ourTell
ourTell
Token: OTL


Sentiment Intelligence For Cryptocurrency Trading

ICO dates
Start date: 2018-04-01
End date:

Registrated in: Korea

Platform: Ethereum
Type: ERC20

PREMIUM ICO

ourTell categories
Artificial Intelligence Big Data Cryptocurrency Investment
ourTell token sale
KYC passing required Yes | Whitelist Yes | Restriction for countries USA, China
Hard cap 28,000,000 USD
Tokens for sale 600,000,000
Token distribution in ICO
80%
80%
Price 1 OTL = 0.047 USD
Acceppting ETH
ourTell news, social
ourTell search trends in Google
Random whitepaper excerpts

Contents
1. Introduction to ourTell......................................... 3
1.1.1 Overview ................................................... 3
1.2 Current Use .................................................. 4
1.2.1 Sentiment Analysts ....................................... 4
1.2.2 Social Media Channels .................................... 5
1.2.3 Platform ................................................. 5
2. Sentiment Intelligence Prediction Platform...................... 6
2.1 Sentiment Analysis ........................................... 6
2.1.1 Telegram ................................................. 6
2.2 Token/Project Data ........................................... 7
2.2.1 Telegram ................................................. 7
2.3 Historical and Current Market Data .......................... 13
2.3.1 Bull/Bear ............................................... 13
2.3.2 Trend/Technical Analysis ................................ 14
2.3.3 Volume Analysis ..........

1. Introduction to ourTell
1.1.1 Overview
We provide a next generation, decentralized, iterative sentiment
analysis for optimal cryptocurrency trading.
We use a system called Sentiment Intelligence which is a combination
of sentiment analysis from real certified analysts, big data and
machine learning in order to create an artificial intelligence
platform which allows for market predications with near perfect
information.
Financial markets are turbulent and require a huge vat of knowledge
to even begin trying to predict outcomes. This is the case in the
cryptocurrency market and is often even more severe due to the
casual and uninformed nature of many traders.
Markets are moved in the cryptocurrency and blockchain sphere
largely due to public perception. This is often left unknown in
analysis and predictions. With a few common issues arising,
sentiment data is:

Severely under sourced

Without context

Misinterpreted ...

1.2 Current Use
1.2.1 Sentiment Analysts
In February 2017, 50 sentiment analysts were onboarded. Analysts had
to have obtained a Bachelor’s degree at a minimum in a related field
in order to be eligible. From there, analysts underwent a 5 week
intensive course and a written examination in order to qualify them
to be a sentiment analyst on the platform.
As of April 2017 we have on boarded over 200 analysts in total,
under the same rigorous conditions.
Currently, we have 230+ sentiment analysts working on the platform.
We aim to have onboarded over 1000 sentiment analysts by the end of
2018 and remunerated them through our Proof of Sentiment protocol
(discussed later in the paper).
Analysts work on fixed data pieces in order to enable optimizations
and efficiency in review time. Each piece of data is reviewed by
four analysts:
1. Working solely on the fixed data set
2. Working on data set and competitors
3. Working on unrelated data sets
4. Working o...

1.2.2 Social Media Channels
While we are optimizing our initial offering of the platform
Telegram has been the sole source of sentiment data. We subscribe to
the philosophy that
“The future belongs to those who learn more skills
and co
mbine them in creative ways.”
(Greene, 2012, p. 44). We believe
in the merit of optimizing our current systems and ensure we derive the
greatest benefit from them before moving forward.
Once we are happy and can sustain our Telegram sentiment analysis we
will be moving onto other platforms. Namely:
1. Discord
2. Reddit
3. BitcoinTalk
4. 4Chan
5. Steemit (comments+upvotes)
6. Medium (comments+upvotes)
7. Twitter
8. Facebook
1.2.3 Platform
Our predictions platform is currently operational and in beta
testing. It combines all facets that could affect market outcomes
and is able to make complete predictions with near perfect
information. The prediction platform has been used for pre...

2. Sentiment Intelligence
Prediction Platform
2.1 Sentiment Analysis
Throughout our research and development we have encountered numerous
forms of sentiment analysis methodologies and automated systems. All
methods bearing their own merits and shortfalls as discussed in
1.2.1.
We continuously aim to improve and optimize our system and give the
most contextual and complete analysis possible.
2.1.1 Telegram
2.1.1.1 Users
Users on Telegram form the foundation for where all sentiment comes
from. Naturally, this means that user analysis is paramount in order
to understand what they are saying and determine the efficacy of
their information.
2.1.1.1.2 History
Messages are analyzed from our sentiment analysts and passed into
our machine learning model. This model keeps a permanent and running
score for each user and ranks the user based on:

Participation
: How wildly they participate in Telegram
communities. Users that contribute m...


