User Interfaces and Algorithms for Fighting Phishing

1
User Interfaces and Algorithms for Fighting
Phishing
Jason I. HongCarnegie Mellon University
2
Everyday Privacy and Security Problem
3
This entire process known as phishing
4
Phishing is a Plague on the Internet

• Estimated 3.5 million people have fallen for
  phishing
• Estimated 350m-2b direct losses a year
• 31000 unique phishing sites reported in June 2007
• Easier (and safer) to phish than rob a bank

5
Project Supporting Trust Decisions

• Goal help people make better online trust
  decisions
• Currently focusing on anti-phishing
• Large multi-disciplinary team project at CMU
• Computer science, human-computer interaction,
  public policy, social and decision sciences, CERT

6
Our Multi-Pronged Approach

• Human side
• Interviews to understand decision-making
• PhishGuru embedded training
• Anti-Phishing Phil game
• Understanding effectiveness of browser warnings
• Computer side
• PILFER email anti-phishing filter
• CANTINA web anti-phishing algorithm

Automate where possible, support where necessary
7
Our Multi-Pronged Approach

• Human side
• Interviews to understand decision-making
• PhishGuru embedded training
• Anti-Phishing Phil game
• Understanding effectiveness of browser warnings
• Computer side
• PILFER email anti-phishing filter
• CANTINA web anti-phishing algorithm

What do users know about phishing?
8
Interview Study

• Interviewed 40 Internet users (35 non-experts)
• Mental models interviews included email role
  play and open ended questions
• Brief overview of results (see paper for details)
• J. Downs, M. Holbrook, and L. Cranor. Decision
  Strategies and Susceptibility to Phishing. In
  Proceedings of the 2006 Symposium On Usable
  Privacy and Security, 12-14 July 2006,
  Pittsburgh, PA.

9
Little Knowledge of Phishing

• Only about half knew meaning of the term
  phishing
• Something to do with the band Phish, I take it.

10
Little Attention Paid to URLs

• Only 55 of participants said they had ever
  noticed an unexpected or strange-looking URL
• Most did not consider them to be suspicious

11
Some Knowledge of Scams

• 55 of participants reported being cautious when
  email asks for sensitive financial info
• But very few reported being suspicious of email
  asking for passwords
• Knowledge of financial phish reduced likelihood
  of falling for these scams
• But did not transfer to other scams, such as an
  amazon.com password phish

12
Naive Evaluation Strategies

• The most frequent strategies dont help much in
  identifying phish
• This email appears to be for me
• Its normal to hear from companies you do
  business with
• Reputable companies will send emails
• I will probably give them the information that
  they asked for. And I would assume that I had
  already given them that information at some point
  so I will feel comfortable giving it to them
  again.

13
Summary of Findings

• People generally not good at identifying scams
  they havent specifically seen before
• People dont use good strategies to protect
  themselves
• Currently running large-scale survey across
  multiple cities in the US to gather more data
• Amazon also active in looking for fake domain
  names

14
Outline

• Human side
• Interviews to understand decision-making
• PhishGuru embedded training
• Anti-Phishing Phil game
• Understanding effectiveness of browser warnings
• Computer side
• PILFER email anti-phishing filter
• CANTINA web anti-phishing algorithm

Can we train people not to fall for phish?
15
Web Site Training Study

• Laboratory study of 28 non-expert computer users
• Asked participants to evaluate 20 web sites
• Control group evaluated 10 web sites, took 15 min
  break to read email or play solitaire, evaluated
  10 more web sites
• Experimental group same as above, but spent 15
  min break reading web-based training materials
• Experimental group performed significantly better
  identifying phish after training
• Less reliance on professional-looking designs
• Looking at and understanding URLs
• Web site asks for too much information

People can learn from web-based training
materials, if only we could get them to read
them!
16
How Do We Get People Trained?

• Most people dont proactively look for training
  materials on the web
• Companies send security notice emails to
  employees and/or customers
• We hypothesized these tend to be ignored
• Too much to read
• People dont consider them relevant
• People think they already know how to protect
  themselves
• Led us to idea of embedded training

17
Embedded Training

• Can we train people during their normal use of
  email to avoid phishing attacks?
• Periodically, people get sent a training email
• Training email looks like a phishing attack
• If person falls for it, intervention warns and
  highlights what cues to look for in succinct and
  engaging format
• P. Kumaraguru, Y. Rhee, A. Acquisti, L. Cranor,
  J. Hong, and E. Nunge. Protecting People from
  Phishing The Design and Evaluation of an
  Embedded Training Email System. CHI 2007.

