This is a lecture given at "SRV - 2000: Engaged Citizenship" class at Wright State University on 6th March, 2018 on Youth Violence. All the publications related to the work presented in this talk are listed below.
[1] S. Wijeratne, D. Doran, A. Sheth, and J. L. Dustin, “Analyzing the social media footprint of street gangs”, In Proc. of IEEE ISI, 2015, pages 91–96, May 2015.
[2] L. Balasuriya, S. Wijeratne, D. Doran, and A. Sheth, “Finding street gang members on twitter”, In The 2016 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM 2016). San Francisco, CA, USA; 2016.
[3] S. Wijeratne, L. Balasuriya, D. Doran, A. Sheth, “Word Embeddings to Enhance Twitter Gang Member Profile Identification”, In IJCAI Workshop on Semantic Machine Learning (SML 2016). New York City, NY: CEUR-WS; 2016.
[4] Lakshika Balasuriya, Sanjaya Wijeratne, Derek Doran, Amit Sheth. “Signals Revealing Street Gang Members on Twitter”, In Workshop on Computational Approaches to Social Modeling (ChASM 2016) co-located with 8th International Conference on Social Informatics (SocInfo 2016). Bellevue, WA, USA; 2016.
Gang affiliates have joined the masses who use social media to share thoughts and actions. Perhaps paradoxically, they use this public medium to express recent illegal actions, to intimidate others, and to share outrageous images and statements. Agencies able to unearth these profiles may thus be able to anticipate, stop, or hasten the investigation gang related crimes and activities. This talk discusses our efforts in analyzing street gangs on twitter, with an emphasis on discovering their profiles. Our approach, which uses deep learning to embed signals in tweet language, images, shared YouTube links, and emoji use into a vector space for machine learning classifiers, recovers gang member profiles with promising accuracy and a low false positive rate.
Most street gang members use Twitter to intimidate others, to present outrageous images and statements to the world, and to share recent illegal activities. Their tweets may thus be useful to law enforcement agencies to discover clues about recent crimes or to anticipate ones that may occur. Finding these posts, however, requires a method to discover gang member Twitter profiles. This is a challenging task since gang members represent a very small population of the 320 million Twitter users. This paper studies the problem of automatically finding gang members on Twitter. It outlines a process to curate one of the largest sets of verifiable gang member profiles that have ever been studied. A review of these profiles establishes differences in the language, images, YouTube links, and emojis gang members use compared to the rest of the Twitter population. Features from this review are used to train a series of supervised classifiers. Our classifier achieves a promising F1 score with a low false positive rate.
Link to the paper - http://knoesis.org/?q=node/2754
Gang affiliates have joined the masses who use social media to share thoughts and actions publicly. Interestingly, they use this public medium to express recent illegal actions, to intimidate others, and to share outrageous images and statements. Agencies able to unearth these profiles may thus be able to anticipate, stop, or hasten the investigation of gang-related crimes. This paper investigates the use of word embeddings to help identify gang members on Twitter. Building on our previous work, we generate word embeddings that translate what Twitter users post in their profile descriptions, tweets, profile images, and linked YouTube content to a real vector format amenable for machine learning classification. Our experimental results show that pre-trained word embeddings can boost the accuracy of supervised learning algorithms trained over gang members’ social media posts.
Our paper presented at IEEE ISI 2015. Download the paper at - http://www.knoesis.org/library/resource.php?id=2148
Abstract
Gangs utilize social media as a way to maintain threatening virtual presences, to communicate about their activities, and to intimidate others. Such usage has gained the attention of many justice service agencies that wish to create better crime prevention and judicial services. However, these agencies use analysis methods that are labor intensive and only lead to basic, qualitative data interpretations. This paper presents the architecture of a modern platform to discover the structure, function, and operation of gangs through the lens of social media. Preliminary analysis of social media posts shared in the greater Chicago, IL region demonstrate the platform's capability to understand gang members' social media usage patterns.
