OSINT using Twitter & Python

OSINT Footprinting

Using Twitter and Python

Who am I?
 Raymond Lilly

 @37point2

 Analyst at a Social Media/Customer Relations Management
company

 Senior, Eastern Michigan University

 Information Assurance/Network Security

What are we talking about?
 OSINT gathering methods

 Research with implications in
 Intelligence

 Social Engineering

 Marketing

Intelligence
 What are people talking about?

 Intel vs Counter Intel

 Targeting concerns
Individuals/Groups
Geographic regions
Time
Topics

Social Engineering
Leaking information

 What do your co-workers/employees talk about during/after
work?

 IT talking about new tech deployments?

 Any employees venting about internal issues?

 C levels discussing personal hobbies/travel plans?

Marketing
 Can you identify your customers?

 What are they talking about?/What other interests do they
have?

 Can you profile them and use that to reach new potential
customers?

 Find new markets?

 Reduce your customer assistance cost or increase customer
satisfaction?

Fun Stuff
 New Job info

 What’s the corporate culture like?

 Does the company embrace new tech/ideas or shun
them?

 Amplify the reach of your messages

 Find organizations/groups that are interested in the same
things you are

Key Twitter Concepts
 Tweets – 140 characters

 Following
 Friends

 Followers
 Did you pick the user?

Twitter’s API
 https://dev.twitter.com/docs/api

 Authenticated vs. Unauthenticated

 How hard is it to get OAuth Tokens?

 REST

 Streaming

Tweepy!
 Python module for Twitter’s API

 https://github.com/tweepy/tweepy/

 Joshthecoder

GET status/user_timeline
 Takes a user_id or screen_name

 since_id

 count

 exclude_replies

 include_rts

 Tweepy.api.get_status(‘37point2’)

GET users/show
 user_id/screen_name

 include_entities
 ^-- Awesome!

 Tweepy.api.get_user(‘37point2’)

 "id": 286868576,

 "id_str": "286868576",

 "name": "37point2",

 "screen_name": "37point2",

 "location": "",

 "description": "Information Assurance student at Eastern Michigan University. rnIntel
Analysis, Data Viz, Incident Response",

 "url": "http://www.linkedin.com/in/raymondlilly",

 "protected": false,

 "followers_count": 244,

 "friends_count": 992,

 "listed_count": 6,

 "created_at": "Sat Apr 23 21:25:44 +0000 2011",

 "utc_offset": -18000,

 "time_zone": "Eastern Time (US & Canada)",

"description": "Information Assurance
student at Eastern Michigan University.
rnIntel Analysis, Data Viz, Incident
Response",

Method to the Madness
 Information Needed/Gathered

 Tools used

 Visualization

 Analysis

Echo Chamber
 Last 1000 Tweets of everyone followed

 Basic Word Count

 Wordle.net

Tweets per Day
 Individual

 Last 3200 Tweets

 Community

 Last 1000 Tweets (#infosec – May 18-21)

 Plot Tweets over weekdays

Hashtags/Topics

 include_entities! #lazyhacker

 include_rts

 Google Visualization API

 Hashtags & HashtagsWithRetweets

Retweets/Replies


 include_rts!!!

 Retweets

 Replies

 Best time for a response?

Interactions w/ Influence and
Topics
 Klout

 BOO!!!

 Changes algorithm daily

 What is the algorithm? /shrug

 Weights social media sites differently

 Useful

 Topics!!

 Score used as guideline

Model

 Include all the things!

 Add Klout score and topics

Formula for Influence
 Klout^2 * interactions

 50 vs 60

 2500 vs 3600

Clients

 Total Counts

 Client usage over time

Interactions & Topics


 Maltego CaseFile

 Community Edition

Interesting Tools
 Tweetstats.com

 Twopcharts

 Klout

 Kred

 Socialmention

 NetworkX

Contact Info

 Raymond Lilly
 @37point2
 rlilly@emich.edu

OSINT using Twitter & Python

More Related Content

Viewers also liked

Similar to OSINT using Twitter & Python

In this document

OSINT using Twitter & Python