You are here:  Home  >  Data Education  >  Current Article

Cruxly Analytics Technology Drives Actions From Intents

By   /  March 5, 2014  /  No Comments

Image courtesy: Flickr/ M4D GROUP

Image courtesy: Flickr/ M4D GROUP

by Jennifer Zaino

What are your customers – or potential clients – saying or asking online, often in short texts and streaming posts, or in emails about your products, services, or their own particular interests or desires? If you can understand their actionable intents in realtime, then you have a good shot at responding swiftly and appropriately to those expressed intents, requests, or queries. That could add up to new sales, new customers, and better marketing and product management.

Startup Cruxly, which is presenting at this week’s Sentiment Analysis Symposium in NYC, believes it’s taking the oft-touted concept of social media monitoring in a new direction with its platform that applies natural language processing techniques for intent detection in realtime. “The idea is to be actionable,” says CEO Aloke Guha. “If it’s not actionable, at most [monitoring] is a nice-to-have [capability].”

How it’s getting there includes the application of grammar-aware parsing, in which grammar rules are derived from the specific event or event class being defined. Users can set up trackers for categories like Want, Likes, Buying Intent, Question/Request, and so on. Its solution looks not just for sentiment and opinion but for actionable verbs that show things like a strong buying intent even when likes or dislikes aren’t present. And, because a big challenge is to respond as soon as possible to someone, “you have to optimize how quickly you can parse while being grammar-aware, especially with verb classification being critical,” he says.


In a sentence like “I really need a new phone bad,” for example, the word “need” classifies the comment in the “intent to buy” category. And the word “bad,” which might be interpreted by more traditional sentiment analytics solutions as a negative concept, is also indicative of a very strong signal of the writer’s intent to buy a new phone, Guha says. “Without being able to parse grammar, you cannot detect this,” he says.

That said, Cruxly trades off the processing-time-intensive full-parsing of every word in order to detect as accurately as possible actionable intents at high volume in a very short time. The goal is to analyze a comment in 30 ms. “Our job is distinct from others in that we are not doing batch processing,” says CTO and co-founder Kapil Tundwal. “A tweet is only important if it has an intention, [and then it’s important to] detect that sooner. A lot of thought engineering went into make realtime detection possible at scale and that is important to us.”

Cruxly currently uses Amazon’s Web Services including scalable storage with its DynamoDB NoSQL database. The service can capture the data for later analysis; its collection and enrichment of the data can facilitate things like creating mashups to see the distribution of users who’ve expressed strong buying intents.

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...

Thinking Inside the Box: How to Audit an AI

Read More →