Category: Knowledge Graph

The Best Schools For Female Founders According To The Knowledge Graph

Posted on | by Merrill Cook

Upon seeing Crunchbase’s annual ranking of the best schools for graduating entrepreneurs, we wanted to see how our Knowledge Graph results stack up.

The Diffbot Knowledge Graph is sourced from crawling a majority of the web and extracting entities and facts using NLP and machine vision.

Two prominent entity types are person and organization entities. When paired together powerful observations sourced from across the web are possible. In this exploration we returned all person entities within the Knowledge Graph who are currently founders and who are female. We filtered to make sure each organization had at least some publicly disclosed funding, and then we took a look at a summary view of which schools these founders had attended. You can check out the Knowledge Graph query here with a free trial.

While the top schools for female founders were consistent with Crunchbase’s coverage, you may wonder why the numbers vary so dramatically. Crunchbase’s ranking this year was looking at 2019-2020 graduates, and Crunchbase’s data is centered around tech and startup firmographics. While Diffbot’s Knowledge Graph certainly has firmographic details on tech-centered companies, our database of organizations is much wider ranging (over 250M+ orgs at last count). This means our list includes founders of all sorts of endeavors: non-profits, artistic organizations, medical organizations, and tech companies to name a few.

Continue reading →

Monitoring Large Food Retailer Investments With The Knowledge Graph

Posted on | by Merrill Cook

A few weeks ago we published a view into Big Tech investments by industry. In this post we’ll take a similar look at the largest food retailers.

Panning out a bit, there are over 250M organizations within the Knowledge Graph. To obtain this list of large food retailers we first narrowed our search to food retailers with more than 1,000 employees. This query surfaces more than 7,000 fact-rich entities.

From there we simply sorted the results by number of employees to gain the largest food retailers including Walmart, Target, Tesco, Kroger, Carrefour, and Safeway.

With this list in mind, we looked for a list of organizations who had been invested in by one of these organizations. Bounded by calendar years, we then returned a summary view that looked at which industries the invested-in companies represented. If you have a subscription or free trial feel free to check out the resulting query.
Continue reading →

Startup Revenue By County With Diffbot’s Knowledge Graph

Posted on | by Merrill Cook

What can you do with billions of web-sourced facts on hundreds of millions of organizations? Beyond analyzing the facts themselves, you (or a machine of your choice) can learn a lot. Historically, our Knowledge Graph has had one of the largest collections of publicly-disclosed organization revenue. Recently, we’ve applied machine learning processes across many org fields to estimate revenue for private organizations as well.

Continue reading →

Using the Knowledge Graph to Segment Big Tech Investments By Industry

Posted on | by Merrill Cook

Every big tech investment is big news. If your firm raises a funding round with prestigious investors or is acquired, you better bet you’ll spread the news far and wide.

But where can you go for this information en masse? Even covering a handful of big investors over a handful of years can lead to a list of thousands of invested in firms. And a list of firms themselves isn’t that useful. Sure, some big names pop out. But how do you see what “plays” big tech is making?

That’s where our web-reading bots come in. By working through billions of web pages using NLP and machine vision, Diffbot’s Knowledge Graph is the largest public-web sourced database of organizations, articles, people, products, and events. For each entity — organization, articles, people, etc. — facts are vetted and accumulated to create a filterable, searchable database of “things.” So when we wanted to check out which industries big tech has invested in over the last decade, we knew right where to turn. No analyst middlepersons, just public web data structured into a market intel-rich format.

Big Tech Investment By Industry 2010-2021

Distribution of industries of organizations invested in by Facebook, Alphabet, Amazon, Microsoft, Apple, and Netflix from 2010 to July 2021. Firmographic data sourced from Diffbot’s Knowledge Graph.
Continue reading →

Every Company That Sells Organization Data is Biased

Posted on | by Jerome Choo

Yes, even the biggest leaders in market intelligence. Even us.

Some focus solely on startups. Some only on venture-backed companies. But you probably wouldn’t even know. Because most won’t (or can’t) tell you what their data is biased towards! 🤭

“We have over 10M companies in our database!” is a meaningless statement if you can’t tell whether the data is a representative sample of Indian restaurants in the world, or perhaps more realistically, what they just happened to scrape.

Unless we’re talking at least 200M+ unique organizations strong, you’re looking at a biased dataset. And that’s still a conservative minimum.

This is common knowledge for data buyers, who make up for the lack of a known bias by evaluating datasets for known, easily verifiable data, like the Fortune 1000.

Given enough evaluation feedback cycles, most organization data brokers end up biased towards the Fortune 1000.

If your target is enterprise b2b, you’re in luck. You can find that data anywhere. Just check your spam folder.

If it’s anything even remotely more niched, like rubber gasket manufacturers or global non-profits focused on relieving poverty, you’re probably scraping this data yourself off a conference site.

And if your market intelligence application needs the closest thing to a truly representative sample of global organizations, it might seem impossible.

For data brokers, it just doesn’t make any sense to boil the ocean. It’s cheaper and easier to focus data entry resources on a few markets and whatever coverage gap feedback they get from lost deals.

Even if they did manage to compile all the companies on Earth, they would have to do it over and over again to keep their records fresh.

It’s an absurd and impractical human labor cost to maintain. So no one employs hundreds of people just to enter org data. Not even us.

We employ machines instead, which crawl millions of publicly accessible websites, interpret raw text into data autonomously, and structure each detail into facts on every organization known to the public web.

Which, as it turns out, is our known bias.

Generating B2B Sales Leads With Diffbot’s Knowledge Graph

Posted on | by Merrill Cook

Generation of leads is the single largest challenge for up to 85% of B2B marketers.

