The recommender system for identifying a suitable area for exploration of mines by using singular value decomposition (SVD)

Message:
Article Type:
Research/Original Article (دارای رتبه معتبر)
Abstract:
Summary

In this paper, by using the mines information of Iran provinces and singular value decomposition, the provinces of Iran are clustered. By using these clusters information, a recommender system is developed to help geologists and mining engineers.
 

 Introduction

Mining exploration consists of prospecting, early exploration, detailed and complementary stages (even extraction process). Prospecting stage begins with gathering and studying the information of mining area and preliminary sampling. Based on thumbnail technical and economic review, the feasibility of entering the preliminary stage of exploration is realized. At the start of prospecting stage, the total capacity of the area for production of significant and economic mineral deposits must be identified. Usually detecting of valuable deposits is performed according to information and experience of geologists and mining engineers. Therefore, the stage of exploration involves high cost and risk. Thus, the methods that can reduce costs and uncertainties are considered. On the other hand, there is a collection of data related to the studied area and its surrounding with high dispersion that their analysis can lead researchers to the aim. By using mathematical methods in data mining such as clustering data, the mines with specific deposits according to similar geological conditions could be classified.
 

Methodology and Approaches

Singular value decomposition is one of the most powerful tools in mathematics. This method is widely used in clustering problems and recommender systems. In this paper, at the first stage, the provinces of Iran are clustered by using singular value decomposition and metal mines data which are extracted from the mines database. The provinces in the same clusters are similar in geological conditions. Finally, recommender system for exploration of a specific mine, the provinces that are in the cluster having higher  number of this mine are allocated top priority.
 

Results and Conclusions 

The results show that by using singular value decomposition and the mines information of Iran provinces, the provinces can be clustered.  Moreover, the provinces  in the same clusters, are similar in geological conditions. It is expected that, if a mineral deposit was discovered in a province, it would be found in another province in the same cluster too.

Language:
Persian
Published:
Journal of Aalytical and Numerical Methods in Mining Engineering, Volume:9 Issue: 19, 2019
Pages:
55 to 61
https://www.magiran.com/p2029576