Gang affiliates have joined the masses who use social media to share thoughts and actions. Perhaps paradoxically, they use this public medium to express recent illegal actions, to intimidate others, and to share outrageous images and statements. Agencies able to unearth these profiles may thus be able to anticipate, stop, or hasten the investigation gang related crimes and activities. This talk discusses our efforts in analyzing street gangs on twitter, with an emphasis on discovering their profiles. Our approach, which uses deep learning to embed signals in tweet language, images, shared YouTube links, and emoji use into a vector space for machine learning classifiers, recovers gang member profiles with promising accuracy and a low false positive rate.
Most street gang members use Twitter to intimidate others, to present outrageous images and statements to the world, and to share recent illegal activities. Their tweets may thus be useful to law enforcement agencies to discover clues about recent crimes or to anticipate ones that may occur. Finding these posts, however, requires a method to discover gang member Twitter profiles. This is a challenging task since gang members represent a very small population of the 320 million Twitter users. This paper studies the problem of automatically finding gang members on Twitter. It outlines a process to curate one of the largest sets of verifiable gang member profiles that have ever been studied. A review of these profiles establishes differences in the language, images, YouTube links, and emojis gang members use compared to the rest of the Twitter population. Features from this review are used to train a series of supervised classifiers. Our classifier achieves a promising F1 score with a low false positive rate.
Link to the paper - http://knoesis.org/?q=node/2754
Gang affiliates have joined the masses who use social media to share thoughts and actions publicly. Interestingly, they use this public medium to express recent illegal actions, to intimidate others, and to share outrageous images and statements. Agencies able to unearth these profiles may thus be able to anticipate, stop, or hasten the investigation of gang-related crimes. This paper investigates the use of word embeddings to help identify gang members on Twitter. Building on our previous work, we generate word embeddings that translate what Twitter users post in their profile descriptions, tweets, profile images, and linked YouTube content to a real vector format amenable for machine learning classification. Our experimental results show that pre-trained word embeddings can boost the accuracy of supervised learning algorithms trained over gang members’ social media posts.
Our paper presented at IEEE ISI 2015. Download the paper at - http://www.knoesis.org/library/resource.php?id=2148
Abstract
Gangs utilize social media as a way to maintain threatening virtual presences, to communicate about their activities, and to intimidate others. Such usage has gained the attention of many justice service agencies that wish to create better crime prevention and judicial services. However, these agencies use analysis methods that are labor intensive and only lead to basic, qualitative data interpretations. This paper presents the architecture of a modern platform to discover the structure, function, and operation of gangs through the lens of social media. Preliminary analysis of social media posts shared in the greater Chicago, IL region demonstrate the platform's capability to understand gang members' social media usage patterns.
Analyzing the Social Media Footprint of Street GangsSanjaya Wijeratne
Abstract
Gangs utilize social media as a way to maintain threatening virtual presences, to communicate about their activities, and to intimidate others. Such usage has gained the attention of many justice service agencies that wish to create better crime prevention and judicial services. However, these agencies use analysis methods that are labor intensive and only lead to basic, qualitative data interpretations. This paper presents the architecture of a modern platform to discover the structure, function, and operation of gangs through the lens of social media. Preliminary analysis of social media posts shared in the greater Chicago, IL region demonstrate the platform's capability to understand gang members' social media usage patterns.
Link to Paper - http://knoesis.wright.edu/researchers/sanjaya/papers/2015/Wijeratne_ISI_2015.pdf
Social media research is an executable form of market research. The presentation covers how companies can implement a social media research process in their organization.
Twitter's Elections Integrity Datasets (Galvanize; February 22, 2019)Ivan Corneillet
Short presentation of Twitter's publicly available archives of Tweets and media that the company believed resulted from potentially state-backed information operations on its service for the benefit of Galvanize's Data Science Immersive Fellows on February 22, 2019 in San Francisco.
Advertise on Instagram: Content Creation GuidelinesIshan Gandhi
Looking to advertise on Instagram? This presentation would provide valuable inputs for your social media marketing campaign in India.
