MétaCan
Menu
Back to cohort
Record W4411340796 · doi:10.1016/j.grets.2025.100235

Performance of machine learning algorithms to evaluate the physico-mechanical properties of nanoparticle panels

2025· article· en· W4411340796 on OpenAlex
Derrick Mirindi, James Hunter, David Sinkhonde, Tajebe Bezabih, Frédéric Mirindi

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.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.

Bibliographic record

VenueGreen Technologies and Sustainability · 2025
Typearticle
Languageen
FieldEngineering
TopicAdditive Manufacturing and 3D Printing Technologies
Canadian institutionsUniversity of Manitoba
Fundersnot available
KeywordsNanoparticleComputer scienceMachine learningAlgorithmArtificial intelligenceMaterials scienceNanotechnology

Abstract

fetched live from OpenAlex

Nanoparticles significantly enhance the properties of wood-based materials, especially particleboards and wood panels. This review analyzes secondary data on nanoparticle integration in board production, aiming to evaluate the relationships among physical (water absorption (WA) and thickness swelling (TS)) and mechanical (modulus of rupture (MOR), modulus of elasticity (MOE); and internal bond (IB) strength) properties and to predict performance using machine learning (ML) algorithms. These algorithms include Pearson correlation, hierarchical clustering, and decision tree (DT) models. Results indicate that nanoparticles such as graphene oxide (GO), reduced graphene oxide (rGO), hydrolysis lignin, and calcium carbonate improve mechanical properties, with MOR values of 27.38–52.65 MPa and MOE of 2591.6–4680 MPa, meeting EN312 load-bearing standards. Zinc oxide nanoparticles yield superior dimensional stability by achieving a low TS of 9.33%. However, according to the American National Standard for Particleboard (ANSI/A208.1-1999), most nanoparticle boards produced met general-purpose standards except for WA and TS, which exceeded the maximum limits of 8% and 3%, respectively. Only crosslinked chitosan and zinc oxide nanoparticle panels meet the minimum requirements for TS (17%) and the maximum MOR (11.00 MPa) and MOE (1,800.00 MPa) for general purposes in dry conditions (furniture and interior fitments) according to the Brazilian standard (ABNT NBR). The Pearson correlation analysis reveals a strong relationship between board properties (R = 0.94 for WA–TS; R = 0.93 for MOR–MOE), confirming that nanoparticle treatments enhance performance while maintaining inherent material behavior. Hierarchical clustering grouped nanoparticles by performance: zinc oxide and chitosan+UF+epoxy formed a cluster with the lowest WA and TS, indicating optimal dimensional stability, while GO, rGO, and chitosan-based composites clustered with moderate values. For mechanical properties, APTES-modified nanocellulose, aluminum oxide, and zinc oxide formed a high-performance cluster (high MOR, MOE, IB). DT algorithms demonstrated high predictive accuracy (R 2 = 0 . 92 for WA-TS, 0.96 for MOR-MOE, and 0.80 for IB-MOE), identifying critical thresholds: WA below 29.73% corresponded to minimal TS (9.94%), MOR above 38.18 MPa led to MOE above 3598.86 MPa, and IB above 0.88 MPa corresponds to MOE greater than 2,747.99 MPa. This data-driven framework enables targeted nanoparticle selection to fabricate engineered wood products and can be included in industry quality control standards to advance sustainable material development through ML-guided optimization.

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)0.000
metaresearch head score (Gemma)0.001
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Bench or experimental · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.956
Threshold uncertainty score0.365

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.001
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.001
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 head0.014
GPT teacher head0.230
Teacher spread0.217 · 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