PROFILING THE PROFILERS

SOUSVEILLANCE

OF BIG TECH COMPANIES

USING THEIR OWN TOOLS

[DISNOVATION.ORG 2018-2019]

KEYWORDS

SURVEILLANCE CAPITALISM, GLITCH CAPITALISM, SOUSVEILLANCE, COUNTER-PROFILING, USER PROFILING, DATA MINING, BIG DATA ANALYTICS, ATTENTION ECONOMY, PAGERANK, GOOGLE MATRIX, MARKOV CHAIN, BLACK-BOX ALGORITHMS, TRUST IN DATA, FAKE NEWS, POST-TRUTH, WISDOM OF THE CROWD, SEMANTIC WEB

SUMMARY

0. ABOUT PROFILING THE PROFILERS
1. FROM SURVEILLANCE CAPITALISM TO GLITCH CAPITALISM

2. USER PROFILING

3. COUNTER PROFILING

4. TARGETING BIG TECH COMPANIES
5. ANNEX

6. AFTERWORDS

0. ABOUT PROFILING THE PROFILERS

BASED ON STATE OF THE ART BIG DATA ANALYTICS TECHNIQUES, THIS WORK GENERATES A SERIES OF HIGHLY DETAILED DIGITAL PROFILES OF BIG TECH COMPANIES — IE. PSYCHOLOGICAL, CULTURAL AND POLITICAL PROFILES — SIMILAR TO THE ONES CONSTANTLY GENERATED FOR EACH USER BY THESE VERY SAME COMPANIES. IN OTHER WORDS, THIS ALGORITHM PERFORMS A SOUSVEILLANCE OF BIG TECH COMPANIES USING THEIR OWN TOOLS. LINK TO THE PROJECT: TO BE RELEASED SOON

PROFILING THE PROFILERS — A project by DISNOVATION.ORG 2018-2019. Programming Jerome Saint-Clair. With the scientific support of Dr. José Lages, Institut UTINAM, France. The work was realised within the framework of the European Media Art Platforms EMARE program at m-cult with support of the Creative Europe Culture Programme of the European Union. In partnership with NRW-Forum Düsseldorf (DE) and MU Artspace Eindhoven (NL).

1. FROM SURVEILLANCE CAPITALISM
TO GLITCH CAPITALISM

TODAY’S INTERNET PLATFORMS ARE OVERLY FINANCED BY ADVERTISEMENTS.

WITH 80% OF GLOBAL PERSONAL DATA WHICH THEY OFFER TO ADVERTISERS AND THIRD PARTIES UNDER UNREGULATED POLICIES, THE MAJOR BIG TECH COMPANIES (IE. GOOGLE, AMAZON, APPLE, FACEBOOK AND MICROSOFT) BECAME PIVOTAL ACTORS IN THE SHIFT TOWARDS THE ERA OF DIGITAL POST-TRUTH AND THE INFLUENCE INDUSTRY: MICRO-PROFILING, DISINFORMATION CAMPAIGNS, BOTS, ILLEGAL DATA HARVESTING, TROLL FARMS, PERSUASIVE DESIGN.

THIS SHIFT, CONTEMPORARY TO THE EMERGENCE OF THE WEB 2.0, HAS BEEN ANALYZED AND PROBLEMATIZED NOTABLY BY SHOSHANA ZUBOFF. SURVEILLANCE CAPITALISM IS A “RADICALLY DISEMBEDDED AND EXTRACTIVE VARIANT OF INFORMATION CAPITALISM” BASED ON THE COMMODIFICATION OF “REALITY” AND ITS TRANSFORMATION INTO BEHAVIORAL DATA FOR ANALYSIS AND SALES.

THE EXTENT OF THE PROCESS OF AUTOMATIZATION AND SURVEILLANCE IS SUCH, THAT OUR SOCIETY HAS BECOME AN OPEN LABORATORY FOR CAPITALIST OPTIMIZATION. IN APRIL 2018, MALCOLM HARRIS WROTE ABOUT THE CONCEPT OF “GLITCH CAPITALISM” FOR THE INTELLIGENCER. THE WHOLE SILICON VALLEY ETHOS OF “MOVE FAST, BREAK THINGS” IS ESSENTIALLY AN ENDORSEMENT OF THE GLITCH AS A MODE OF PRODUCTION. AMERICA LOOKS LIKE A GLITCHY COMPUTER, AND IT’S BECAUSE CAPITALISM IS A MACHINE LANGUAGE, REDUCIBLE TO NUMBERS. AMERICA EXISTS TO CREATE WEALTH, AND THE SYSTEM ISN’T BROKEN, IT’S JUST OBEYING THE RULES TO DISASTER; AS A COUNTRY, WE’RE MORE OURSELVES THAN EVER.