Following are some of the very interesting reasons why you should choose to advertise on Instagram:
1. Over ⅓ of Instagram Users Have Used The App To Purchase A Product Online.
2. Instagram’s Ad Platform Has Access To The Most Advanced Social Media Advertising Targeting Options Available.
3. Sales and Leads Are Trackable Through Instagram Ads So You Can See Clear ROI.
4. Instagram Differentiates Personal Profiles from Business Profiles.
5. Instagram’s User Base Allows You To Tap Into An Otherwise Unreached Customer Base.
(Reasons by Lyfemarketing)
For any further assistance or query, you can contact me at- ishan711gandhi@gmail.com
Social Websites And Seo Social Dev Camp Chicago2008 By John FairleyJohn Fairley
SocialDevCampChicago 2008 Comparing Search Engine Optimization of social media websites, including digg, stumbleupon and twitter. Learn SEO best practices and pitfalls to avoid.
The Power Point to support my presentation on use of Social Media in higher education. NISOD Conference on Teaching and Leadership Excellence, May 30 2011
This is a presentation I delivered for a group of marketing managers from Ernst & Young on December 3, 2009. These managers are based in the western US and focus primarily on E&Y's Entrepreneur of the Year program.
The presentation covers the basics of social media, stats on leading tools and recommendations for ramping up social media activities individually and for the Entrepreneur of the Year team.
Read| The latest issue of The Challenger is here! We are thrilled to announce that our school paper has qualified for the NATIONAL SCHOOLS PRESS CONFERENCE (NSPC) 2024. Thank you for your unwavering support and trust. Dive into the stories that made us stand out!
Analyzing the Social Media Footprint of Street GangsSanjaya Wijeratne
Abstract
Gangs utilize social media as a way to maintain threatening virtual presences, to communicate about their activities, and to intimidate others. Such usage has gained the attention of many justice service agencies that wish to create better crime prevention and judicial services. However, these agencies use analysis methods that are labor intensive and only lead to basic, qualitative data interpretations. This paper presents the architecture of a modern platform to discover the structure, function, and operation of gangs through the lens of social media. Preliminary analysis of social media posts shared in the greater Chicago, IL region demonstrate the platform's capability to understand gang members' social media usage patterns.
Link to Paper - http://knoesis.wright.edu/researchers/sanjaya/papers/2015/Wijeratne_ISI_2015.pdf
Social media research is an executable form of market research. The presentation covers how companies can implement a social media research process in their organization.
Twitter's Elections Integrity Datasets (Galvanize; February 22, 2019)Ivan Corneillet
Short presentation of Twitter's publicly available archives of Tweets and media that the company believed resulted from potentially state-backed information operations on its service for the benefit of Galvanize's Data Science Immersive Fellows on February 22, 2019 in San Francisco.
Advertise on Instagram: Content Creation GuidelinesIshan Gandhi
Looking to advertise on Instagram? This presentation would provide valuable inputs for your social media marketing campaign in India.
Following are some of the very interesting reasons why you should choose to advertise on Instagram:
1. Over ⅓ of Instagram Users Have Used The App To Purchase A Product Online.
2. Instagram’s Ad Platform Has Access To The Most Advanced Social Media Advertising Targeting Options Available.
3. Sales and Leads Are Trackable Through Instagram Ads So You Can See Clear ROI.
4. Instagram Differentiates Personal Profiles from Business Profiles.
5. Instagram’s User Base Allows You To Tap Into An Otherwise Unreached Customer Base.
(Reasons by Lyfemarketing)
For any further assistance or query, you can contact me at- ishan711gandhi@gmail.com
Social Websites And Seo Social Dev Camp Chicago2008 By John FairleyJohn Fairley
SocialDevCampChicago 2008 Comparing Search Engine Optimization of social media websites, including digg, stumbleupon and twitter. Learn SEO best practices and pitfalls to avoid.
The Power Point to support my presentation on use of Social Media in higher education. NISOD Conference on Teaching and Leadership Excellence, May 30 2011
This is a presentation I delivered for a group of marketing managers from Ernst & Young on December 3, 2009. These managers are based in the western US and focus primarily on E&Y's Entrepreneur of the Year program.
The presentation covers the basics of social media, stats on leading tools and recommendations for ramping up social media activities individually and for the Entrepreneur of the Year team.
