KEY PROJECT DATA
- Name of Project : Datapeek
- Team Website : https://datapeek.org
- Proposal Wallet Address : 0xF40b005FFE2Db0197b8c301e1C966C2cb3B59A08
- Current Country of Residence : France
- Contact Email : email@example.com
- Twitter Handle : @datapeek
- Discord Handle : kemur#7399
Which Category best describe your project :
- [x] Increase Awareness
- [x] Unleash data
- Funding Amount : $17,600
- Current Remaining Grant Treasury Balance :0
- Have you Previously received an OceanDAO Grant ? No
PROPOSAL IN ONE SENTENCE
Increase public interest for datasets by linking them to relevant social media comments.
We are bulding a platform where it will be easy to find data
that confirm or contradict arguments and reasoning found on the web, on many topics.
We think it’s a natural way to make Ocean protocol and the data economy part of the internet day-to-day use.
WHAT PROBLEM IS YOUR PROJECT SOLVING
Our main goal is to improve discoverability of datasets.
Datasets are the raw materials for analysts to make decisions.
But only a small subset of professionals, can spend days
thinking about what they could do with one in particular.
Most people, outside of a few use cases, cannot guess
the usefulness of a new dataset nor its value.
What we mean to do is to build bridges between datasets and social media.
It will add a semantic layer around datasets and makes it easier of an analyst
to imagine what she could do with it.
An interesting part of what happens on social media ( twitter, reddit, hacker news ) ,
are debates on various topics. Some of the arguments made in those debates
are quantitative and can be backed by data.
In particular debates trying to assess the value of newly created cryptoassets are often quantitative in their nature but the needed data is not trivial to guess, that is part of the usefulness of a reasonning engine.
We will foster the addition of data to online debates by making a platform to fact-check
with data. It will works in two step. The first step is to identify weakly backed
arguments and to request a link to data analysis to strongarm them.
In a second step we will add relevant datasets, every time we can find one on ocean marketplace. To do that we have to develop a set of algorthms that use the description of a dataset to find matches with debates and arguments identified on the platform.
Akin to finding the nearest conversation in a meaning graph.
Here is an example of how a user can get introduced to a new dataset by reading an answer to a request:
HOW DOES THE PROJECT DRIVE VALUE FOR OCEAN PROTOCOL
The value we will to drive to the ocean protocol ecosystem comes in multiple stages:
By building the habit of checking the data on anything that get argued online,
we increase the total market of consumers for the datasets hosted on Ocean Protocol
and entice potential data providers to put datasets on the marketplace. For example a
data analyst will be able to build up her profile and reputation by consuming
datasets to produce analysis. As a return on her investment, it can lead to job opportunities.
We will strive make Ocean Protocol a Data Layer for online conversations.
The advantage of having Ocean Protocol as a reference layer for online debate is to
improve both reproductibility and trust in figures thrown around in public debates.
It increases the usefulness of Ocean Protocol to the internet as a whole,
and that will get translated in its token price.
If we take datasets and their relevant commentary in social media,
what we build is a kind of “social network for datasets”.
It gives a better marketing profile to datasets. That in turn is an incentive for potential data providers to use their data as a new marketing channel. Later, the social footprint of a dataset
will become a signal used to price datatoken.
BANG FOR BUCK
We will offer an experience superficially similar to the one on stackoverflow, to have a smooth onboarding.
We aim to attract the fraction of their audience made of people who are more quantitative minded and to make them familliar with running computation on datasets. They will go on datapeek to find answers to quantitative questions about the state of the world and be encouraged to run an analysis by themselves whenever a dataset on matching that question is available.
About 50 millions people visit stackoverflow in a month. Around 20 millions of them are
thought to be developpers . We hope to bring 0.1% of them to view analysing datasets on ocean protocol as part of their thinking toolset. That’s about 20 000 developpers
with a quantitative bent.
Of those 20 000 in the target audience, We estimate about 1% will become active investors.
with a mean value of 2000 $OCEAN. It translates to around 400 000 $OCEAN .
We are anticipating some staking behavior and potential datasets coming from that audience
and are still evaluating how much we can expect.
- BANG: 400 000 $OCEAN
- BUCK: 40 000 $OCEAN
- BANG/BUCK Ratio: 10
PROJECT DELIVERABLES :
App will be published at: https://datapeek.org/
We have started to work
- on the core ReQuest-Answer mechanic , gathering user feedback.
- on the automatic detection of data link on social media
- on the automatic detection of quantitative arguments.
We are going to work
- on refining the previous points
- producing marketing content through interviews with data economy stakeholders.
- work on automatic suggestion of datasets.
- building the ontology and the automated reasonning architecture
- provide tools for easy publishing of simple visual analysis from datasets
- work on topical “social interest” radar for datasets. ( a preview of the first steps is available here if you want to give some feedback: https://datapeek.org/static/dataset-interest/index.html?ds_name=german-traffic-data-for-machine-learning )
- Kevin Muur, developer
- discord: kemur#7399
- twitter: https://twitter.com/datapeek