2. USER PROFILING

DIGITAL PROFILING IS THE PROCESS OF GATHERING AND ANALYZING INFORMATION ABOUT AN INDIVIDUAL THAT EXISTS ONLINE. THIS IS THE USE OF ALGORITHMS OR OTHER MATHEMATICAL TECHNIQUES THAT ALLOW THE DISCOVERY OF PATTERNS OR CORRELATIONS IN LARGE QUANTITIES OF DATA, AGGREGATED IN DATABASES. A DIGITAL PROFILE CAN INCLUDE INFORMATION ABOUT PERSONAL CHARACTERISTICS, BEHAVIORS, AFFILIATIONS, CONNECTIONS AND INTERACTIONS.

Video: Facebook ads categories (US 2018)

EASILY ACCESSIBLE DIGITAL RECORDS OF BEHAVIOR (IE. FACEBOOK LIKES) CAN BE USED TO AUTOMATICALLY AND QUITE ACCURATELY PREDICT A RANGE OF HIGHLY SENSITIVE PERSONAL ATTRIBUTES INCLUDING: SEXUAL ORIENTATION, ETHNICITY, RELIGIOUS AND POLITICAL VIEWS, PERSONALITY TRAITS, INTELLIGENCE, HAPPINESS, USE OF ADDICTIVE SUBSTANCES, PARENTAL SEPARATION, AGE, AND GENDER.

THE INFORMATION ENVIRONMENT HAS BECOME THE NEW MODERN BATTLEFIELD WHERE STATE AND NON-STATE ACTORS EMPLOY SOPHISTICATED TECHNIQUES FOR TARGETING, PROPAGANDA AND DISINFORMATION (DARK ADVERTISEMENTS, NUDGING, ALGORITHMIC BIASES, SOCIAL BOTS, SOCKPUPPETS, BLACK PROPAGANDA, CLICK FARMS…).

Private traits and attributes are predictable from digital records of human behavior, Michal Kosinski, David Stillwell, and Thore Graepel, Edited by Kenneth Wachter, University of California, Berkeley, CA, and approved February 12, 2013 (received for review October 29, 2012)

3. COUNTER PROFILING

AS A RESPONSE TO THIS INFORMATION ASYMMETRY, WE SEIZED THE MEANS OF DATA ANALYTICS TO CREATE A SERIES OF PSYCHOLOGICAL, CULTURAL AND POLITICAL PROFILES OF THE MOST DATA-EXTRACTIVIST BIG TECH COMPANIES OF OUR TIME: GAFAM, NATU, BATX, AND OTHERS.

TO DO SO, WE WORKED FOR A YEAR WITH DR. JOSÉ LAGES AND HIS RESEARCH TEAM FROM INSTITUT UTINAM, BESANÇON, FRANCE. IN ORDER TO “INFER HIDDEN CAUSAL RELATIONS” BETWEEN BIG TECH COMPANIES AND SPECIFIC SOCIETAL AND POLITICAL ISSUES, WE ARE USING AN ALGORITHMIC METHOD DERIVED FROM PAGERANK (REDUCED GOOGLE MATRIX ANALYSIS) TO ANALYSE THE MATRIX OF EVERY POSSIBLE LINK BETWEEN EVERY SINGLE WIKIPEDIA ARTICLE.

SIMILAR ALGORITHMIC METHODS ARE OFTEN USED IN DATA SCIENCE, DATA JOURNALISM, AND FOR PROBABILISTIC USER PROFILING. IT ALLOWS TO ESTIMATE THE STRENGTH OF THE HIDDEN RELATIONS BETWEEN VARIOUS MEMBERS (ARTICLES, PAGES, USERS) OF THE STUDIED NETWORK (FOR INSTANCE BETWEEN A USER AND AN ITEM FOR THE PURPOSE OF PRODUCT RECOMMENDATION).

THESE AUTOMATED ACTIONS RESULT IN A SERIES OF HIGHLY DETAILED, AND BIASED, DIGITAL PROFILES OF BIG TECH, SIMILAR TO THE ONES CONSTANTLY GENERATED FOR EACH USER BY THESE VERY SAME COMPANIES.

Larry Page talks about  the PageRank algorithm, in Larry Page and Sergey Brin interview on Starting Google, 2000 (video excerpt)

Left: Google matrix of Wikipedia articles network, written in the bases of PageRank index; fragment of top 200 X 200 matrix elements is shown. — Right: Google matrix of Cambridge University network (2006), coarse-grained matrix elements are written in the bases of PageRank index.

4. TARGETING BIG TECH COMPANIES

GAFAM (GOOGLE, APPLE, FACEBOOK, AMAZON, MICROSOFT)

NATU (NETFLIX, AIRBNB, TESLA, UBER)

BATX (BAIDU, ALIBABA, TENCENT, XIAOMI)...

RATHER THAN SIMPLY FOLLOWING THE SAME CATEGORIES AS THE ONES USUALLY TRACKED FOR THE PROFILING AND PREDICTION OF USERS’ ACTIVITY (AGE GROUP, DEMOGRAPHIC, CONSUMER BEHAVIOUR, LOCATION, INCOME GROUP, ETC), WE AUGMENTED THESE CATEGORIES WITH ADDITIONAL CRITICAL INSIGHTS, SPECIFICALLY RELEVANT FOR BIG TECH (POLITICAL ORIENTATION, ETHICAL ORIENTATION, PROPAGANDA TECHNIQUES, TYPE OF INDUCED ADDICTIONS, TYPES OF BIASES, ETC).