Similar to Analyzing the Social Media Footprint of Street Gangs (20)
Read| The latest issue of The Challenger is here! We are thrilled to announce that our school paper has qualified for the NATIONAL SCHOOLS PRESS CONFERENCE (NSPC) 2024. Thank you for your unwavering support and trust. Dive into the stories that made us stand out!
Unit 8 - Information and Communication Technology (Paper I).pdfThiyagu K
This slides describes the basic concepts of ICT, basics of Email, Emerging Technology and Digital Initiatives in Education. This presentations aligns with the UGC Paper I syllabus.
A review of the growth of the Israel Genealogy Research Association Database Collection for the last 12 months. Our collection is now passed the 3 million mark and still growing. See which archives have contributed the most. See the different types of records we have, and which years have had records added. You can also see what we have for the future.
Safalta Digital marketing institute in Noida, provide complete applications that encompass a huge range of virtual advertising and marketing additives, which includes search engine optimization, virtual communication advertising, pay-per-click on marketing, content material advertising, internet analytics, and greater. These university courses are designed for students who possess a comprehensive understanding of virtual marketing strategies and attributes.Safalta Digital Marketing Institute in Noida is a first choice for young individuals or students who are looking to start their careers in the field of digital advertising. The institute gives specialized courses designed and certification.
for beginners, providing thorough training in areas such as SEO, digital communication marketing, and PPC training in Noida. After finishing the program, students receive the certifications recognised by top different universitie, setting a strong foundation for a successful career in digital marketing.
Operation “Blue Star” is the only event in the history of Independent India where the state went into war with its own people. Even after about 40 years it is not clear if it was culmination of states anger over people of the region, a political game of power or start of dictatorial chapter in the democratic setup.
The people of Punjab felt alienated from main stream due to denial of their just demands during a long democratic struggle since independence. As it happen all over the word, it led to militant struggle with great loss of lives of military, police and civilian personnel. Killing of Indira Gandhi and massacre of innocent Sikhs in Delhi and other India cities was also associated with this movement.
Normal Labour/ Stages of Labour/ Mechanism of LabourWasim Ak
Normal labor is also termed spontaneous labor, defined as the natural physiological process through which the fetus, placenta, and membranes are expelled from the uterus through the birth canal at term (37 to 42 weeks
Macroeconomics- Movie Location
This will be used as part of your Personal Professional Portfolio once graded.
Objective:
Prepare a presentation or a paper using research, basic comparative analysis, data organization and application of economic information. You will make an informed assessment of an economic climate outside of the United States to accomplish an entertainment industry objective.
A workshop hosted by the South African Journal of Science aimed at postgraduate students and early career researchers with little or no experience in writing and publishing journal articles.
South African Journal of Science: Writing with integrity workshop (2024)
Analyzing the Social Media Footprint of Street Gangs
1. Analyzing the Social Media Footprint of
Street Gangs
Ohio Center of Excellence in Knowledge-enabled Computing (Kno.e.sis)
Wright State University, Dayton, OH, USA
Presented at the SRV 2000 – Engaged Citizenship (Youth Violence) Class at Wright State University
Dayton, OH, USA, March 06, 2018
Sanjaya Wijeratne
PhD Student @ Kno.e.sis Center
Department of Computer Science and Engineering
Wright State University, Dayton, OH
sanjaya@knoesis.org
3. Finding Street Gang Members on Twitter – Sanjaya Wijeratne 3
Tweets Source – http://www.wired.com/2013/09/gangs-of-social-media/all/
Image Source – http://www.7bucktees.com/shop/chi-raq-chiraq-version-2-t-shirt/
*Example tweets shown above are public data reported in the cited news paper article. They are not part of our research dataset.
4. 4
Image Source – National Gang Threat Assessment Report, 2015.
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
5. 5
What does gang related research tell us?
