Scholarlink

<p>Machine learning (ML) algorithms have demonstrated significant potential for making accurate predictions in drug formulation processes. This study aimed to evaluate the performance of five different ML models in predicting naproxen solubility and to identify the most accurate algorithms that closely align with experimental results. Using a dataset of 724 samples and 26 features, five ML algorithms—Linear Regression (LR), Decision Tree (DT), Random Forest (RF), K-Nearest Neighbors (KNN), and Neural Networks—were trained to predict naproxen solubility across various solvents and temperatures. All algorithms achieved low error rates, with DT, RF, and KNN showing particularly strong agreement with experimental data. The study also compared thermodynamic models with ML algorithms, revealing that ML approaches provide more accurate predictions. Overall, machine learning offers powerful tools to improve solubility prediction in the pharmaceutical industry. By providing more reliable and efficient methods, ML can enhance the drug development process, potentially reducing both time and costs.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Prediction of Naproxen Solubility at Various Temperatures and Concentrations using Machine Learning Approaches

  • 期刊论文

  • 发表日期：

    2025年6月30日

查看全文
  • Zahra Besharati,
  • Seyed Hossein Hashemi,
  • Aziz Babapoor,
  • Ahmad Seyfaee

摘要

Machine learning (ML) algorithms have demonstrated significant potential for making accurate predictions in drug formulation processes. This study aimed to evaluate the performance of five different ML models in predicting naproxen solubility and to identify the most accurate algorithms that closely align with experimental results. Using a dataset of 724 samples and 26 features, five ML algorithms—Linear Regression (LR), Decision Tree (DT), Random Forest (RF), K-Nearest Neighbors (KNN), and Neural Networks—were trained to predict naproxen solubility across various solvents and temperatures. All algorithms achieved low error rates, with DT, RF, and KNN showing particularly strong agreement with experimental data. The study also compared thermodynamic models with ML algorithms, revealing that ML approaches provide more accurate predictions. Overall, machine learning offers powerful tools to improve solubility prediction in the pharmaceutical industry. By providing more reliable and efficient methods, ML can enhance the drug development process, potentially reducing both time and costs.