Preact (Predict + Act) is a prediction service for Crypto currencies, exclusive to Total Cryptos.
Total Cryptos has developed a price prediction tool based on the most advanced Machine Learning techniques available. The Total Cryptos Machine Learning Engine (TC-MLE) is an artificial intelligence engine that runs the most efficient form of Recurrent Neural Network (RNN), Gated Recurrent Unit (GRU). Data sources include semantic data (text, keywords, phrases) taken from social media sources in real time via API. This is combined with price data and trained in the engine, which outputs future prices based on this model. Prices are given in varying time sets; 1 day, 3 day, 1 week, 1 month, 3 months, 1 year, 3 years. Probabilities are provided based on internal data metrics. Predictions are recorded for accuracy and for future analysis. As of now, this is a read-only service, it requires traders to buy/sell based on the signals by themselves. In the future it may incorporate an execution algorithm which automatically trades based on user parameters.
More about Gated Recurrent Unit
From Wikipedia: Gated recurrent units (GRUs) are a gating mechanism in recurrent neural networks, introduced in 2014 by Kyunghyun Cho et al. Their performance on polyphonic music modeling and speech signal modeling was found to be similar to that of long short-term memory. However, GRUs have been shown to exhibit better performance on smaller datasets. They have fewer parameters than LSTM, as they lack an output gate. There are several variations on the full gated unit, with gating done using the previous hidden state and the bias in various combinations, and a simplified form called minimal gated unit.
Crypto Currency Price Machine Learning Precognition Engine
Crypto Currency is a mysterious technology, we don't even know who originally created Bitcoin. As of 2018, many institutions are building their own enterprise grade Blockchain solutions. But with more than 2,000 + Crypto Currencies and growing, there still lacks a cohesive doctrine outlining the architecture of the Crypto Currency market. Like the Foreign Exchange market, the Crypto Currency market has no design, it is an amalgamation of 'exchanges,' coin developers, users, investors, and other participants.
Traditional market analysis, such as technical analysis, fundamental analysis, or other methods - simply will not work on Crypto. Bitcoin rose to 20,000 for absolutely no reason - it was emotional. Emotions and social trends are not really predictable, at least insofar as they cannot be quantified and analyzed on a global scale. Even if this were possible, good Macro analysis would not have likely predicted the quick rise of Bitcoin in the fall of 2017. With this Machine Learning engine we are going to the heart of the market and creating a 'robot' which will automatically 'learn' what's going on in the market. Some investors, not the original whales (early adopters), noticed Bitcoin in 2014, 2015, and 2016 and bought. Their HODL (Hold on for Dear Life) strategy worked, as Bitcoin finally topped 20,000 (depending on which exchange is referenced).
It was Google actually that allowed Bitcoin to flourish, as referral marketing kicked in and investors told their friends about it, finally leading to a supply squeeze driving the price higher. We are using data from Google as input into the engine, and the engine is running on Google Cloud.
This Machine Learning Engine (MLE) is available to all paid Total Cryptos Members (only $49/month). Click here to learn more about the benefits of Premium Membership.
To deal with the problem of time being fluid, we have developed a system to manage this logically we call "Time Slices" - we have divided time into past, present, and future. From these 3 values, past and future can have additional values such as F+1 F+2 etc. The present is always the present, it is similar to zero in the Complex Number system. Zero is a reference point, where the rest of the Euclidean Space starts - it is the origin.
Server Infrastructure Diagram
Total Cryptos has 3 servers running the necessary infrastructure to handle the entire service full cycle; 2 on AWS and 1 on Google Cloud. The 'brain' running the Machine Learning Engine is at Google Cloud. The Total Crypto data engine is run at AWS using the RDS service. The Portal structure is a modified Joomla on LAMP stack for ease of use and customization ability.
While it is correct to say that the future does not exist, it soon will. Quantum Physics has begun to attempt to understand time as a dimension (see figure):
There are a limited amount of possible futures based on the present, with the current laws of physics and our current timeline. The potential futures grow as time increases in the future.