Powered by Machine Learning, PitchBook Suggestions Helps Clients Discover New Companies
June 20, 2018
New Feature Automatically Refines Search Results so Users Can Discover New Companies and Capitalize on Investment Opportunities
SEATTLE, NEW YORK, SAN FRANCISCO, LONDON – June 20, 2018 – PitchBook, the premier data provider for the private and public equity markets, today launched PitchBook Suggestions, its newest feature that recommends relevant companies to users during advanced searches in PitchBook Desktop. Leveraging the new tool, PitchBook users will now be able to automatically add relevant companies recommended by PitchBook’s Suggestions algorithm directly to search results, without modifying search criteria. As a result, clients are empowered to create the most complete and up-to-date analysis while researching a specific industry, creating a comp set, analyzing trends, sourcing investments or keeping tabs on the competition. PitchBook has invested heavily in machine learning and large-scale cloud computing technology to gather hard-to-find financial data at scale. Recent enhancements specifically target PitchBook’s web mining tools and automated news collection technology, which can now process 30 billion words – a 1000% increase from 2016. These enhancements strengthen PitchBook’s word embedding and word vector models that power PitchBook Suggestions and the high-quality comparable companies it suggests to users. “The entrepreneurial ecosystem in the US is thriving as evidenced by the healthy pace of venture capital investment. With larger sums of funding being deployed into niche and emerging industries—VCs deployed more than 28 billion overall last quarter—market dynamics are constantly shifting as winners and losers emerge,” said John Gabbert, founder and CEO of PitchBook. “For investors tracking industry activity, PitchBook Suggestions eliminates inefficiencies tied to identifying all the companies in an emerging industry and ensures users obtain an accurate understanding of a rapidly changing space.”
After running an advanced Companies & Deals search, PitchBook Suggestions will recommend related companies that were inadvertently filtered out by initial search criteria and with a single click, users can easily add companies directly in their search results without further modification of their search criteria. Behind the scenes, the algorithm that powers Suggestions starts with an intersection of machine learning technology and natural language processing. All descriptive text within the Platform is encoded into vectors to create PitchBook’s word embedding model. During an advanced search, the Suggestions algorithm encodes the entire search as a vector and compares it against the larger word embedding model to locate similar companies. For users, this removes any guess work with executing advanced searches in the PitchBook Platform and ensures the most up-to-date snapshot of a specific market. PitchBook Suggestions is the first in a series of machine learning powered discovery tools PitchBook is developing to meet evolving needs of today’s investment professionals.
For more information about PitchBook Suggestions and its machine learning enhancements, click here.
About PitchBook PitchBook is a financial data and software company that provides transparency into the capital markets to help professionals discover and execute opportunities with confidence and efficiency. PitchBook collects and analyzes detailed data on the entire venture capital, private equity and M&A landscape—including public and private companies, investors, funds, investments, exits and people. The company’s data and analysis are available through the PitchBook Platform, industry news and in-depth reports. Founded in 2007, PitchBook has offices in Seattle, San Francisco, New York and London and serves nearly 16,000 professionals around the world. In 2016, Morningstar acquired PitchBook, which now operates as an independent subsidiary.
PitchBook Press Contact Bailey Fox PR Group Manager firstname.lastname@example.org +1 206.823.3022