Types of users’ data and metrics targeted by Facebook, Google... & types of Big Tech’s data and metrics targeted by Disnovation

THIS COUNTER-PROFILING DATA WILL BE CONTINUOUSLY RELEASED ON A DEDICATED PLATFORM AS NOTIFICATIONS, OPTIMIZED FOR SOCIAL MEDIA SHARING BY EACH VISITOR.

THIS WILL RESULT IN A DISTRIBUTED COUNTER-PROPAGANDA CAMPAIGN, EVENTUALLY POLLUTING THE SOCIAL FEEDS OF BIG TECH COMPANIES.

Left: OCEAN Personality — Right: Extraction of hidden correlations from within Wikipedia using ‘Google Matrix’

5. ANNEX

SELECTION OF SCIENTIFIC STUDIES THAT APPLIED GOOGLE MATRIX TO DATAMINE IN VARIOUS FIELDS:

5.1  CAPTURING THE INFLUENCE OF GEOPOLITICAL TIES FROM WIKIPEDIA WITH REDUCED GOOGLE MATRIX [PAPER]
SAMER EL ZANT, KATIA JAFFRES-RUNSER, DIMA L. SHEPELYANSKY

Position of EU countries in the local (K, K*) plane. EnWiki (left), FrWiki (middle) and DeWiki (right) networks. Countries are marked by their flags.

5.2  INFERRING HIDDEN CAUSAL RELATIONS BETWEEN PATHWAY MEMBERS USING REDUCED GOOGLE MATRIX OF DIRECTED BIOLOGICAL NETWORKS [PAPER]

JOSÉ LAGES, DIMA L. SHEPELYANSKY, ANDREI ZINOVYEV

AKT-mTOR pathway reconstructed using SIGNOR database and by inferring indirect connections using reduced Google matrix approach.

5.3  INTERACTIONS OF CULTURES AND TOP PEOPLE OF WIKIPEDIA FROM RANKING OF 24 LANGUAGE EDITIONS [PAPER]
YOUNG-HO EOM, PABLO ARAGÓN, DAVID LANIADO, ANDREAS KALTENBRUNNER, SEBASTIANO VIGNA, DIMA L. SHEPELYANSKY

List of global historical figures by PageRank and 2DRank for all 24 Wikipedia editions. All names are represented by the corresponding article titles in the English Wikipedia. Here, ΘA is the ranking score of algorithm A (3); NA is the number of appearances of a given person in the top 100 rank for all editions.

6. AFTERWORD

ACCORDING TO YOUR TASTE � BASED ON YOUR BROWSING HISTORY � BASED ON YOUR BROWSING PURCHASE � BASED ON YOUR HISTORY � BASED ON YOUR INTEREST IN � BASED ON YOUR RECENT VISIT � BASED ON YOUR VIEWING HISTORY � (BECAUSE YOU PURCHASED) RECOMMENDED FOR YOU � COMPANION PRODUCTS � CUSTOMERS ALSO BOUGHT � CUSTOMERS ALSO VIEWED � CUSTOMERS WHO PURCHASED � ALSO ENJOYED �  FIRSTNAME, TRY � HAVE YOU SEEN THESE YET? � � � IF YOU LIKE THIS, TRY THESE � � � IF YOU LIKE THIS, YOU’LL LOVE � INSPIRED BY YOUR BROWSING HISTORY � INVESTMENT RECOMMENDATION � JUST FOR YOU � MAKE IT EXTRA SPECIAL! ADD � MORE FANCY THINGS � � � MORE GREAT ITEMS � MORE ITEMS LIKE � MORE LIKE � NEARBY GROUPS � OTHERS LIKED THESE SIMILAR ITEMS � � � PEOPLE ALSO BOUGHT � PERSONALIZED PRODUCT RECOMMENDATIONS � PERSONALIZED RECOMMENDATIONS � � � RECOMMANDATION, JUST FOR YOU! � RECOMMENDATIONS FOR YOU � RECOMMENDED FOR YOU: � RECOMMENDED VIDEOS FOR YOU � RELATED ITEMS � RELATED PRODUCTS � � � SIMILAR PRODUCTS � � � SPONSORED CONTENT � SUGGESTED FOR YOU � SUGGESTED GROUP � SUGGESTED POST � THIS CAUGHT YOUR EYE: � YOU MIGHT ALSO LIKE THESE � TRY FOR FREE � WE RECOMMEND � WE SUGGEST THESE ADDITIONAL ACCESSORIES � YOU MAY ALSO LIKE � YOU MAY ALSO NEED ONE OF THESE ITEMS � YOU MAY BE INTERESTED IN � YOU MAY LIKE � YOU MIGHT BE INTERESTED IN THESE ITEMS � � � YOUR FRIENDS ALSO LIKED � YOUR RECENT PURCHASES ARE SIMILAR TO � YOUR TASTE PREFERENCES CREATED THIS ROW � � �