---
title: Comparing ML Algorithms for Item Difficulty Prediction
url: https://www.ml-quant.com/papers/repec/gam-jmathe-v-11-y-2023-i-19-p-4104-d-1249993/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-09-26
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: RePEc:gam:jmathe:v:11:y:2023:i:19:p:4104-:d:1249993
source_url: https://econpapers.repec.org/scripts/redir.pf?u=https%3A%2F%2Fwww.mdpi.com%2F2227-7390%2F11%2F19%2F4104%2Fpdf%3Bh%3Drepec%3Agam%3Ajmathe%3Av%3A11%3Ay%3A2023%3Ai%3A19%3Ap%3A4104-%3Ad%3A1249993
featured: 2023-10-12
citations: unknown
topic: ML & AI Methods
---


# Comparing ML Algorithms for Item Difficulty Prediction

A comparison of machine learning methods for predicting English reading comprehension test difficulty found that elastic net was best for continuous prediction, while random forests excelled in classification tasks.

- Source: https://econpapers.repec.org/scripts/redir.pf?u=https%3A%2F%2Fwww.mdpi.com%2F2227-7390%2F11%2F19%2F4104%2Fpdf%3Bh%3Drepec%3Agam%3Ajmathe%3Av%3A11%3Ay%3A2023%3Ai%3A19%3Ap%3A4104-%3Ad%3A1249993
- Identifier: RePEc:gam:jmathe:v:11:y:2023:i:19:p:4104-:d:1249993
- Released: 2023-10-12
- First featured: Quant Letter No. 20 (2023-10-12): https://www.ml-quant.com/issues/2023-10-12/
- Citations (Semantic Scholar): not tracked
- Published in: not yet
- Topic: ML & AI Methods

## Related

- [Enhanced Local Explainability and Trust Scores with Random Forest Proximities](https://www.ml-quant.com/papers/arxiv/2310.12428/): The study introduces a new method to explain the performance of random forest models by viewing them as adaptive weighted K nearest-neighbors models, offering a localized understanding of model predictions.
- [Quantile Regression using Random Forest Proximities](https://www.ml-quant.com/papers/arxiv/2408.02355/): The article introduces a new method for calculating quantile regressions from random forests, showing improved performance and efficiency in predicting the average daily volume of corporate bonds.
- [Case-based Explainability for Random Forest: Prototypes, Critics, Counter-factuals and Semi-factuals](https://www.ml-quant.com/papers/arxiv/2408.06679/): The study explores Explainable Case-Based Reasoning methods to clarify the results of black-box machine learning algorithms, using a technique to extract the learned distance metric from Random Forests, and assesses their explanatory power.
- [Financial Fraud Detection System Based on Improved Random Forest and Gradient Boosting Machine (GBM)](https://www.ml-quant.com/papers/arxiv/2502.15822/): The paper suggests a financial fraud detection system that uses an improved Random Forest and Gradient Boosting Machine model, offering an efficient and reliable solution for detecting financial fraud.
- [Survival Models for Startup Failures](https://www.ml-quant.com/papers/repec/bba-j00005-v-1-y-2023-i-3-p-1-15-d-264/): The research finds that advanced machine learning models like MTLR and Random Forest are more accurate in predicting startup failures than standard models.
- [Predicting Corporate Fraud with Machine Learning](https://www.ml-quant.com/papers/repec/kap-jbuset-v-186-y-2023-i-1-d-10-1007-s10551-022-05120-2/): The study applies a machine learning model using the GONE framework to predict corporate fraud in China, revealing that the Random Forest model is superior and that exposure variables are vital for accurate prediction.