Abstract :

Mineral Hardness And Density Serve As Fundamental Physical Characteristics In Fields Such As Geology, Mining, And Material Science, Where They Are Widely Used For Mineral Identification And Classification. Conventionally, Hardness Was Evaluated Using The Mohs Scale Through Manual Scratch Testing, In Which Minerals Were Compared Against Standard Reference Specimens. This Process Typically Demanded Laboratory Facilities, Domain Expertise, And Considerable Time To Achieve Reliable Results. Moreover, It Was Often Labor-intensive And Prone To Inconsistencies Arising From Human Error And Varying Experimental Conditions. To Overcome These Challenges, The Proposed Work Presents A Machine Learning Driven Multimodel Regression Framework Designed To Estimate Mineral Hardness And Specific Gravity Based On Their Chemical And Physical Attributes. The System Is Developed As A Web-based Application Using The Django Framework, Enabling An Interactive And Accessible Platform. Multiple Regression Techniques Are Employed To Enhance Predictive Accuracy, Including CatBoost Regressor (CBR), AdaBoost Regressor (ABR), Random Forest Regressor (RFR), And A Hybrid ML And Deep Learning (DL) Model As Greedy Tab Transformer (GTT). These Models Learn The Relationships Between Mineral Features And Target Variables To Generate Accurate Predictions For Mohs Hardness And Specific Gravity. Experimental Results Indicate That The GTT Model Achieves Superior Predictive Performance Compared To The Other Approaches. In Addition To Prediction, The System Incorporates Functionalities Such As Model Training, Comparative Analysis Of Algorithms, And Prediction Modules For Evaluating Mineral Properties. By Combining ML Techniques With A Web-based Interface, The System Provides A More Efficient, Scalable, And Reliable Alternative To Traditional Manual Testing Methods For Mineral Property Prediction.

Published:

09-4-2026

Issue:

Vol. 26 No. 4 (2026)

Page Nos:

2117-2126

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite