You are here:  Home  >  Data Education  >  Current Article

Sentiment Analysis Spotlight: New Symposium Explores Consumer, Business Usage

By   /  April 5, 2010  /  No Comments

sethgrimes.jpg The Sentiment Analysis Symposium will debut next week In New York, under the direction of advanced analytics consultancy Alta Plana founder Seth Grimes. The new conference speaks to the growing interest in the sentiment space, which has applications on a number of fronts, Grimes says:

▏ Consumer services: Whether it’s books or music or hotels to stay at, consumers trust other consumers, as seen by the success of ratings-based recommendation engines in services such as Amazon and Netflix. The demand from consumers for peer recommendations even is playing out on Google and Bing, where search returns can include sentiment summaries from properly constructed sites. There’s a lot of possibility for giving what Grimes calls “analytical lift” to recommendation engines by folding in and assessing information drawn from free text into the opinion mix. “Often the most interesting way to offer opinions is just through narrative – and then you get into the question of volume,” he says. “Users need to be able to get at that content in a more focused way than just by surfing and browsing.”

▏ Business services: Companies can deliver business value in customer service and support as well as in managing their brands and their reputations online. Sentiment analysis has a role to play in all these cases. Think about the value that can be derived from using technology to automatically glean sentiments from the unstructured text of call center recordings or notes, email, and survey responses, for instance. Companies can discover they might have issues they didn’t know about, and then appropriately route them to be addressed. “Text analytics tools can do more sophisticated analysis and can get at the sentiment on the entity or feature level, and they can be coupled with transactional systems to drill through and find root causes,” Grimes says. “If someone filled out a hotel survey and said the room was not clean, there’s a transaction level associated with that person, and so you can find out who the housekeeper at the time of his stay was, and correct that problem.”

Analytics listening platforms that monitor Twitter or Facebook postings are increasingly used by companies to see what’s being said about them on social networks, but generally they don’t have the same reach into the surveys and other information maintained by organizations, as well as the links into transactional records. More sophisticated text analytics options hail from providers such as Attensity, Clarabridge, NetBase, and SAP Business Objects. But Grimes allows that they serve a different use case “where you want to understand what people are saying about your brands, and have some of tracking, maybe correlating that with your marketing campaigns.” Ultimately both forms need to come together to get at sentiment in a variety of forms, he notes.

▏ Specialty niches: Law enforcement, intelligence and counter-terrorism forces, clearly, also see needs to leverage sentiment analysis around all the “chatter” on the web. So too can political campaigns; B2B marketing efforts by organizations seeking to influence narrow channels (and their influencers); and financial services.

The financial services market, in fact, has a few aspects where sentiment analysis can help. One of them is similar to what CPG companies want, around satisfying customers’ issues. But another area where text analytics can help is around meeting compliance mandates, and also around trading and the opportunities that exist to draw sentiment information out of social networks. The idea is that one might be able to form conclusions about what direction a stock may soon trend.

“The financial markets stuff pushes the envelope in a number of directions,” Grimes says. “One is velocity. When you do trading, being the first to detect signals in information gives you a competitive edge,” he says. “Once everyone knows something you no longer have that, so they are working on very fast ingestion of news and other information. It’s high-volume, very fast analysis, as well as deep analysis, to create models.”

Semantics Even If Not In Name
The text analytics tools that may lie behind these efforts are semantic technologies even if they are not labeled that way, Grimes says – text analytics is entirely capable of driving the creation of semantics out of the existing web. The solutions encompass semantic information retrieval or semantic search, and also essentially business intelligence on so-called unstructured information, he says. “Some of them, like Clarabridge, call themselves customer experience management companies, and that’s a good sign when they sell solutions rather than technology,” Grimes says.

As compared to the upper-case Semantic Web, “this is here and now vs. Semantic Web stuff where you’re sucking data into RDF triple stores. That’s not delivering business value yet…. This is here and now technology being delivered by companies who are not waiting for a pie in the sky vision to be realized.”

The show next week, Grimes says, will be about trying to help existing and prospective users, especially in areas such as pharma, financials and public relations, to get a handle on the technology, and to understand the service offerings.

What possibilities do you see in sentiment analysis? Let the Semantic Web Blog know your thoughts.

About the author

Jennifer Zaino is a New York-based freelance writer specializing in business and technology journalism. She has been an executive editor at leading technology publications, including InformationWeek, where she spearheaded an award-winning news section, and Network Computing, where she helped develop online content strategies including review exclusives and analyst reports. Her freelance credentials include being a regular contributor of original content to The Semantic Web Blog; acting as a contributing writer to RFID Journal; and serving as executive editor at the Smart Architect Smart Enterprise Exchange group. Her work also has appeared in publications and on web sites including EdTech (K-12 and Higher Ed), Ingram Micro Channel Advisor, The CMO Site, and Federal Computer Week.

You might also like...

Machine Learning vs. Deep Learning

Read More →