18
Embedded Training Example
Subject Revision to Your Amazon.com Information
Please login and enter your information
http//www.amazon.com/exec/obidos/sign-in.html
19
Intervention 1 Diagram
20
Intervention 1 Diagram
Explains why they are seeing this message
21
Intervention 1 Diagram
Explains what a phishing scam is
22
Intervention 1 Diagram
Explains how to identify a phishing scam
23
Intervention 1 Diagram
Explains simple things you can do to protect self
24
Intervention 2 Comic Strip
25
Intervention 2 Comic Strip
26
Intervention 2 Comic Strip
27
Embedded Training Evaluation 1

• Lab study comparing our prototypes to standard
  security notices
• Group A eBay, PayPal notices
• Group B Diagram that explains phishing
• Group C Comic strip that tells a story
• 10 participants in each condition (30 total)
• Screened so we only have novices
• Go through 19 emails, 4 phishing attacks
  scattered throughout, 2 training emails too
• Role play as Bobby Smith at Cognix Inc

28
Embedded Training Results
29
Embedded Training Results

• Existing practice of security notices is
  ineffective
• Diagram intervention somewhat better
• Though people still fell for final phish
• Comic strip intervention worked best
• Statistically significant
• Combination of less text, graphics, story?

30
Evaluation 2

• New questions
• Have to fall for phishing email to be effective?
• How well do people retain knowledge?
• Roughly same experiment as before
• Role play as Bobby Smith at Cognix Inc, go thru
  16 emails
• Embedded condition means have to fall for our
  email
• Non-embedded means we just send the comic strip
• Also had people come back after 1 week
• To appear in APWG eCrime Researchers Summit (Oct
  4-5 at CMU)

31
(No Transcript)
32
Results of Evaluation 2

• Have to fall for phishing email to be effective?
• How well do people retain knowledge after a week?

33
Results of Evaluation 2

• Have to fall for phishing email to be effective?
• How well do people retain knowledge after a week?

Correctness
34
Results of Evaluation 2

• Have to fall for phishing email to be effective?
• How well do people retain knowledge after a week?

Correctness
35
(No Transcript)
36
(No Transcript)
37
(No Transcript)
38
Anti-Phishing Phil

• A game to teach people not to fall for phish
• Embedded training focuses on email
• Our game focuses on web browser
• Goals
• How to parse URLs
• Where to look for URLs
• Use search engines for help
• Try the game!
• http//cups.cs.cmu.edu/antiphishing_phil

39
Anti-Phishing Phil
40
(No Transcript)
41
(No Transcript)
42
(No Transcript)
43
(No Transcript)
44

45
Evaluation of Anti-Phishing Phil

• Test participants ability to identify phishing
  web sites before and after training up to 15 min
• 10 web sites before training, 10 after,
  randomized order
• Three conditions
• Web-based phishing education
• Printed tutorial of our materials
• Anti-phishing Phil
• 14 participants in each condition
• Screened out security experts
• Younger, college students

46
Results

• No statistically significant difference in false
  negatives among the three groups
• Actually a phish, but participant thinks its not
• Unsure why, though game group had fewest false
  positives
• Press release last month, 50k new users
• Still analyzing results
• High knowledge retention 1 week later by
  participants
• Faster at identifying phish (12 seconds to 6
  seconds)
• Banks, non-profits, consulting firms, Air Force,
  ISPs

47
(No Transcript)
48

49
Outline

• Human side
• Interviews to understand decision-making
• PhishGuru embedded training
• Anti-Phishing Phil game
• Understanding effectiveness of browser warnings
• Computer side
• PILFER email anti-phishing filter
• CANTINA web anti-phishing algorithm

Do people see, understand, and believe web
browser warnings?
50
Screenshots
Internet Explorer Passive Warning
51
Screenshots
Internet Explorer Active Block
52
Screenshots
Mozilla FireFox Active Block
53
How Effective are these Warnings?

• Tested four conditions
• FireFox Active Block
• IE Active Block
• IE Passive Warning
• Control (no warnings or blocks)
• Shopping Study
• Setup some fake phishing pages and added to
  blacklists
• Users were phished after purchases
• Real email accounts and personal information
• Spoofing eBay and Amazon (2 phish/user)
• We observed them interact with the warnings

54
How Effective are these Warnings?
55
How Effective are these Warnings?
56
Discussion of Phish Warnings

• Nearly everyone will fall for highly contextual
  phish
• Passive IE warning failed for many reasons
• Didnt interrupt the main task
• Slow to appear (up to 5 seconds)
• Not clear what the right action was
• Looked too much like other ignorable warnings
  (habituation)
• Bug in implementation, any keystroke dismisses

57
Screenshots
Internet Explorer Passive Warning
58
Discussion of Phish Warnings

• Active IE warnings
• Most saw but did not believe it
• Since it gave me the option of still proceeding
  to the website, I figured it couldnt be that
  bad
• Some element of habituation (looks like other
  warnings)
• Saw two pathological cases

59
Screenshots
Internet Explorer Active Block
60
A Science of Warnings

• See the warning?
• Understand?
• Believe it?
• Motivated?
• Planning on refining this model for computer
  warnings

61
Outline

• Human side
• Interviews to understand decision-making
• PhishGuru embedded training
• Anti-Phishing Phil game
• Understanding effectiveness of browser warnings
• Computer side
• PILFER email anti-phishing filter
• CANTINA web anti-phishing algorithm

Can we automatically detect phish emails?
62
PILFER Email Anti-Phishing Filter

• Philosophy automate where possible, support
  where necessary
• Goal Create email filter that detects phishing
  emails
• Spam filters well-explored, but how good for
  phishing?
• Can we create a custom filter for phishing?
• I. Fette, N. Sadeh, A. Tomasic. Learning to
  Detect Phishing Emails. In W W W 2007.

