Kessler

Kim

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The random forest algorithm requires an enormous difference between the decision trees with no correlation. If there is no strong dependency between the weak classifiers, the trees can be generated in parallel [28]. Random forest uses autonomous sampling to extract multiple samples from the original data.…”

Section: Methodsmentioning

confidence: 99%

Development of heart attack prediction model based on ensemble learning

Hasan

Saleh

2021

EEJET

View full text Add to dashboard Cite

With the advent of the data age, the continuous improvement and widespread application of medical information systems have led to an exponential growth of biomedical data, such as medical imaging, electronic medical records, biometric tags, and clinical records that have potential and essential research value. However, medical research based on statistical methods is limited by the class and size of the research community, so it cannot effectively perform data mining for large-scale medical information. At the same time, supervised machine learning techniques can effectively solve this problem. Heart attack is one of the most common diseases and one of the leading causes of death, so finding a system that can accurately and reliably predict early diagnosis is an essential and influential step in treating such diseases. Researchers have used various data mining and machine learning techniques to analyze medical data, helping professionals predict heart disease. This paper presents various features related to heart disease, and the model is based on ensemble learning. The proposed system involves preprocessing data, selecting attributes, and then using logistic regression algorithms as meta-classifiers to build the ensemble learning model. Furthermore, using machine learning algorithms (Support Vector Machines, Decision Tree, Random Forest, Extreme Gradient Boosting) for prediction on the Framingham Heart Study dataset and compared with the proposed methodology. The results show that the feasibility and effectiveness of the proposed prediction method based on group learning provide accuracy for medical recommendations and better accuracy than the single traditional machine learning algorithm.

show abstract

“…This ensures that the model can handle the correlation between features and grows somewhat uncorrelated trees. See Casarin et al (2021) for an introduction to random forests with applications.…”

Section: Random Forestmentioning

confidence: 99%

Learning from Experts: Energy Efficiency in Residential Buildings

Billio,

Casarin,

Costola

et al. 2023

SSRN Journal

View full text Add to dashboard Cite

Decision trees and random forests *

Cited by 5 publications

References 1 publication

Identification of cyclic-di-GMP-modulating protein residues by bi-directionally evolving a social trait in Pseudomonas fluorescens

Identification of cyclic-di-GMP-modulating protein residues by bi-directionally evolving a social trait in Pseudomonas fluorescens

Development of heart attack prediction model based on ensemble learning

Learning from Experts: Energy Efficiency in Residential Buildings

Contact Info

Product

Resources

About