Vested Interest
: Does the user have a stake in the sentiment
they are discussing. I.e. does the user own ExampleCoin when
discussing it. This is only ascertainable if the user has
directly mentioned it. Typically, users with a vested interest
are less authentic.

Influence
: How much influence does the user have in the
community. Users have varying influence and can move
opinions/spark conversations more easily than others.
Typically, these users have a higher value than those with low
influence.
2.1.1.1.2 Messages
Messages are where we derive our actual sentiment data that forms
the core of our sentiment analysis. While being integral, it has
been kept as simple as possible to optimize the process. Working
with all other metrics and sentiment metadata such as user data
allow for a contextual and complete analysis.

Sentiment
: Does the message portray a positive, negative or
neutral sentiment. This is ranked...


Size
: How large is the Telegram group. The size of the groups
gives indications of the interest levels in the community.

Age/Growth:
We analyze how organically a group has grown based
on its age and events that cause unexpected growth.

Bought/Fake Users:
We have analyzed hundreds of groups and
have models to determine whether a group has fake user and
bots. These methods included growth curve analysis and group
activity/responsiveness.

Activity:
How much activity is there in the group in
proportion to their users and other groups of similar size.
Typically, this is indicative of how anticipated the ICO is.
This metric is useless on its own. It needs to be combined
with the sentiment that makes up the activity.

Responsiveness and communication:
How quickly do
administrators respond to questions and is there often a state
of “not knowing” in the
group. Typically, the mo...

Will unsold tokens be burnt in order to reduce total supply.
2.2.1.1.4.4 Lockup
How long tokens are locked up from the crowd sale before
transferable.
2.2.1.1.4.5 Discount/Bonus
What was the discount/bonus for the ICO and/or presale. A higher
bonus reflects poorly as there is selling pressure once the token
lists.
2.2.1.1.4.5 Sale Structure
We have analyzed the results of numerous sale structures such as
First Come, First Serve, tiered rounds, etc. There is clear
correlation on the outcomes for each structure.
2.2.1.1.4.6 Raising Amount
How much is the ICO trying to raise compared to the current market
standard and similar projects.
2.2.1.1.4.6 Openness
What restrictions/limitations were there for investors to
participate in the sale. The more open a sale is the less likely it
has for potential gain as the buying pressure is low.
2.2.1.2 Qualitative
Not all measures and metrics that form part of an ICO are purely
quantitative. For quali...

ourTell Roadmap

1
July 2016
Product Idea Initial idea first conceived
2
Nov 2016 Scraping
Basic data capture and scraping implemented across exchanges and Telegram
3
February 2017 Sentiment Analysts
50 Analysts added to review data after passing examination to determine qualification
4
March 2017 Analysts Platform
Beta platform for analysts to review data in an optimised and efficient fashion
5
April 2017 Additional Analysts
200+ analysts on the platform
6
May 2017 ML/AI
Machine Learning implemented with Sentiment Analysis for predictions
7
July 2017 Fund Strategy Testing
Implemented tests to run over the next year and determine best fund strategy and returns
8
April 2018 Token Sale
Public and private token sale
9
May 2018 Prediction Platform
Open the prediction platform to OTL holders
10
July 2018 Fund Investment
Fund investment open to OTL holders
11
Q4 2018 Open Fund Investment
Allow externals to invest into the Sentiment Intelligence fund
Luke Kim
Luke Kim CEO, Co-Founder

Mike Yoo
Mike Yoo CTO, Co-Founder

Luqman Suleman
Luqman Suleman Technical Lead

Michael Kim
Michael Kim Full-stack Developer

Richard Goh
Richard Goh Full-stack Developer

Jake Harrington
Jake Harrington Backend Developer

Vincent Park
Vincent Park Data Scientist

Jackie Gang
Jackie Gang Data Scientist

Harry Lim
Harry Lim Data Scientist

Paul Kim
Paul Kim Data/Trade Analyst

Jack Hahn
Jack Hahn Marketing Manager

Mo Ahmadi
Advisor Mo Ahmadi Tech Evangelist, Capital Expert

Virat Naidoo
Advisor Virat Naidoo Blockchain Investor, Growth Specialist

Khaleel Osman
Advisor Khaleel Osman Venture Capital, AI/ML Investor

Michael Kim
Michael Kim
Full-stack Developer
CEO and Founder of CoinInside
Product Development
CTO, Co-Founder
CTO, Co-Founder
Business Development Advisor
CEO & Founder of CoinInside
CEO & Founder of CoinInside Former executive of EA, Microsoft, Blizzard, Wargaming, and Havok
CTO, Co-Founder
GAMING ADVISOR
Paul Kim
Paul Kim
Data/Trade Analyst