Big Data Requires a Text Analytics Approach that Scales
It’s hard to pinpoint when and how a small project had morphed into an unwieldy, time-consuming distraction. What started out as a small project to analyze consumer text and tag certain content pieces with a specific annotation had morphed and grown into an almost unmanageable manual effort. At the start, a small team had been able to manually review each piece of consumer text and using a defined tagging system flag different pieces of content with an annotation.
Big Data Requires a Text Analytics Approach that Scales
It’s hard to pinpoint when and how a small project had morphed into an unwieldy, time-consuming distraction. What started out as a small project to analyze consumer text and tag certain content pieces with a specific annotation had morphed and grown into an almost unmanageable manual effort. At the start, a small team had been able to manually review each piece of consumer text and using a defined tagging system flag different pieces of content with an annotation. But now they were beginning to spend more time reviewing and tagging the content rather than analyzing the findings from this effort; the effort to organize the information was quickly overwhelming the resulting benefits.
We worked with our client to understand not only their annotation system but also the purpose of the effort and best possible outcome. We configured our semantic engine to not only correctly annotate the text but our language processing engine was able to achieve a more accurate result, more quickly. Data was fed directly into our semantic engine, tagged, then exported back to the client. Once the analysis was complete, the resulting information would be fed into the client’s existing data system.
Interested in learning more, read the full Text Analytics Case Study. The case study is an excellent example of applying statistical language processing to a large volume of text.
Thanks for reading.
