Overview
ETHERNOS OFFICIAL WHITE PAPER DOCUMENT, last updated January 2021
Last updated
ETHERNOS OFFICIAL WHITE PAPER DOCUMENT, last updated January 2021
Last updated
Calin-Laurentiu Popescu, Founder of ETHERNOS
https://www.linkedin.com/in/calin-popescu-96670374/
ETHERNOS - an Ai-based technology and platform that stores, processes, and classifies human memories based on non-biological captured data inputs from third-party sources.
The paper and concept propose a new way of efficiently collecting and processing both real-time and historical data that overlaps on one or many specific sub-sets of attributes such as timestamp, geolocation, and different classes of similarities with the day-to-day activities and with the labeled human memories generated on the chronological path of the daily life of the user.
With the help of the described technology, the day-to-day cognitive and semantic memories of a user are backed up in a digital format where data is indexed and processed in parallel to the moment of the aware experiencing each specific memory or group of memories. The similarity and relation between common elements within the memories are also indexed and stored. Data similarities and their relation to memories represent an important part of this paper and are detailed in the pages below.
The technology introduces a per-user user storage approach and aims to organize as efficiently as possible the raw data sources and files and their relation with the user memories by generating a time-stamped backup matrix stored in a blockchain ledger called the user memory meta chain ledger. The matrix of stored and incoming data streams is organized in the meta chain ledger and facilitates a proof of indexing to date of all the data that was fed to the wallet of a user up to a certain moment in time. Memories of multiple users that share common attributes such as geolocation, timestamp, or specific ML model signatures can be recreated on the fly without the need to store publicly each user entry.
Each memory is represented by blocks of meta chains containing factual information and cognitive meaning. In parallel to the way we naturally remember things, ETHERNOS uses a concept called affinity pools which are groups of similar memories that share similar factual blocks &, or cognitive blocks that make sense together. It is like thinking about your favorite food and remembering the memories of when you had that food. These memories are indexed by Ethernos in affinity pools.
New data packages are analyzed and processed in real-time in a memory wallet. This data is processed and split into factual and cognitive blocks and compared against historical existing memories using a set of chronological processes (pre-cortex checking, active cortex checking, and memory meta chains staking). The organized new memories are logged in a ledger called memory meta chain ledger that stores time-based historical information of all the memories logged, together with their presence in affinity pools and the signature of the semantic, factual, and cognitive models that helped to identify their position in the block during the staking process.
The relation between multiple user memories can be recreated on a higher network level by colliding and mapping multiple user ledgers or memory meta chains based on location, timestamp, device, and device connectivity area. Multiple users memories in the same date, time, or location can easily be reconstructed by merging the meta chains of these users and validating where the overlapping moments are and what was backed up for each of the users. The source of alignment is always time as it handles the proof of history of the events and mapped memories and indicates the present moment of the verification.
A user can establish certain privacy and anonymity measures and ultimately can choose to be discovered or share his memories or collections of memories with other users.
