26
Identifying Warning Behaviors of Violent Lone Offenders in Written Communication 1 Amendra Shrestha 1 Lisa Kaati 2 Tony Sardella 1 Uppsala University 2 Washington University December 12, 2016

Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Embed Size (px)

Citation preview

Page 1: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Identifying Warning Behaviors of Violent LoneOffenders in Written Communication

1Amendra Shrestha

1 Lisa Kaati 2 Tony Sardella

1Uppsala University

2Washington University

December 12, 2016

Page 2: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

1 IntroductionExampleViolent lone offenders

2 Countering VLOsVLOsLIWC

3 Data

4 Experiments

5 Conclusion

- 1 -

Page 3: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

Example

School shootings

https://everytownresearch.org/school-shootings/

- 2 -

Page 4: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

Example

Lone actor terrorist attacks

- 3 -

Page 5: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

Example

Mass murderers

- 4 -

Page 6: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

Violent lone offenders

Violent Lone Offenders (VLO)

• VLOs : school shooters, lone actor terrorists, mass murderers

• wide factors : social status, ideology, mental health,personality type

• rare events

• pose a serious security threat to a society

• shows sign of psychological warning behaviours

• challenging to detect prior to an event

• challenge to identify, target and arrest

• common that they leave digital trace prior to attack

- 5 -

Page 7: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

Violent lone offenders

Mass murderer : Dylan Roof

• killed 9 persons in a church shooting in Charleston, SouthCarolina

• published a manifesto on a website supporting whitesupremacy

- 6 -

Page 8: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

Violent lone offenders

Lone actor terrorist: Anders Breivik

• killed 8 people by detonating a van bomb in Oslo

• shot dead 69 participants of a Workers’ Youth League

• distributed a compendium of texts describing his militantideology

- 7 -

Page 9: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

VLOs

Countering VLOs

• analyze and understanding potential signals in writtencommunication

• can be used to stop these attacks

• combine weak signals and gain informations about intentions

• weak signals• signs of an individuals radical beliefs and extreme hate• knowledge about how to produce homemade explosives• interest in firearms and signs of rehearsal• signs of warning behaviours from written text

- 8 -

Page 10: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

VLOs

- 9 -

Investigate possibilities to identify potentialviolent lone offenders based on writtencommunication using machine learning

Page 11: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

VLOs

- 10 -

• Electronic and written text(manifestos, letters, blogs, etc.)

comparision←−−−−−→

Profile of VLOs text Profile of non-VLOs users text

Page 12: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

LIWC

LIWC

- 11 -

• Linguistic Inquiry and Word Count

• a computerized word counting tool

• counts words in psychologically meaningful categories

Page 13: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

LIWC

- 12 -

Psychologist User

Page 14: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

LIWC

- 13 -

Page 15: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

LIWC

- 14 -

Page 16: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

LIWC

- 15 -

Page 17: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

LIWC

LIWC Categories

- 16 -

Page 18: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

Data

- 17 -

Figure : Jose Reyes’s Letters

Page 19: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

Data

• VLOs• manifesto, personal letter, suicide letter written by school

shooters, mass murderers and lone offenders• 32 violent lone offenders : 46 documents

• Non-VLOs• 54 blogs written about personal interests, news, fashion and

photography• 108 stormfront users and their posts• 108 boards.ie users and their posts

- 18 -

Page 20: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

Experiment Setup

• Feature selection : Mahalanobis relevance estimate

• Synthetic Minority Over-sampling Technique (SMOTE)

• Leave-One-Out Cross-Validation (LOOCV)

• Adaboost

• Java and R

- 19 -

Page 21: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

Experiment 1: Weak signals of warning behavior

• if it is possible to separate texts written by VLO

• combined lone offenders into one set

• combined blogs, Stormfront and Boards.ie data into one set

• 11 important features used

• results :• Accuracy : 0.8766• Blogs + Forums : 254 out of 270 are correctly classified• VLO : 33 out of 46 are correctly classified

- 20 -

Page 22: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

Experiment 2: Bloggers

• possibility to identify lone offenders from bloggers

• 12 important features used

• results : blog vs VLO• Accuracy : 0.89• Blogs : 50 out of 54 are correctly classified• VLO : 39 out of 46 are correctly classified

- 21 -

Page 23: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

Experiment 3: Stormfront users

• identify lone offenders from Stormfront users

• 10 important features used

• results : Stormfront vs VLO• Accuracy : 0.9026• Forum : 100 out of 108 are correctly classified• LO : 33 out of 35 are correctly classified

- 22 -

Page 24: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

Experiment 4: Boards.ie users

• identify lone offenders from boards.ie users

• 10 important features used

• results : boards.ie vs VLO• Accuracy : 0.9221• Forum : 100 out of 108 are correctly classified• LO : 42 out of 46 are correctly classified

- 23 -

Page 25: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

Conclusion

• machine learning can be use to identify texts written byviolent lone offenders

• consider ethical issues

• aid for human analyst

- 24 -

Page 26: Identifying Warning Behaviors of Violent Lone Offenders in Written Communication

Outline Introduction Countering VLOs Data Experiments Conclusion

- 25 -

Thank You