Decentralizing the generation of market alpha using blockchain data

Transcript

present No yes there will be streaming hello Bonjour and maximal ostrich front

of sentiment and we are working on a specific topic for three years already so I wanted to share something what we've been able to achieve and how it's possible to decentralize as a trick here is crypto economics whenever we talk about any crypto economical systems there should be some some kind of economics behind I strongly believe the creep when they're talking about crypt economic system as the main the main focus what value are we creating there and once we find out what value agreat since then we can talk about the centralization now for for for the topic we as a sentiment as a platform focus for the last three years it's specific

until now no one has achieved to build a decentralized system how in any market or in our particular case in the cryptocurrency market can you create strategies or signals which do include proven alpha alpha in in the financial sense means you can predict the future with some some results and there are some parameters how you defined what is your value of your prediction so when is the value of your system creates in such an environment where you regularly make a prediction so good timely coverage of the market and you

wanted to do isn't decentralized it means there are different participants and they share vary between themselves it doesn't exist yet there is not a single successful project at the moment we did however achieve on our side in centralized way to extract this value of the market and there are few examples what we were able to build and for the purpose of this presentations I can show example explain a little bit when we been thinking what is the most important to understand the market in our

perspective is the most important part you understand and you can measure the risks and the risk can be in different areas and for investors of participants the biggest risk you are not a way of top has come in or crashes coming or risk of underlying protocols already there you are simply not aware of them and we've built a few models and over the last two years for the one I mentioned here we spent maybe one year to collect needed data sources to clean up some and make sure pipelines are working in the last year we've been applying it all over again and we made more than one hundred hundred sixty

around predictions and this is very quite impressive results at 76% of time we were on mining means whenever we say something it was a top and when I was a clearly stop I think it was only 12 times when we did direct my prediction so all of them were right and so we have now a kind of our self proves that we can build it in centralized way we know what data we need we know what framework can be applied and approximately know how we can measure the results but we're interested as we also since year number one or one and a half is the theorem I

remember I've been participates in def cons since Shanghai time and few people here I met already not the first time so very interested how we can do the same things achieving the predictability or risk measurement of the crypto networks in a decentralized way and this is exactly the step we are going now in our team and we are looking for the ways how we can do it and here's just an example how it looks now on our platform if you look you have to look for it because it has big value we are not literally like pushing it everywhere we believe that if values here you are we'll have to look for it and try to use it for your benefits but it's as a

Hinton and the inside you can find the market coverage and tops predictions and they look like some text at the moment when they say Samson you will see a red point and of course I took the best example the way a few times were not so good but you will see a repeating pattern when we say something it's very often pretty much unit of class - one day so the centralization of the whole process initially I was thinking that we can decentralized data connectors I did not work it's very find very difficult to find incentive or long-term

sustainable model for people who can attach a very nice important data to pay them it defines the price especially it's very difficult to define the price so this part we still believe will have to do mostly centralized and especially engineering part is very complicated if you want to build based on one chain especially on chain data some some of high scalability and low latency I don't think any single person can do it it's in our side we have team of at least four or five high skilled engineers who would take care of it so this but extracting the data doesn't seem to be very easy to decentralize so what can we do next we can leave it for a moment

no needs to decentralize data source we can still produce it as a team but what can be done and we here like going exactly now these steps we have created a frameworks how we analyze the data it's a like quite a notebook of people that assigns know how it works you extract the data you pull in all needed data points and you apply your analysis so this part already can be partially decentralized and it can be on a level where we start here in our Python notebooks with the frameworks or results can be part of the marketplace someone come out with something he can publish it and if someone want to get sexist okay we can probably establish

peer-to-peer marketplace you pay as much as you think is available results of this researcher and this is one of the way we like be pursued as the next maybe at least half a year want to try it out now there is but this you see amount of data scientists who understand how you do rigorous data analysis is not so big and to pull it to the Crypt is also not so simple there is a even simplified more more simple part of what can be done is the way we are doing now in our team where we put this inside so analysis we see the data we understand the market and we put it as a report

mostly like a market researcher this can be decentralized and this will be I think will deliver it may be well optimistically two months it means maybe four months where anyone who is part of a community or platform he gets access to the data we can teach how he can analyze or he can bring his knowledge and he can publish this insights and they already now been developed only closed access so peer to peer payment and this seems to be in the easier way if people sit in an auditorium or listen to this video later on if you believe you are in a position you can grasp complexity of the data simplify it and

make some nice crypto network analysis where you use these risks and you publish it and the system will take on our side time stamp and make sure in a mutable and we'll leave it to the market to decide for the price of your analysis and the second part can begin fight or another step more user-friendly version of prediction markets they put in this way we have not gotten any big traction for prediction markets yet even though all gorongosa's exist for quite some time but whatever is created there we hope is a trump election and over the years there will be some good traction or

people will use it but for the market or risk predictions I haven't seen any real usage it can be done in a gamified way and I will have some ideas we'll probably do it in this way to our contribution to the centralization of risk analysis yeah and just last if you feel like you could be interested on point of analysis the data or making your kryptonite networks research publishing in a way and say my scientific way beaten Python notebook or just in written analysis

please join goes the link and there you will see maybe a few more examples on this landing page and we can start building together we will share our knowledge we should get it over the two years and you can be able to produce some value and capture this value and peer-to-peer way and be rewarded from the community or users indirectly way I think that's pretty much it yep we can come to a question section is there any questions how does the sentiment token tap into this I did not mention it yeah we have

our token right depends if it's a more prediction market then it can be used as a and we have a kind of exchange for for built in exchange for your opinions then it as a exchange fee can be used in this way paid or possible collateral even though I'm like a little bit cautious as after the debacle with synthetics doesn't seem to be very good idea so most likely it's a vehicle for establishing prediction exchange and on if it's a peer-to-peer

reward I would think that we will then limit on the usage you can only get access to the system pain with the Sun token yeah and actually this type of woman to just pretty much soon because there's been so much in my opinion of course as a founder really generated a net weight form and we still now a lot of people they can pay in a Fiat and any cryptocurrency and on a Sun token you can burn it but we never limited on the only Tucson token holders only you can get exercise so we were like do it as in in the next one two months and other they experimented you know usage of the

token unless it's a very established token model like is a proof-of-work mind algorithms it's you can change it or actually you should change it regularly so the first we will try it a limit access to some parts of san token and if goes to prediction use it as a fee on exchange and then we might change it in the next two three months again as a question [Applause]