ScholarLens Back to search
2007•Xinjiang GeologyRequires access

THE APPLICATION OF BP NET IN CHEMICAL CLASSIFICATION OF IGNEOUS ROCKS

Feng Liu

Open publisher page 0 citations

Abstract

This essay briefly presents some problems on the classification of igneous rocks nowadays, such as some kinds of rocks have no place in the traditional classification diagrams, some overlapping area exists because of less rock chemical parameters, and some wrong classifications resulted because of the existence of glass matter. Taking the ingredients of the eleven oxides of the igneous rocks as the variable, which include SiO2、TiO2、Al2O3、FeO,etc., the paper puts forward a new way to set up a chemical classification model of igneous rocks by using artificial neural net-the BP Net. With correct judgment rate of 97.8% on 46 rock samples, the experiment results show that the method is reasonable and feasible.

About this research paper

What this paper is about

This essay briefly presents some problems on the classification of igneous rocks nowadays, such as some kinds of rocks have no place in the traditional classification diagrams, some overlapping area exists because of less rock chemical parameters, and some wrong classifications resulted because of the existence of glass matter. Taking the ingredients of the eleven oxides of the igneous rocks as the variable, which include SiO2、TiO2、Al2O3、FeO,etc., the paper puts forward a new way to set up a chemical classification model of igneous rocks by using artificial neural net-the BP Net. With correct judgment rate of 97.8% on 46 rock samples, the experiment results show that the method is reasonable and feasible.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

This essay briefly presents some problems on the classification of igneous rocks nowadays, such as some kinds of rocks have no place in the traditional classification diagrams, some overlapping area exists because of less rock chemical parameters, and some wrong classifications resulted because of the existence of glass matter. Taking the ingredients of the eleven oxides of the igneous rocks as the variable, which include SiO2、TiO2、Al2O3、FeO,etc., the paper puts forward a new way to set up a chemical classification model of igneous rocks by using artificial neural net-the BP Net. With correct judgment rate of 97.8% on 46 rock samples, the experiment results show that the method is reasonable and feasible.

Key concepts: Igneous rock, Geology, Net (polyhedron), Artificial neural network, Set (abstract data type), Variable (mathematics), Geochemistry, Mineralogy

Related papers

Related paper 1Related paper 2Related paper 3
Back to paper searchBrowse research topicsOriginal source
THE APPLICATION OF BP NET IN CHEMICAL CLASSIFICATION OF IGNEOUS ROCKS — Research Paper | ScholarLens