Social Media Data Collection and Persona Development 

There is an immense amount of data in the social media space that can be 

collected and analyzed for valuable information that can aid cyber operations.  

Much of the information that can be gleaned is not at first obvious but 

materializes through link analysis and data correlation across social media 

platforms.  To do this work today is a very manual process, but much of it can be 

automated.  Once the capability is built you could point the collection at any 

target, the output of which would be a detailed target profile with link diagrams 

showing the social connections between entities.  You can take this data and 

develop social media platform capabilities (persona s, landing pages) that fit 

within the target profiles.  A developed capability to manage these profiles and 

keep them fresh with be the last stage of development.

PHASE 1

The first phase of our effort will be to develop the social media data collectors 

and analysis tools.  For the purposes of initial development we will focus on the 

following platforms:

1. Facebook

2. LinkedIn

3. Twitter

4. Blogs

5. Myspace

6. Forums

7. Google Buzz

The research and development of this phase will take the following path.

1. Determine what social media data to collect from the different platforms 

(profession, location, friends, posts, education, subscriptions).

2. Develop a database schema to store collected data.

3. Develop the collectors using the social media APIs.  For those sites that 

don t have robust APIs to allow for easy collection we will develop 

scrapers.

4. Develop the analytics for data correlation and link analysis.  Determine 

what is statistically normal for a given persona type by profession, 

demographics, etc.   Also to determine what is trending conversationally 

by topic.

PHASE II

During the second phase we will build the persona management capability.  The 

methods for managing a set of personas for future use on Social Media platforms 

are simple.  Most of the social media platforms do not necessarily front the name 

of an individual, often times the accounts are handles for users, often the names 

represent the focus of the page or individual, such as “socialmediaguru”.  The 

only two platforms that are an exception are Facebook and LinkedIn, which 

typically run off of real names.  Often times you might want to mature a set of platforms but you are not completely clear how those platforms will be used in 

the future.