Database Reference
In-Depth Information
game changer is in fraud detection. To understand the complete story around
transactions like securities trading, insurance claims, or mortgage applica-
tions, all of the data surrounding these transactions needs to be analyzed. In
addition to relational data, this invariably includes raw text (from emails or
text fields in forms) and semistructured data (from log files).
Data redaction
is another area in which a text analytics platform greatly
enhances the capability of the discipline. A great deal of data being stored
today includes personally identifiable information (PII), which in many
countries must be adequately governed to protect citizens' privacy. To ensure
regulatory compliance, this data (for example, from forms, records, or legal
documents) must be redacted to hide key elements, which would otherwise
reveal people's identities. In fact, the IBM InfoSphere Guardium Data Redac-
tion product uses the text analytics technology that's core to the IBM Big
Data platform for this exact purpose.
For business-to-consumer-oriented companies, especially within the ser-
vice industries, having a complete picture of an individual customer's account
is essential—this is the domain of customer relationship management (CRM)
analytics. Many valuable activities, such as targeted marketing and customer
churn prediction, depend on understanding a customer's behavior, and because
this involves not only their transactional history, but also transcriptions of their
call center interactions, and even clickstream logs of their visits to the corporate
web presence, text analytics are required to find “that next gear.”
Text analytics can boost a Big Data project in a very obvious way in the
area of social media analytics
.
With the advent of online communities such as
Facebook, and micro-blogging services such as Twitter, people are publicly
expressing their personal feelings on a scale we've never seen before. Businesses
could gain a greater understanding of individual customers by understanding
not just what they're saying in social media, but why they are saying it, too. In
addition, knowing what groups of people are saying on social media can
revolutionize how marketers assess the reach of and response to their
campaigns.
While it's obvious that people quickly envision sentiment use cases when
they hear the phrase “text analytics,” we want to make very certain that you
understand it's so much more; from piracy detection, sentiment analysis,
investment research, and more, a Big Data platform requires a sound text
analytics ecosystem.
