Attivio, a company changing the speed and time to value of Big Data deployments, announced on May 29 that it was awarded U.S. Patent No. 8725494 for entity-level sentiment analysis. The patent addresses the market’s need to more accurately analyze, assign and understand customer sentiment within unstructured content, where multiple brands and people are referenced and discussed.
Most sentiment analysis today is conducted on a broad level to determine, for example, if a review is positive, negative or neutral, explains Attivio in a news release. The entire entry or document is assigned sentiment uniformly, regardless of whether the feedback contains multiple comments that express a combination of brand and product sentiment.
For example, a customer may post the following feedback on a review site:
“I was wrong to think I would love the new product just because it came from a trusted brand I use all the time. I love Brand X but I don’t like what they are doing with Product X compared to what Brand Y and Z are doing with their offerings. Those seem to work better.”
In the above example, the entire entry sentiment is not completely negative. The comment shows strong brand affinity and trust that could be leveraged by the brand to better engage with the customer.
“Evaluating sentiment in unstructured content without also analyzing entities gives companies a false and inaccurate blanket view of how customers feel about a product, service or brand,” said Sid Probstein, chief technology officer of Attivio. “The Attivio entity-level sentiment technology provides deeper insight into how the reviewer really feels — what they value, and what could be improved. What was previously considered too difficult or time consuming by other Big Data vendors, is accomplished within seconds with Attivio.”