Tuesday Oct 27, 2009

Identity Trend 9: Identity Analytics

This post is the ninth in a series of eleven posts I am writing about key trends in the Identity Management industry.

Whenever data is amassed and made available for analysis, the odds are great that someone will  figure out ways to derive new meaning from this data.  So it is with data related to personal Identity.  I believe we will see an explosion of data analytics being applied to Identity-related data for a number of applications.  Three emerging areas are briefly described in this post.

Authentication/Discovery

imageConsiderable evidence is available to show how each of us is progressively establishing a historical, logical  “fingerprint” based on our personal patterns of accessing online resources.   In a blog post entitled, “Anonymized Data Really Isn't,” I discussed how correlating “anonymized” data with seemingly unrelated publicly available data can pinpoint personal identities with frightening accuracy. 

In his address at Digital ID World, Jeff Jonas’ discussion about using data analytics to discover space-time-travel characteristics of individuals was both challenging and disturbing.  Mobile operators are accumulating 600 billion cellphone transaction records annually and are selling this data to third parties who use advanced analytics to identify space/time/travel characteristics of individual people, to be used for authentication and focused marketing activities.

I expect we will soon see many ways data analytics will be used for both positive and negative purposes, to very accurately identify individual people and leverage that identification for authentication and personalization purposes.

Context/Purpose

imageJust like data analytics can be used to identify who we really are, these methods can be leveraged to personalize the experience online users have with each other and with online applications.  As I discussed in my Identity Trend blog post about Personalization and Context, personalization increases the value of online user experience by presenting relevant content to a specific user at a particular time and tailoring the user experience  to fit what a user is doing at that time.  Data analytics can be used to evaluate both real time and historical information to answer questions such as:

  • What are you doing now?
  • What did you do recently in a similar circumstance?
  • Will historical patterns predict your preferences?

Perhaps the best-known example of this is Amazon.com’s recommendation service illustrated in the photo above.  In this case, based on my historical purchase pattern, Amazon recommended two books to me.  Ironically, Amazon recommended I purchase Seth Godin’s book entitled “Permission Marketing, which addresses some of these very issues we are addressing in this post.  In the next few years, we will most likely see more powerful and refined recommendation engines based on complex data analytics, adapted to a wide variety of user interfaces.

Auditing

imageThe big question surrounding IT auditing is, “Who really did what, when and where?”  While many tools exist for maintain audit trails and evaluating compliance with audit policy, I believe we will see and emerging class of tools to evaluate audit trails and logs in ways not anticipated by current tools.  A few examples:

Sophisticated ad hoc analytics may make it easier to discover patterns of fraudulent access that may be missed by more structured audit tools. 

Enhanced analytics may help improve the business role discovery process by detecting obscure usage trends in log data.

Recommendations:

Some questions you may consider to explore how Identity Analytics may affect your enterprise include:

  1. What Identity data do you currently store?
  2. What related data do you store that could be correlated with Identity data?
  3. Can data analytics be used to correlate data you store with publicly-available data to provide value to your enterprise and your customers?
  4. What additional business value could accrue to your organization base on such analytics?
  5. That privacy and security threats may exist to your employees and your organization if advanced analytics are used to correlate publicly-available data with data you make available?
  6. How could data analytics related to Context and Preference be used to enhance the way users interact with your organization?
  7. How can advanced analytics help you combat fraud or other cybercrime?
  8. How can you use advanced analytics to improve corporate processes?

Friday Oct 02, 2009

Identity Trend 2: Authentication

This post is the second in a series of eleven articles I am writing about trends in the Identity Management industry. 

After all is said and done, Authentication continues to be right at the heart of Identity Management.  Determining whether the correct set of Identity credentials is presented, so a person or process can be granted access to the correct system, application or data, is critical to the integrity of the online experience.   Authentication is like the gatekeeper or enforcer who determines who gets in the door. 

  1. Demand for strong authentication is accelerating as the sophistication and sheer numbers of people who would defraud or damage online systems continue to grow.  More effort is being focused on just how to economically, but securely, implement strong authentication methods to protect confidential information.
  2. As the need for strong authentication grows, there has been considerable conversation about whether the pervasive use of passwords is headed for extinction.  Is the password really on its deathbed? In a Network World column posted earlier this year, Dave Kearns equated passwords to buggy whips.  In my response entitled Passwords and Buggy Whips, I challenged “Replace username/password with what?"  Until we get wide acceptance of alternate methods, it is unlikely that passwords will join buggy whips in the dustbin of history.
  3. In a subsequent post entitled, Seat Belts and Passwords ... and Buggy Whips, I proposed that “until ease of use makes passwords irrelevant, people will continue to use buggy whips or drive without seat belts.”  The key issue dogging the industry is how to provide identity credentials that are so easy to use that the technical unsavvy majority can easily use them while providing a level of security commensurate with the rising tide of online threats.

Recommendations:

  1. Assess what level of security is needed for different areas of your enterprise.  In some cases, authentication must protect high value information.  In other cases, less strong authentication may be appropriate.
  2. Seek to understand what your users need.  What methods are both secure and easy to use for them?
  3. Is the cost of strong authentication commensurate with the risk of data loss or compromised system access?
  4. What is the best combination of authentication methods to serve my user community and protect my business interests?

Many years ago, while involved in a large physical security project, we joked that you need to invest enough in your security system so it is cheaper to bribe the guard than to breach the electronic system.  The same principle may be true with Identity Authentication.

About

Discovering Identity was founded on blogs.sun.com in May 2005 as a means of documenting my exploration of the field of Identity and Access Management. In February, 2010, I switched to hosting the blog at DiscoveringIdentity.com. In March 2012, I began posting Oracle-related information in both places.

Thanks for stopping by.

Please connect with me in cyberspace at LinkedIn or Twitter.


The views expressed on this blog are my own and do not necessarily reflect the views of my employer, Oracle Corporation, or any other person or organization.

Search

Archives
« April 2014
SunMonTueWedThuFriSat
  
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
   
       
Today