The Mito Knowledge Graph
- Carefully engineered

The core foundation of the Mito.ai data processing platform is our knowledge graph. At is core it’s an expression of how the world is connected - a graph of entities, represented as nodes, connected by relations.

By analyzing and adding millions of multilingual documents through this graph daily, it becomes an evolving and self-learning structure that, in real-time, represents what’s going on in the world.

Using this graph and understanding the hidden connections that make up the business landscape, Mito is able to know how relevant and significant each event is for any company, product or industry.

Former New York Attorney General
instance of
position
subclass of
state attorney general
Relevance
0.00139
Offset
0.98480
Mention
0.01538
Eric Schneiderman
country of citizenship
United States of America
educated at
Amherst College
Harvard Law School
position held
member of the State Senate of New York
place of birth
New York City
Facebook profile ID
SchneidermanNY
Twitter handle
@Schneiderman
Relevance
0.01217
Offset
0.97960
Mention
0.03077
alleged Exxon
chief executive officer
Darren Woods
country
United States of America
nics
Integrated oil Companies
headquarters location
Irving
stock exchange
New York Stock Exchange
ticker symbol
XOM
Relevance
0.01165
Offset
1.00000
Mention
0.23077
withheld information about its internal climate change
instance of
concept
Relevance
0.02138
Offset
0.99068
Mention
0.10769
discussions and misled the public about what it knew. Exxon
chief executive officer
Darren Woods
country
United States of America
nics
Integrated oil Companies
headquarters location
Irving
stock exchange
New York Stock Exchange
ticker symbol
XOM
Relevance
0.01165
Offset
1.00000
Mention
0.23077
has long denied the charges. Exxon
chief executive officer
Darren Woods
country
United States of America
nics
Integrated oil Companies
headquarters location
Irving
stock exchange
New York Stock Exchange
ticker symbol
XOM
Relevance
0.01165
Offset
1.00000
Mention
0.23077
CEO
instance_of
corporate title
legal concept
profession
said to be the same as
director
Relevance
0.01491
Offset
0.99880
Mention
0.04615
Darren Woods
country of citizenship
United States of America
educated at
Kellogg School of Management
Texas A&M University
employer
ExxonMobile
place of birth
Wichita
occupation
chief executive officer
Relevance
0.00142
Offset
0.94240
Mention
0.03077
said he has not reached out to Barbara Underwood, who was appointed to replace Schneiderman.
Story impact: 0.68609

Extreme self-learning capabilities

Mito’s Context Relevance system uses state-of-the-art machine learning to assess the relevance and impact of each event. This usually requires massive amounts of manually generated training data for it to work properly.

Mito—on the other hand—has proven its ability to create customized machine learning models in a matter of days. By using our proprietary knowledge graph and a world-class entity linking system Mito is able to leverage its self-learning capabilities—even with only limited access to training data.

This means you don’t have to waste time sifting through unnecessary information, get only what matters, and what your business can use.

Technology that scales
with the speed of your company

Mito uses state-of-the-art technologies that can be combined and extended in extraordinary ways. The workflows of our tech team facilitate extreme agility and round the clock monitoring ensures high availability. What’s more its scalable. Our adaptable system is able to meet both current and future demands of our customers.
  • Entity Linking

    Our world-class language agnostic Entity Linking delivers structure, clarity, and insight to the constant stream of data. Mito currently covers more than 18 million different meaningful entities, such as people, places, companies, and even abstract concepts.

  • Clustering

    Duplicate articles covering a single story is a thing of the past. By performing real-time clustering on all ingested content, Mito is capable of identifying important events as they happen.

  • Categorization

    A combination of state-of-the-art content categorization techniques and our sophisticated Knowledge Graph makes Mito capable of classifying textual content. It works on pre-defined categories like finance and sports, as well as previously unknown content categories.

  • Sentiment analysis

    We fuel our Sentiment Analysis service with world-class Entity Linking to provide the highest possible granularity.

  • Summarization

    Tired of reading through stacks of long documents to get the essence? Mito.ai provides summarization. And as the rest of our platform, it works across language barriers.

  • Flexible data sources

    We are constantly adding more data sources to our system improving our analysis constantly.