Quadrant Protocol - The Need for a Democratic and Authentic Data Marketplace

in ai •  last year  (edited)

“Data!data!data!" he cried impatiently. "I can't make bricks without clay.”

― Sir Arthur Conan Doyle, The Adventure of the Copper Beeches

As evident from this quote from a book which was written way back in 1892, "Data" has always been an important subject serving a variety of purposes. As time has gone by, data became more and more relevant. In today's digital era, you could undoubtedly claim that Data drives the world economy with every industry - manufacturing, engineering, healthcare, energy, mobility etc depending on it, one way or the other.

Therefore, Enterprises both big and small, need to employ data-driven models and sophisticated technologies such as machine learning & artificial intelligence to target their customers, drive growth and become sustainable. However, there is a huge disparity between big and small companies when it comes to the issue of access to data (data asymmetry). This apparent asymmetry of power is having a detrimental effect on most industries, especially the tech industry, where the likes of Google & Facebook have the monopoly. This highlights the need for democratic data marketplace that caters to the needs of small companies. Coincidentally, we have an emerging, blockchain based decentralized protocol called Quadrant that can fix these issues.

ai-monitoring-3262759_640.jpg

Image Courtesy - sujins, Pixabay

Quadrant's primary goal is to address the data gap between the haves and have-nots in the AI segment. AI and Machine Learning are extremely sophisticated technology segments that require enormously large datasets for operations such as "supervised learning" and "model training". The problem is not so much regarding the existence of this enormous data. In fact, data exists but is spread across diverse sectors with very little organization. So, the key is to organize this huge amounts of data spread across multiple industries and make it available for companies. Quadrant is aiming to do this by creating a decentralized ecosystem in which all the stakeholders (creators, verifiers, buyers of data) and their respective tasks are well defined. Earmarking tasks will remove the ambiguity associated with the marketplace as the participant knows their role well. Further, the Quadrant protocol advocates a policy of fair compensation to the original creators of data. This is often the missing link for many marketplaces as the absence of a proper creator incentivization process negatively affects the integrity of the marketplace.

Besides, Quadrant Protocol also upholds the principle of Data Authenticity i.e the data supplied needs to be profoundly authentic. Why? - When it comes to AI and Machine Learning applications, only authentic data can be used to train models for getting accurate results. Bogus data means the entire model becomes flawed and the system shall eventually fail. This could be true for any AI/Machine learning application - Visual Recognition, Speech synthesis, Machine Translation Natural language processing etc. To address this issue, Quadrant will implement a Data Stamping technique wherein the data will be stamped by certain stakeholders (known as Nurseries) for authenticity. This will enable the data recipients (Elons) to trace their data sources, thus eliminating frauds.

authentic.jpg

Conclusion

Data-driven models powered by Machine Learning and AI will eventually take over every industry and enterprises will need to find ways to build systems utilizing these models/algorithms to succeed. The data asymmetry can only be solved by having a democratic marketplace that upholds the principles of data authenticity or else the results will be extremely poor. Quadrant is trying to incorporate these principles into a decentralized ecosystem that will also fairly compensate the original data creators. Let's hope Quadrant can disrupt the existing status quo of centralized data economy dominated by the likes of Facebook and Google.

To learn more about Quadrant, visit the Website

Join the Telegram Group
Read the White Paper
BitcoinTalk ANN Thread
BitcoinTalk Bounty Thread
My BitcoinTalk Profile

Authors get paid when people like you upvote their post.
If you enjoyed what you read here, create your account today and start earning FREE STEEM!
Sort Order:  

Cool, thanks for sharing.

You're welcome 😊