首页|Researcher at Universiti Teknologi PETRONAS Details Research in Machine Learning (Fundamental error in tree-based machine learning model selection for reservoir characterisation)
Researcher at Universiti Teknologi PETRONAS Details Research in Machine Learning (Fundamental error in tree-based machine learning model selection for reservoir characterisation)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning DailyNews Daily News - Investigators publish new report on ar tificial intelligence. According to news reportingfrom Seri Iskandar, Malaysia, by NewsRx journalists, research stated, “Over the past two decades, machinelea rning techniques have been extensively used in predicting reservoir properties. While this approachhas significantly contributed to the industry, selecting an appropriate model is still challenging for mostresearchers.”The news journalists obtained a quote from the research from Universiti Teknolog i PETRONAS:“Relying solely on statistical metrics to select the best model for a particular problem may not alwaysbe the most effective approach. This study e ncourages researchers to incorporate data visualization intheir analysis and mo del selection process. To evaluate the suitability of different models in predic tinghorizontal permeability in the Volve field, wireline logs were used to trai n Extra-Trees, Ridge, Bagging, andXGBoost models. The Random Forest feature sel ection technique was applied to select the relevant logsas inputs for the model s. Based on statistical metrics, the Extra-Trees model achieved the highest testaccuracy of 0.996, RMSE of 19.54 mD, and MAE of 3.18 mD, with XGBoost coming in second. However,when the results were visualised, it was discovered that the X GBoost model was more suitable for theproblem being tackled. The XGBoost model was a better predictor within the sandstone interval, whilethe Extra-Trees mode l was more appropriate in non-sandstone intervals. Since this study aims to pred ictpermeability in the reservoir interval, the XGBoost model is the most suitable.”
Universiti Teknologi PETRONASSeri IskandarMalaysiaCyborgsEmerging TechnologiesMachine Learning
NETL
NSTL
