• October 14, 2016
  • By Leonard KlieEditor, CRM magazine and SmarCustomerService.com

Lexalytics Releases Salience 6.2

Lexalytics, a provider of text analytics, has bolstered the machine learning (ML) capabilities of its Salience text analytics platform. Salience 6.2 also now enables companies to more accurately analyze email communications and text that includes the latest emojis.

"Machine learning is driving everything, everywhere today," says Jeff Catlin, CEO of Lexalytics. "We've been doing a lot of work with machine learning for a long time, and this release sets us up to do even more with it."

Salience 6.2 features improved machine learning algorithms and grammars, but also includes the new HSDTrainer, a tool to train sentiment analysis, combining machine learning and natural language processing (NLP) technology. The Lexalytics HSDTrainer can consume any text corpus that has been marked up for sentiment and return a list of phrases and suggested scores for that text corpus. 

The HSDTrainer can also suggest new words and phrases for companies to add to their dictionaries, Catlin points out, noting that some words or phrases can have different meanings or invoke different reactions depending on the domain or industry in which they are used.

For companies that use Salience 6.2, it's all about helping them "understand who's talking and what they're talking about and then linking the sentiment to them," Catlin says.

Salience 6.2’s Emoji Analytics has also been updated to include the latest emojis released in Unicode 9.0.

"We update our systems every year to take in all the new approved and recognized emojis," Catlin says.

Emojis, he adds, are a great source of customer sentiment. "They typically have a very strong and obvious emotional component, and different grades will score higher," he says.

Salience 6.2 also includes improvements in email processing, enabling systems to ingest email databases while stripping out headers and footers, eliminating duplicate emails, and analyzing email threads. Also included is improved named entity recognition, combining machine learning and known lists of people, places, and things. With these improvements, Lexalytics has increased its precision and recall scores, known as F1 scores, by up to 25 percent. The product has also been upgraded to better recognize people and place names from much of Asia, particularly China, Japan, and Korea. This follows Lexalytics's strong growth in that part of the world.

Overall, the greatest strength of Salience 6.2 is that it will allow users "to do plenty of stuff that they might not have been able to do before, and machine learning has been the groundwork for a lot of that," Catlin says.

"With Salience 6.2, we are dramatically bolstering our ML capabilities, making it easier for [companies] to teach and tune the software to meet their unique needs," Catlin adds. "Ultimately we're improving the way machines and humans interact by expanding the capabilities of what text analytics and NLP can accomplish.

"We're putting a tremendous amount of effort into machine learning," he adds. "We’ll be putting all of our eggs in the machine learning basket for the next year or so."

Lexalytics's last update to Salience, version 6.1, was released in June 2015. 

CRM Covers
Free
for qualified subscribers
Subscribe Now Current Issue Past Issues

Related Articles

Lexalytics Releases Text Analytics for Airlines

Lexalytics' Airlines Industry Pack is honed to meet the specific needs of customer experience management and social media marketing professionals in the airlines industry.

Lexalytics Brings Text and Sentiment Analytics to Mac

Lexalytics has released a version of its Semantria platform for Mac desktop computers.

Lexalytics Launches Pharmaceutical Industry Pack

Lexalytics' Pharmaceutical Industry Pack provides out-of-the-box text analytics for pharmaceutical pros.

Lexalytics Moves Semantria to Microsoft's Azure Cloud Platform

Lexalytics' Semantria text analytics platform can now be deployed on Microsoft Azure.