“Gangs use social media mainly to post videos depicting
their illegal behaviors, watch videos, threaten rival gangs
and their members, display firearms and money from
drug sales” [Patton 2015, Morselli 2013]
Studies have shown,
82% of gang members had used the Internet and 71%
of them had used social media [Decker 2011]
45% of gang members have participated in online
offensive activities and 8% of them have recruited new
members online [Pyrooz 2013]
Image Source – http://www.sciencenutshell.com/wp-content/uploads/2014/12/o-GANG-VIOLENCE-facebook.jpg
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
6. 6
Involvement of the Law Enforcement
Source – http://www.lexisnexis.com/risk/downloads/whitepaper/2014-social-media-use-in-law-enforcement.pdf
Image Source – http://www.officialpsds.com/images/stocks/crime-scene-stock1016.jpg
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
7. 7
But there’s a problem…
Image Source – http://www.officialpsds.com/images/stocks/crime-scene-stock1016.jpg
Source – http://www.lexisnexis.com/risk/downloads/whitepaper/2014-social-media-use-in-law-enforcement.pdf
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
8. 8
A Platform to Analyze Social Media
Monitor negative community effects of gang
activities
Discover opinion leaders who influence the
thoughts and actions of other gang members
Evaluate the sentiment of posts targeting
communities, locations, and groups
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
9. 9
Architecture of Our Platform
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
Wijeratne, Sanjaya et al. "Analyzing the Social Media Footprint of Street Gangs" in IEEE ISI 2015, doi: 10.1109/ISI.2015.7165945
10. 10
Data Collection
Gang Related Dataset Location Related Dataset
Data Collection 91 known gang member profiles
from a Chicago gang (Gang X) +
Gang related keywords (eg. #BDK,
#GDK etc.) [Decary-Hetu 2011]
10 neighborhoods in Chicago
- South Landale, North
Landale, West Elsdon, Gage
Park, West Lawn, Chicago
Lawn, New City, Humboldt
Part, Logan Square and
Belmont Cragin
APIs Twitter REST + Twitter Streaming Twitter Streaming
Size 56,368 + 49,079 = 105,447 383,656
Geo Information 57.07% 100%
Image Source – http://goo.gl/iCTDj4
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
11. 11
Data Analysis – Twitter Profiles
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
Chicago-based Gang Member Dataset Analysis
12. 12
Data Analysis – Tweets
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
Chicago-based Gang Member Dataset Analysis
13. 13
Data Analysis – Follower Network
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
Chicago-based Gang Member Dataset Analysis
14. 14
Data Analysis – Sentiment Analysis
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
Chicago-based Gang Member Dataset Analysis
15. 15
Finding Street Gang Members on Twitter
Image Source – http://www.sciencenutshell.com/wp-content/uploads/2014/12/o-GANG-VIOLENCE-facebook.jpg
Balasuriya, Lakshika et al. “Finding Street Gang Members on Twitter”, In IEEE/ACM ASONAM 2016 (To Appear)
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
16. 16
Data Collection
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
FreeDaGuys, RIPDaGuys,
FuckDaOpps etc.
Gang Member Dataset Creation
17. 17
Gang Member Dataset
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
Description Gang Member
Class
Non-gang
Member Class
Total
Number of Profiles 400 2,865 3,265
Number of Tweets 821,412 7,238,758 8,060,170
We collected 400 gang members profiles
Non-gang member profiles were collected from Twitter
Streaming and REST APIs
Filtered only US profiles – 2000 profiles collected
We added 865 manually verified non-gang member
profiles collected from four seed words
To capture language used by family/friends of gang members
18. 18
Data Analysis – Tweets
Words from Gang Member’s Tweets Words from Non-gang Member’s Tweets
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
19. 19
Data Analysis – Profile Descriptions
*Example Twitter profiles shown above are public data reported in news articles. They are not part of our research dataset.
Word usage in profile descriptions: Gang Vs. Non-gang members
Percentofprofilesthatcontainthewordinprofiledescription
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
20. 20
Data Analysis – Emoji
Emoji usage distribution: Gang Vs. Non-gang members
PercentofprofilesthatcontainstheemojiinTweets
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
21. 21
Data Analysis – Emoji Cont.
