Scholarly record
LOGICAL INFORMATION MODELS FOR PREDICTION AND EXPRESS-EVALUATION OF NEW GOLD ORE DEPOSITS IN THE ARCTIC ZONE OF RUSSIA
Abstract
Mathematical data processing has allowed us to build logical information models based on machine learning (select a lot of informative features (elements), indicating their separating weights and ranges of changes in their values (intervals-indicators), typical for each of the groups of deposits of various formation types). The constructed logical-information models are based on a representative analytical database of 95 gold deposits and gold occurrences in the North-East of Russia. Samples were studied by modern analysis methods (AAS, ICP-MS and RFA) to identify the geochemical features of gold ore deposits of different formation types from database. Ores were analyzed for 52 elements. The models are constructed for five formation types of deposits: Au-Ag, epithermal; Au-quartz; Au-sulfide (disseminated ores); Cu-Mo-Au-porphyry; pyrite-polymetallic, enriched with Au and Ag. Developed rules reliably identify the formation type of new objects (recognition quality = 0.85). It is shown that the created models can be used for the rapid assessment of new gold ore occurrences in the Arctic zone of Russia. In order to determine the formation type of ore occurrence by peat samples, it is necessary to calculate the total weight of the indicator data of the samples for each model using the elements for which the value in the sample falls within the interval indicator of a certain formation type. The estimated new object refers to the formation type for which the total weight of indicator data will have a maximum value.
Publication Impact Profile
Publication details
References0
Structured references will appear here after the reference import pass. The count is preserved now so the scholarly record is not incomplete.
View or Download full articleAccess options
SWS access login
Login as SWS Scientific CommitteeLogin as SWS Scientific PartnerLogin as SWS AuthorAuthors and approved SWS contributors will read and export their own linked papers after identity matching by SWS profile, email and SGEM GlobalID.
For librarian assistance: [email protected]
Purchase Instant Access
- Article can be downloaded after successful payment.
- Article may be used according to SWS library access terms.
- Article cannot be redistributed.

