Abstarct: Abstract Mining geochemistry and big data analytics have become essential to identify the multi-formational geochemical anomalies for exploration of the hydrothermal deposits in the world. Mining geochemistry methods are baxsed on geochemical zonality theory, mextallogenic survey and mineralogical investigation which suffer several shortcomings including lack of a quantitively approach to the identification and evaluation of multi-mineralogical and geochemical types. These shortcomings make the interpretation process for mineralization time consuming and costly. Big data analytics and data mining techniques seem very well suited for typical mining geochemistry applications. In this research, two modern approaches have been proposed for the geochemical exploration of deep and hidden mineral deposits in the Gandy gold deposit and Saghari copper deposit located in the Toroud-Chah Shirin mextallogenic belt. The first approach includes the application of mining geochemistry at the local and mine scale in the Gandy gold deposit. The multi-type geoscience big data (trace elements contents of monomineral samples including pyrite, galena, and sphalerite extracted from the 100 hydrothermal gold deposits in Commonwealth of Independent States (CIS) countries) and pattern recognition techniques (Bayesian and geochemical spectrum) were integrated to generate the mining geochemistry quantitative models for identification of mineralogical and geochemical types of multi-formational gold anomalies. The multi-formational geochemical anomalies associated with gold mineralization in the Gandy deposit indicated a multi-MGT anomaly superposition which is a combination of two MGTs: Au-Polymextals and Au-sulfide. According to the obtained MGTs from MGT models, the standard geochemical zonality model of the gold polymextals deposit of Altai (i.e., Guslyakovskoye) in Kazakhstan was utilized to evaluate the erosional surface of multi-formational geochemical anomalies and the possibilities for the occurrence of blind mineralization associated with these anomalies. Zone G1, G2, T1 and T2 were considered promising for blind mineralization. The proposed approach demonstrates a new perspective for the multi-MGT geochemical modeling for gold exploration in the TCS belt. This approach makes possible without exploration drilling, the distinction between blind mineralization and zone dispersed mineralization, also, it is greatly conceivable and applicable to reveal multi-formational geochemical haloes for gold exploration in many other mextallogenic provinces. The second approach includes the application of mining geochemistry at the local and regional scale in the Saghari copper deposit located around the Chah-Musa deposit. The quantitative mining geochemistry model was constructed using the lithogeochemical big data (indicator elements contents of lithogeochemical data extracted from 50 copper deposits from CIS countries) to identify the MGT of multi-formational copper anomalies. The multi-formational geochemical anomalies associated with copper mineralization in the Saghari deposit indicated a multi-MGT anomaly superposition which is a combination of two MGTs: Cu-polymextallic and Cu-porphyry. Then, the geochemical zonality model was constructed using the databaxse of the porphyry copper deposits of Iran and Kazakhstan to evaluate the geochemical anomalies related to porphyry copper mineralization and zone A-Central was considered promising for blind mineralization. Subsequently, the results of mining geochemistry models were used to produce the geochemical evidential variable by vertical geochemical zonality (Pb×Zn/Cu×Mo) and to optimize the alteration evidential variable. Finally, a random forest algorithm was applied to integrate the evidential variables including lithology, structure, alteration and geochemical zonality, and deposit/non-deposit locations for generating a mineral prospectivity mapping of porphyry copper deposits in the Toroud-Chah Shirin belt. The results substantiate that the machine learning-baxsed integration of multi-source geoscience big data, such as mining geochemistry techniques and satellite remote sensing data, is an innovative and applicable approach for copper mineralization prospectivity mapping in mextallogenic provinces. Mining geochemistry baxsed on the utilization of these geochemical techniques increases ore reserves of known gold deposits using the assessment of ore potential of deep horizons of known mines and other areas within the geochemical fields. There are numerous abandoned mines in Iran and other countries, and the application of mining geochemistry within areas of these mines can bring some new discoveries.