Page 8 - SPi Global Whitepaper_Transform, Content to Actionable Data and Unlock New Business Opportunities Through AI

8 Transform, Content to Actionable Data and Unlock New Business Opportunities Through AI Transform, Content to Actionable Data and Unlock New Business Opportunities Through AI 9

Identifying Concepts

to Improve Database

Search & Discovery

For several years, SPi Global has provided In 2018, SPi Global augmented this process
Concept Identification services to scientific with a machine learning solution. “We created
and medical information business. Historically, a supervised model to predict the concepts
these publishers would rely on a team of based on the document content,” says Venky.
highly qualified subject matter experts to
read research, identify core concepts within The model is built on all of the research
that research, and link to related concepts in SPi Global’s subject matter experts have
other content. analyzed and tagged in the past. Those
millions of data points teach the machine
“Let’s say we extract from a research article learning model how to recognize and extract
a certain disease. We’ll also extract the the most important concepts within a
symptoms associated with that disease, research article. The technology is calibrated
the medication for that disease, the so that it can distinguish between concepts The Concept Identification tool is improving the quality and relevancy of concepts, often
chemical compounds of that medication,” and recognize which concepts are most identifying concepts that subject matter experts might have missed. That’s improving
explains Jishnu. “All of this data is organized valuable and relevant to the researcher. the taxonomy tree, making search and discovery within medical databases more relevant
and ingested into a taxonomy tree. and valuable for professionals. It also adds a level of consistency and reduces deviation,
That taxonomy tree powers search and which is inherent when SMEs select concepts.
discovery within a database so that medical
or pharmaceutical professionals can find the The application is also increasing the volume of research SPi Global can analyze for
most relevant information.” its clients, as well as the rate at which it can develop taxonomy trees. For example, one
client has increased the volume of research it analyzes through SPi Global by 50% in
“Let’s say on average there are about 20 important the last year.

concepts within an article,” says Jishnu. “Today, the In the near future, SPI Global plans to employ deep learning technology to automate

machine is able to extract 16 out of 20 of those concepts Concept Identification even further, says Venky. With these advancements, SPi Global
automatically. It’s achieved 80% automation, meaning our will effectively eliminate the need to identify concepts manually and subject matter
experts will simply quality check the concept recommendations.
subject matter experts can spend the majority of their time
validating the machine’s work, as opposed to identifying

concepts themselves and going through the entire article.”

© SPi Global Content Solutions © SPi Global Content Solutions