SAS Turns its Analytics to Social Networks
Bank Technology News | August, 2009
|
Print
Email
Reprints
FeedbackIDC says the global volume of social network use in 2008 reached 500 billion gigabytes from Web sites, emails, RFID chips, networks, YouTube videos, industrial processors, and other sources. SAS believes there’s a huge opportunity to make fraud fighting more efficient by tapping and analyzing that volume, so it’s launched a new Social Network Analysis Tool for both CRM and security purposes.
“Banks spend a lot of time and resources on false positives,” says Ellen Joyner, a manager in SAS’ global financial services unit, who would not divulge adoption numbers for the new product. “It’s a problem that social network analysis can help by correctly identifying ‘positive’ fraud cases, allowing more efficient use of risk and loss prevention resources.”
SAS’s new tool, which will face competition from similar products offered by Detica, Memento and Norkom, helps investigators search beyond transaction and customer views to analysis of all related activities and relationships within a network, such as shared addresses, phone numbers, employment information, account ownership and other transactional data. “You may be tied to an individual, a family, a small business or an organization, eventually you may be linked to a known fraudster,” Joyner says. “You can get a sense of how people may be tied to fraud.”
Avivah Litan, vp and distinguished analyst for Gartner, says making the “invisible visible” social networking analysis is an effective and emerging tool for fighting fraud and understanding customers.
SAS CEO Jim Goodnight helped write the code for SAS’ new software, which also provides graphical network analysis allowing banks to identify relationships that might not have been apparent. The firm has filed for a patent for the technology, called NetCHAID, which attempts to examine large data clusters for trends and relationships.
An additional upgrade scheduled for the fourth quarter will leverage SAS Sentiment Analysis Manager to capture consumer product reviews and brand comments from mainstream sites such as Amazon, Overstock, message boards, blogs and Twitter to locate digital content in real time to get a sense of the writer’s opinion, spot trends and uncover product defects. IDC analyst Sue Feldman says using text mining and natural language processing can help marketers glean intelligence from social media sites and gain a more complete view of customer sentiment.
| More articles in Bank Technology News |
| Subscribe to Bank Technology News |
