MétaCan
Menu
Back to cohort
Record W1980703148 · doi:10.5589/m12-046

Variable selection strategies for nearest neighbor imputation methods used in remote sensing based forest inventory

2012· article· en· W1980703148 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

venuePublished in a venue whose home country is Canada.
no affNo Canadian affiliation: this work is invisible to an affiliation-only frame.
No Canadian affiliation. An affiliation-only frame, the usual design, would never have seen this work. It is one of the works that make the case for inverting the frame.

Bibliographic record

VenueCanadian Journal of Remote Sensing · 2012
Typearticle
Languageen
FieldEnvironmental Science
TopicRemote Sensing and LiDAR Applications
Canadian institutionsnot available
Fundersnot available
KeywordsImputation (statistics)OverfittingRandom forestGeographyMathematicsStatisticsForestryComputer scienceMissing dataArtificial intelligence

Abstract

fetched live from OpenAlex

AbstractWe examined the problem of selecting predictor variables for Nearest Neighbor (NN) imputation in remote sensing based forest inventory. Eighty-three variables were calculated from Airborne Laser Scanning data and aerial images, with responses being either dominant height or a set of five common stand attributes. Three different approaches were compared with select predictor variables. Analyses were repeated with three different NN imputation methods using a varying number of predictor variables. Results indicated that variable selection is justified, but it must be done properly. The most accurate method to select predictors was to minimize error using Simulated Annealing. For a single response, the most accurate imputation method was Random Forest proximity matrix-based imputation, whereas Most Similar Neighbor was the most accurate for five responses. An optimization-based distance metric also worked well. We also examined the degree to which different imputation methods are prone to overfitting as well as how to properly do cross-validation in NN imputation.On a examiné la problématique de la sélection des variables prédictives dans la procédure d'imputation par la méthode du plus proche voisin dans le contexte des inventaires forestiers réalisés par télédétection. Quatre-vingt trois variables ont été calculées à partir de données SLA (scanneur laser aéroporté) et d'images aériennes, les réponses étant soit la hauteur dominante ou un ensemble de cinq attributs courants de peuplement. Trois approches différentes ont été comparées pour la sélection des variables prédictives. Les analyses ont été répétées à l'aide de trois méthodes différentes d'imputation par le plus proche voisin en utilisant un nombre variable de variables prédictives. Les résultats ont montré que la sélection variable est justifiée, mais que celle-ci doit être faite correctement. La méthode la plus précise pour sélectionner les variables prédictives consistait à minimiser l'erreur à l'aide de la technique de recuit simulé. Pour une réponse unique, la méthode d'imputation la plus précise était l'imputation basée sur la matrice de proximité de type «Random Forest» (forêt aléatoire) alors que la méthode la plus précise pour les cinq réponses était la méthode d'imputation par le voisin le plus semblable «Most Similar Neighbor». Une mesure de distance basée sur une méthode d'optimisation a également donné de bons résultats. On a aussi étudié la propension des différentes méthodes d'imputation au sur-ajustement de même que la façon d'exécuter correctement une validation croisée dans le contexte de l'imputation par le plus proche voisin.[Traduit par la Rédaction] AcknowledgementsWe acknowledge Prof. Timo Pukkala and Dr. Tero Heinonen for their insights while implementing optimization routines. We also thank Prof. Jukka Tuomela and Mr. Pekka Savolainen for mathematical support.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.002
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.916
Threshold uncertainty score0.957

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0020.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.001
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.026
GPT teacher head0.287
Teacher spread0.261 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it