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
Emoji Chain Gang Member Usage Non-gang Member Usage
32.25% 1.14%
53% 1.71%
Emoji usage: Gang Vs. Non-gang members
22. 22
Data Analysis – YouTube Links
51.25% of gang members post at least one
YouTube video
76.58% of them are related to gangster hip-hop
A gang member shares 8 YouTube links on average
Top 5 words from gang members – shit, like,
nigga, fuck, lil
Top 5 words from non-gang members – like,
love, peopl, song, get
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
23. 23
Data Analysis – Profile Images
Image tags distribution: Gang Vs. Non-gang members
Percentofprofilesthatcontainstheimagetag
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
24. 24
Data Analysis – Profile Images Cont.
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
Gang member tags – trigger, bullet, worship
Non-gang member tags – beach, seashore, pet
Gang Members’ Twitter Profile and Cover Images
25. 25
Classification Approach
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
Used five feature types
Tweets, Profile Descriptions, Emojis, Profile and
cover images, YouTube videos
Converted all non textual features to text
Emoji for Python API for emoji1
Clarifai API for profile and cover images2
Removed stop word and stemmed the remaining
Represented Twitter profiles using unigrams
Trained four types of Classifiers (NB, LR, RF, SVM)
1https://pypi.python.org/pypi/emoji | 2http://www.clarifai.com/
27. 27
Our Classifier in Action
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
Twitter profiles collected from neighborhoods
that are known for gang-related activities
Los Angeles, CA – 15,662 Profiles
Southside of Chicago – 8,500 Profiles
Classified using Model(2) RF Classifier and
qualitatively studies the “Gang” class
Displayed weapons in their profile images
Gang-related names in their profile descriptions
Frequent use of curse words
Referred to known gang members as “my hommies”
28. 28
Our Classifier in Action Cont.
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
Most frequent words in Tweets
shit, nigga, got, bitch, go, fuck
Most frequent words in user profiles
free, artist, shit, fuck, freedagang, ripthefallen
Most frequent emoji –
29. 29
Enhance Twitter Gang Member Profile
Identification using Word Embeddings
Image Source – http://www.sciencenutshell.com/wp-content/uploads/2014/12/o-GANG-VIOLENCE-facebook.jpg
Wijeratne, Sanjaya et al. “Word Embeddings to Enhance Twitter Gang Member Profile Identification”, SML Workshop @ IJCAI, 2016
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
30. 30
Gang Member Dataset
Number of Words
from each Feature
Gang Member
Class
Non-gang Member
Class
Total
Tweets 3,825,092 45,213,027 49,038,119
Profile Descriptions 3,348 21,182 24,530
Emoji 732,712 3,685,669 4,418,381
YouTube Videos 554,857 10,459,235 11,041,092
Profile Images 10,162 73,252 83,414
Total 5,126,176 59,452,365 64,578,536
Description Gang Member
Class
Non-gang
Member Class
Total
Number of Profiles 400 2,865 3,265
Number of Tweets 821,412 7,238,758 8,060,170
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
31. 31
Classification Approach
Converted all non textual features to text
Emoji for Python API for emoji1
Clarifai API for profile and cover images2
Removed stop word and stemmed the remaining
Trained a Skip-gram model using textual features
Negative sample rate = 10
Context window size = 5
Ignore words with min count = 5
Number of dimensions = 300
1https://pypi.python.org/pypi/emoji | 2http://www.clarifai.com/
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
33. 33
Representing Training Examples
For each Twitter profile p, with n unique
words, and the vector of the ith word in p
denoted by wip, we compute the feature
vectors for Vp in the following ways.
Sum of word embeddings – Sum of word vectors for
all words in Twitter profile p
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
34. 34
Representing Training Examples Cont.
Mean of word embeddings – Mean of word vectors
for all words in Twitter profile p
Sum of word embeddings weighted by term
frequency where term frequency of ith word in p is
denoted by cip
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
35. 35
Representing Training Examples Cont.
Sum of word embeddings weighted by tf-idf where
tf-idf of ith word in p is denoted by tip
Average sum of word embeddings weighted by term
frequency where term frequency of ith word in p is
denoted by cip.
