The Exorde WorkSystems are the main components and act as the core for the whole ecosystem.
This platform is decentralized, open, and, as we have seen in a previous article, transparent.
This is where all contributors will work together to index the web as a whole, to extract its unstructured information, relationships, similarities, trends, and any pattern in the information flowing anywhere on the web. It is not important whether it is a platform or a media source. The Exorde platform continuously extracts information from the content being indexed by its contributors, through continuous and decentralized data science analysis.
Exorde is governed by its DAO (decentralized autonomous organization) and will use community votes and polls. Governance is decentralized among all community members, whether they are investors, participants, or workers.
Collectively, they can modify the rules and the internal parameters of the systems (rewards, constraints, deadlines, schedules, etc.) and have an integrated reputation system.
These mechanisms are designed to keep the community’s interests and governance aligned at all times, for the good of the community.
On a functional level, the working systems, designed in layers, are at the heart of the platform.
The working systems act as a digital data factory that feeds the Exorde knowledge graph. These are the layers of Exorde:
1) Working Systems: the first layer of the system where participants index different URLs extracted from the web and their data relationships according to predefined rules and guidelines. This set of rules is defined and approved by the Exorde community and can evolve to enhance the relevance and maximize the value creation of the data.
2) Data analysis system: the second layer of the system where participants perform continuous analysis of different types of data that make up the central Exorde database (a knowledge graph). It includes data clustering, trend analysis, labeling, tracing, and partitioning, and is performed using NLP sentence (or document) encoding models such as BERT (or other transformer-based deep learning models).
Textual entities (sentences, paragraphs, headings, etc.) are transformed into numerical vectors and then added to the central neural database, allowing indexing and querying by Exorde contributors.
These data operations will increase in quantity and diversity
diversity, over time, to meet demand and maximize the relevance of Exorde’s services and products.
➡️Feel free to join our active community on :
🟪Discord: https://discord.gg/AgRdQaMz5d
🟦Twitter: https://twitter.com/ExordeLabs
🟥Linkedin: https://www.linkedin.com/company/exorde
🟩Medium: https://medium.com/@ExordeLabs
🟧Website: https://exorde.network/