Ensemble machine learning to accelerate industrial decarbonization: Prediction of Hansen solubility parameters for streamlined chemical solvent selection
Bibliographic record
Abstract
Several processes and strategies have been developed to promote the utilization of lignin and to facilitate its market adoption across a broad spectrum of applications within the expanding lignin bioeconomy. However, the inherent variability in lignin properties, resulting from diverse feedstock sources and varied recovery and downstream processing methods, remains a significant challenge. This highlights the critical need to investigate lignin's miscibility and reactivity with polymers and solvents, as most lignin valorization pathways involve mixing, blending, or solubilization. Accurate estimation of Hansen solubility parameters (HSP) is crucial for solvent selection in several fields such as polymer science, coatings, adhesives, lignin-based biorefineries and solvent-based carbon capture. Traditional methods for predicting HSP are time-consuming and involve complex experiments, especially in applications dealing with carbon dioxide and lignin solubility. This paper introduces a novel ensemble modeling methodology based on machine learning (ML) techniques for accurate HSP prediction using Simplified Molecular Input Line Entry System (SMILES) codes as entries. The methodology integrates different ML approaches, including deep and shallow learning, to enhance prediction accuracy. Decision fusion of individual ML models is achieved through a hybrid approach combining non-learnable and learnable methods, resulting in reduced errors and enhanced accuracy. The results highlight the effectiveness of the ensemble-based methodology, which achieved 99% accuracy in predicting dispersion solubility parameters, outperforming other individual ML techniques. The proposed generic methodology, from data preprocessing to decision fusion through diverse ML algorithms, can be applied to various chemical analytics beyond HSP prediction.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".