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
37. 37
Results Interpretation
Averaging the vector sum weighted by term
frequency performs the best
Out of the five vector based operations on
word embeddings;
Four of them gave classifiers that beat baseline
model(1)
Two of them performed as nearly equal or beat
baseline model(2)
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
38. 38
Results Interpretation Cont.
Word vector (top 10) for BDK showed
Five were rival gangs of Black Disciples
Two were syntactic variations of BDK
Three were syntactic variations of GDK
Word vector (top 10) for GDK showed
Six were Gangster Disciple gangs where others
have expressed their hatred towards them
Word vector (top 10) for CPDK showed
Other keywords showing hatred towards cops
such as fuck12, fuckthelaw, fuk12 and gang names
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
39. 39
Recap
Gang members are actively using Twitter
Discussed our approach to identify gang
members on Twitter (ASONAM 2016)
What do they post on Twitter and what types of
features could help to identify their profiles
Discussed how word embeddings can be used
to improve gang member Twitter profile
identification (SML@IJCAI 2016)
Looked at examples of interesting information that
word embeddings can tell us about gangs
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
40. 40
Challenges and Future work
Build image recognition systems that can
accurately identify images commonly posted by
gang members
Guns, Drugs, Money, Gang Signs
Integrate slang dictionaries and use them to pre-
train word embeddings
Hipwiki.com
Utilize social networks of gang members to
identify new gang members
Image Source – http://i.ytimg.com/vi/dqyYvIqjuFI/maxresdefault.jpg
Finding Street Gang Members on Twitter – Sanjaya Wijeratne
42. 42SML @ IJCAI 2016 Wijeratne, Sanjaya et al. Word Embeddings to Enhance Twitter Gang Member Profile Identification
43. References
[1] D. U. Patton, “Gang violence, crime, and substance use
on twitter: A snapshot of gang communications in
Detroit,” Society for Social Work and Research 19th
Annual Conference: The Social and Behavioral
Importance of Increased Longevity, Jan 2015
[2] C. Morselli and D. Decary-Hetu, “Crime facilitation
purposes of social networking sites: A review and
analysis of the cyber banging phenomenon,” Small
Wars & Insurgencies, vol. 24, no. 1, pp. 152–170, 2013
[3] S. Decker and D. Pyrooz, “Leaving the gang: Logging off
and moving on. Council on foreign relations,” 2011
43Finding Street Gang Members on Twitter – Sanjaya Wijeratne
44. References Cont.
[4] D. C. Pyrooz, S. H. Decker, and R. K. Moule Jr, “Criminal
and routine activities in online settings: Gangs,
offenders, and the internet,” Justice Quarterly, no.
ahead-of-print, pp. 1–29, 2013
[5] S. Wijeratne, D. Doran, A. Sheth, and J. L. Dustin,
“Analyzing the social media footprint of street gangs”,
In Proc. of IEEE ISI, 2015, pages 91–96, May 2015.
[6] L. Balasuriya, S. Wijeratne, D. Doran, and A. Sheth,
“Finding street gang members on twitter”, In The 2016
IEEE/ACM International Conference on Advances in
Social Networks Analysis and Mining (ASONAM 2016).
San Francisco, CA, USA; 2016 (To Appear).
44Finding Street Gang Members on Twitter – Sanjaya Wijeratne
45. References Cont.
[7] S. Wijeratne, L. Balasuriya, D. Doran, A. Sheth, “Word
Embeddings to Enhance Twitter Gang Member Profile
Identification”, In IJCAI Workshop on Semantic Machine
Learning (SML 2016). New York City, NY: CEUR-WS;
2016.
[8] Lakshika Balasuriya, Sanjaya Wijeratne, Derek Doran,
Amit Sheth. “Signals Revealing Street Gang Members
on Twitter”, In Workshop on Computational
Approaches to Social Modeling (ChASM 2016) co-
located with 8th International Conference on Social
Informatics (SocInfo 2016). Bellevue, WA, USA; 2016.
45Finding Street Gang Members on Twitter – Sanjaya Wijeratne