Vertica

The Web provides us with a myriad of ways to express opinion and interest—from social sites such as Twitter and Facebook to blogs and community forums to product reviews in ecommerce sites to many more. As a result, customers have significant influence in shaping the perceptions of brands and products. The challenge for the managers of those entities on which opinion and interest is expressed is to understand, in an automated way and in as close to real-time as possible, what people are talking about and how they feel about those topics, so that they can better understand and respond to their community.

HP Vertica Pulse — now in private beta — is HP’s scalable, in-database answer to the problem of harnessing the volume and velocity of social media data. Executed through a single line of SQL, HP Vertica Pulse enables you to extract “attributes,” or the aspects of a brand, product, service, or event that your users and customers are talking about; and the ability to assign a sentiment score for each of these attributes, so that you can track your community’s perception on the aspects of your business that your community cares about. Understand whether your customers are looking for a particular feature, how they react to a facet of your product or service as you anticipated, or if they are suddenly encountering problems. See how their perceptions change over time.

We used HP Vertica Pulse at HP Discover to capture attendee sentiment. We captured tweets related to HP Discover, the Tweeter’s screen name, and the timestamp. We ran the tweets through HP Vertica Pulse and visualized the results in Tableau. The whole effort was up and running in just a few hours. The screenshot below shows the major aspects:

• We watched the interest of the crowd change over time. With each successive keynote, we saw new initiatives and people appear in the word cloud. Meg Whitman received a lot of press, as did HP’s New Style of IT and HAVEn. So did Kevin Bacon, who participated in Meg’s keynote.
• HP Vertica Pulse surfaced news in the data analytics world. In a trial run using tweets related to data analytics, we saw “Walmart” — not a common name in the world of analytics — appear in the word cloud. A quick drilldown in Tableau revealed that Walmart recently purchased data analytics company Inkiru.
• We captured the most prolific Tweeters. We could expand on this data to include influencer scores and reach out to the most influential posters.
• We captured sentiment on all of the tweets. In a friendly forum like HP Discover, we expect the majority of the tweets to be neutral or positive in nature.

HP Vertica Pulse is a result of an ongoing collaboration with HP Labs and is built on Labs’s Live Customer Intelligence (LCI) technology. The Labs team has already had great success with LCI as evidenced in part by their Awards Meter application. HP Vertica has also built a social media connector that loads tweets of interest directly from Twitter into the HP Vertica Analytics Platform, allowing you to start understanding your community right away.

HP Vertica Pulse is yet another example of how you can use the HP Vertica Analytics Platform to bring analytics to the data, speeding analysis time and saving the effort of transferring your data to an external system. Because HP Vertica Pulse is in-database, you can store your text data and the associated sentiment alongside sales, demographic, and other business data. HP Vertica Pulse will help you to harness the voice of your customer so that you can better serve them.

We are now accepting applications to trial Pulse in our private beta. To participate, contact me, Geeta Aggarwal, at gaggarwal@vertica.com.