NetBase: Natural Language Processing to Decipher Customers’ Social Media Sentiment

by   |   June 8, 2012 10:48 am   |   0 Comments

At the start of a recent lunch hour, Twitter was abuzz with  #ThingsWeDo2LoseWeight. Some 20 posts per minute touted crazy diets, suggested questionable forms of exercise and joked about fast-food. “Walk [into] McDonalds instead of using the drive thru,” quipped @MeloMotivation.

Click on almost any Twitter hashtag or trending topic, in fact, and you’re awash in opinions about consumer products, companies and celebrities. Marketers, customer service reps and product developers want to make sense out of these references: are tweeters making fun of McDonalds, or of people on diets? They’re finding some answers with the help of text analytics software from vendors such as Mountain View, Calif.-based NetBase Solutions.

“We’re in the insight discovery business,” says NetBase CMO Lisa Joy Rosner. The company’s flagship product, the Social Intelligence Platform, was developed with the help of marketers from leading consumer packaged goods companies, including Coca-Cola and Kraft.

About the technology: NetBase uses a technology called natural language processing to understand sentences in context. Its analytics engine can tell, for example, that two statements—”I want an iPhone because it’s adorable,” and “I covet an iPhone as it is adorable”—are both positive statements about the popular smartphone, but that “I want my iPhone to show you the adorable shoes I want” isn’t related.

Natural language processing goes beyond merely recognizing patterns in a text by including “connector” words (such as conjunctions and prepositions) in its analysis, along with slang and alternative spellings and abbreviations, Rosner says. It can also distinguish when a negative word is part of a positive statement or used in a positive way. These capabilities earned NetBase recognition from market research company IDC as a company to watch in 2010.

Customers identify “topics” they want to analyze, similar to how business intelligence users might define a complex query. (A topic might include a company or product name and its common varioations.) An analysis and reporting tool, Insight Workbench provides templates—which users can customize–for analyzing data and for configuring charts and word clouds to visualize the results.

NetBase represents one of two approaches to social analytics, says Susan Etlinger, industry analyst with Altimeter Group  (Etlinger consults with NetBase and other vendors). Companies such as Lithium and Radian6 (which is owned by Salesforce.com), “provide an easy-to-use dashboard to multiple users,” she says. On the other hand, companies like NetBase and Crimson Hexagon offer “the ability to delve deeply into the voice of the customer,” and “tend to have significant IP in text analytics.”

Use case: Taco Bell was able to uncover that customers “crave” its crunch burrito, and also tease out broader insights related to the item and to the chain’s new Cantina menu. In the video below, NetBase delves into social media chatter about five grocery stores to assess how passionately consumers feel about those brands and why. According to the analysis, consumers talked the most about Wal-Mart, but “hate” it more than the other brands, partly due to “poor treatment” of employees and customers. They loved Costco the most, for its “quality, quantity and discount prices.”

Companies can use such information, Rosner says, to alert their call centers about possible questions or complaints, plan promotions or make pricing and production decisions.

About the company: Jonathan Spier and Michael Osofsky, both engineers at Ariba, founded the company, which they called Accelovation, in 2004 to market search software that would crawl the web for statements identifying a need for a new product or service. The company changed its name to NetBase in 2008.

Early customers used it to identify markets for new products and solutions to research problems. Under an agreement with publisher Reed Elsevier, the technology, branded as Illumin8, was also used to comb peer reviewed journals and patents. “It was sitting on the desktop of research scientists of many large CPG companies,” says Rosner. A marketing executive at one of the companies saw it and realized it could be used to cull insights from social media.

The company receives funding from Thomvest Ventures, and from Altos Ventures, where Spier is now entrepreneur in residence. Osofsky remains on NetBase’s leadership team as co-founder.

Elana Varon is a freelance writer based in the Boston area. Follow her on Twitter at @elanavaron.

Correction: The original version of this story has been corrected to reflect that NetBase was founded in 2004, not 2003.







Tags: ,

Post a Comment

Your email is never published nor shared. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>