63
PILFER Email Anti-Phishing Filter

• Heuristics combined in SVM
• IP addresses in link (http//128.23.34.45/blah)
• Age of linked-to domains (younger domains likely
  phishing)
• Non-matching URLs (ex. most links point to
  PayPal)
• Click here to restore your account
• HTML email
• Number of links
• Number of domain names in links
• Number of dots in URLs (http//www.paypal.update.e
  xample.com/update.cgi)
• JavaScript
• SpamAssassin rating

64
PILFER Evaluation

• Ham corpora from SpamAssassin (2002 and 2003)
• 6950 good emails
• Phishingcorpus
• 860 phishing emails

65
PILFER Evaluation
66
PILFER Evaluation

• PILFER now implemented as SpamAssassin filter
• Alas, Ian has left for Google

67
Outline

• Human side
• Interviews to understand decision-making
• PhishGuru embedded training
• Anti-Phishing Phil game
• Understanding effectiveness of browser warnings
• Computer side
• PILFER email anti-phishing filter
• CANTINA web anti-phishing algorithm

How good is phish detection for web sites? Can
we do better?
68
Lots of Phish Detection Algorithms

• Dozens of anti-phishing toolbars offered
• Built into security software suites
• Offered by ISPs
• Free downloads (132 on download.com)
• Built into latest version of popular web browsers

69
Lots of Phish Detection Algorithms

• Dozens of anti-phishing toolbars offered
• Built into security software suites
• Offered by ISPs
• Free downloads (132 on download.com)
• Built into latest version of popular web browsers
• But how well do they detect phish?
• Short answer still room for improvement

70
Testing the Toolbars

• November 2006 Automated evaluation of 10
  toolbars
• Used phishtank.com and APWG as source of phishing
  URLs
• Evaluated 100 phish and 510 legitimate sites
• Y. Zhang, S. Egelman, L. Cranor, J. Hong.
  Phinding Phish An Evaluation of Anti-Phishing
  Toolbars. NDSS 2006.

71
Testbed System Architecture
72
Results
38 false positives
1 false positives
PhishTank
73
APWG
74
Results

• Only one toolbar gt90 accuracy (but high false
  positives)
• Several catch 70-85 of phish with few false
  positives

75
Results

• Only one toolbar gt90 accuracy (but high false
  positives)
• Several catch 70-85 of phish with few false
  positives
• Can we do better?
• Can we use search engines to help find phish?
• Y. Zhang, J. Hong, L. Cranor. CANTINA A
  Content-Based Approach to Detecting Phishing Web
  Sites. In W W W 2007.

76
Robust Hyperlinks

• Developed by Phelps and Wilensky to solve 404
  not found problem
• Key idea was to add a lexical signature to URLs
  that could be fed to a search engine if URL
  failed
• Ex. http//abc.com/page.html?sigword1word2...
  word5
• How to generate signature?
• Found that TF-IDF was fairly effective
• Informal evaluation found five words was
  sufficient for most web pages

77
Adapting TF-IDF for Anti-Phishing

• Can same basic approach be used for
  anti-phishing?
• Scammers often directly copy web pages
• With Google search engine, fake should have low
  page rank

Fake
Real
78
How CANTINA Works

• Given a web page, calculate TF-IDF score for
  each word in that page
• Take five words with highest TF-IDF weights
• Feed these five words into a search engine
  (Google)
• If domain name of current web page is in top N
  search results, we consider it legitimate
• N30 worked well
• No improvement by increasing N
• Later, added some heuristics to reduce false
  positives

79
Fake
eBay, user, sign, help, forgot
80
Real
eBay, user, sign, help, forgot
81
(No Transcript)
82
(No Transcript)
83
Evaluating CANTINA
PhishTank
84
Weaknesses in CANTINA

• Bad guys may try to subvert search engines
• Only works if legitimate page is indexed
• Intranets
• May be confused if same login page in multiple
  places

85
Summary

• Whirlwind tour of our work on anti-phishing
• Human side how people make decisions, training,
  UIs
• Computer side better algorithms for detecting
  phish
• More info about our work at cups.cs.cmu.edu

86
Acknowledgments

• Alessandro Acquisti
• Lorrie Cranor
• Sven Dietrich
• Julie Downs
• Mandy Holbrook
• Norman Sadeh
• Anthony Tomasic
• Umut Topkara

• Serge Egelman
• Ian Fette
• Ponnurangam Kumaraguru
• Bryant Magnien
• Elizabeth Nunge
• Yong Rhee
• Steve Sheng
• Yue Zhang
• Shelley Zheng

Supported by NSF, ARO, CyLab, Portugal Telecom
87
http//cups.cs.cmu.edu/
CMU Usable Privacy and Security Laboratory
88
(No Transcript)