Cryptocurrency mining selection system and method

Abstract

A system and method of optimizing cryptographic mining yields includes analyzing, by a cryptocurrency mining selection system, data associated with factors of interest for one or more cryptocurrencies using machine learning algorithms. Data that is determined to be predictive of the future value of newly mined tokens is used to determine which tokens will have the highest and lowest future values. Based on the predicted value of tokens in the future and the current value of those tokens for each cryptocurrency, the system outputs one or more instructions to buy tokens in cryptocurrencies predicted to increase in value, to sell tokens in cryptocurrencies predicted to decrease in value, and to instruct associated cryptocurrency mining hardware to switch to generating new tokens in one or more selected cryptocurrencies to maximize yields.

1 - 20 . (canceled) 
     
     
         21 . A token mining selection system, comprising a server, wherein the server comprises a processor and a memory, and a plurality of mining devices, wherein each of the plurality of mining devices is configured to perform token mining activities; wherein the processor is configured to:
 (a) receive a market input dataset from one or more market data sources, and determine a set of market characteristics for one or more mineable tokens based on the market input dataset, wherein the set of market characteristics includes at least a value;   (b) receive a crowd input dataset from one or more crowd data sources, and determine a set of crowd characteristics for each of the one or more mineable tokens based on the crowd input dataset, wherein the set of crowd characteristics includes at least a sentiment;   (c) monitor at least one public distributed ledger and at least one private distributed ledger and produce a distributed ledger dataset;   (d) determine a set of mining characteristics for each of the one or more minable tokens, wherein the set of mining characteristics includes at least a degree of difficulty, a hashrate, and a number of wallets;   (e) create a predictive model for the one or more minable tokens based on the set of market characteristics, the set of crowd characteristics, and the set of mining characteristics for each of the one or more minable tokens, wherein the processor is configured to, when creating the predictive model:
 (A) create the predictive model using a deep neural network based on the set of market characteristics, the set of crowd characteristics, and the set of mining characteristics; 
 (B) periodically validate the predictive model based on a sensitivity analysis of parameters used by the predictive model; 
 (C) periodically perform new training of the predictive model based upon the periodic validation; and 
   (g) provide an instruction for each of the plurality of mining devices based on analysis by the predictive model of the one or more mineable tokens, wherein the plurality of mining devices are configured to, in response to the instruction, perform token mining activities on a distributed ledger associated with one or more of the mineable tokens.