Simultaneously, marketing and sales dashboards are filled with ever more data. There are more ways to get in front of a potential lead than ever before. And nearly every org of interest has a digital footprint.

So what’s the deal? 🤔

Firmographic, demographic, technographic (components of quality market segmentation) data are spread across the web. And even once they’re pulled into our workflows they’re often siloed, still only semi-structured, or otherwise disconnected. Data brokers provide data that gets stale more quickly than quality curated web sources.

But the fact persists, all the lead generation data you typically need is spread across the public web.

You just needs someone (or something 🤖) to find, read, and structure this data.

Continue reading →

Towards A Public Web Infused Dashboard For Market Intel, News Monitoring, and Lead Gen [Whitepaper]

Posted on | by Merrill Cook

It took Google knowledge panels one month and twenty days to update following the inception of a new CEO at Citi, a F100 company. In Diffbot’s Knowledge Graph, a new fact was logged within the week, with zero human intervention and sourced from the public web.

The CEO change at Citi was announced in September 2020, highlighting the reliance on manual updates to underlying Wiki entities.

In many studies data teams report spending 25-30% of their time cleaning, labelling, and gathering data sets [1]. While the number 80% is at times bandied about, an exact percentage will depend on the team and is to some degree moot. What we know for sure is that data teams and knowledge workers generally spend a noteworthy amount of their time procuring data points that are available on the public web.

The issues at play here are that the public web is our largest — and overall — most reliable source of many types of valuable information. This includes information on organizations, employees, news mentions, sentiment, products, and other “things.”

Simultaneously, large swaths of the web aren’t structured for business and analytical purposes. Of the few organizations that crawl and structure the web, most resulting products aren’t meant for anything more than casual consumption, and rely heavily on human input. Sure, there are millions of knowledge panel results. But without the full extent of underlying data (or skirting TOS), they just aren’t meant to be part of a data pipeline [2].

With that said, there’s still a world of valuable data on the public web.

At Diffbot we’ve harnessed this public web data using web crawling, machine vision, and natural language understanding to build the world’s largest commercially-available Knowledge Graph. For more custom needs, we harness our automatic extraction APIs pointed at specific domains, or our natural language processing API in tandem with the KG.

In this paper we’re going to share how organizations of all sizes are utilizing our structured public web data from a selection of sites of interest, entire web crawls, or in tandem with additional natural language processing to build impactful and insightful dashboards par excellence.

Note: you can replace “dashboard” here with any decision-enabling or trend-surfacing software. For many this takes place in a dashboard. But that’s really just a visual representation of what can occur in a spreadsheet, or a Python notebook, or even a printed report.

Continue reading →

How Employbl Saved 250 Hours Building Their Career-Matching Database

Posted on | by Merrill Cook

We started with about 1,000 companies in the Employbl database, mostly in the Bay Area. Now with Diffbot we can expand to other cities and add thousands of additional companies. 

Connor Leech – CEO @Employbl

Fixing tech starts with hiring. And fixing hiring is an information problem. That’s what Connor Leech, cofounder and CEO at Employbl discovered when creating a new talent marketplace meant to connect tech employees with the information-rich hiring marketplace they deserve.

Tech job seekers rely on a range of metrics to gauge the opportunity and stability of a potential employer.

While information like funding rounds, founders, team size, industry, and investors are often public, it can be hard to grab the myriad fields candidates value in a up-to-date format from around the web.

These difficulties are amplified by the fact that many tech startups are often “long tail” entities that also regularly change.

Continue reading →

Context Matters, Tracking Quote Spread Across The Web In A Historic Year

Posted on | by Merrill Cook

Hindsight is 20/20. And as we usher in a new president in what has been one of the most tumultuous years in American history, we can begin to see clarity about the forces that moved throughout our jobs, our lives, and our collective imagination.

Another way to put this is that over time we tend to have more context.

Within Diffbot’s Knowledge Graph, one unique lens through which we can leverage the context of semantic data is by looking at the speakers of quotes.

When our AI reads articles it pulls out quotes, and when it can it attributes a speaker to these quotes. As our crawlers traverse the entirety of the public web, sources of quotes are validated and over time some quotes circulate more than others.

When performing a facet search, this lets us essentially show something like a retweet count for the entire web. This answers questions like whose voices are being heard? And what speakers are the most widely cited in a given topic?

To commemorate the end of an era, let’s take a look at a few of the most circulated statements of the last 365 days.

What were the 10 most circulated quotes across the web by President Joe Biden in the last 365 days?

Continue reading →

From Knowledge Graphs to Knowledge Workflows

Posted on | by Mike Tung

2020 was undeniably the “Year of the Knowledge Graph.”

2020 was the year that Gartner put Knowledge Graphs at the peak of its hype cycle.

It was the year where 10% of the papers published at EMNLP referenced “knowledge” in their titles.

It was the year over 1000 engineers, enterprise users, and academics came together to talk about Knowledge Graphs at the 2nd Knowledge Graph Conference.

There are good reasons for this grass-roots trend, as it isn’t any one company that is pushing this trend (ahem, I’m looking at you, Cognitive Computing), but rather a broad coalition of academics, industry vertical practitioners, and enterprise users that generally deal with building intelligent information systems.

Knowledge graphs represent the best of how we hope the “next step” of AI looks like: intelligent systems that aren’t black boxes, but are explainable, that are grounded in the same real-world entities as us humans, and are able to exchange knowledge with us with precise common vocabularies. It’s no coinincidence that in the same year that marked the peak of the deep learning revolution (2012), Google introduced the Google Knowledge Graph as a way to provide interpretability to its otherwise opaque search ranking algorithms.

The Risk Of Hype: Touted Benefits Don’t Materialize

Continue reading →