mlevcm

mlevcm is an R package for ML-based Epidemiological Vector Control Monitoring using FDA techniques for NIRS data.

The package includes functions to train, predict and assess the performance of machine learning models for spectral data using techniques from Functional Data Analysis---namely functional representation of spectra, spectra smoothing and penalised estimation of the coefficient function.

Citation

PM Esperança, DF Da, B Lambert, RK Dabiré, TS Churcher (2020) "Functional data analysis techniques to improve the generalizability of near-infrared spectral data for monitoring mosquito populations", BioRxiv [link]

Main Functions

The main functions in this package are:

• fdaML_train() trains a machine learning model and assesses its performance;
• fdaML_predict() produces predictions for a new set of observations, given a previously trained model;
• fdaPlot() displays visual diagnostics information and performance measures for objects generated by fdaML_train();
• fdaPlotPred() ---UNDER DEVELOPMENT--- displays visual diagnostics information and performance measures for objects generated by fdaML_predict();