Corporate Surveillance in the Age of Digital Tracking, Big Data & Internet of Things

Video in TIB AV-Portal: Corporate Surveillance in the Age of Digital Tracking, Big Data & Internet of Things

Formal Metadata

Corporate Surveillance in the Age of Digital Tracking, Big Data & Internet of Things
Title of Series
Part Number
Number of Parts
CC Attribution - ShareAlike 3.0 Germany:
You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor and the work or content is shared also in adapted form only under the conditions of this license.
Release Date
Production Place

Content Metadata

Subject Area
Today virtually everything we do is monitored in some way. Nearly every device we use is connected to the Internet. Thousands of companies are analyzing our everyday behavior. Businesses are using this data to make predictions, to manage risk and to motivate behavioral change. To what extent do companies really track our daily lives in 2015? How is predictive analytics based on personal data already being used in the fields of insurance, banking and human resources? And what is to be done?
Population density Computer animation Meeting/Interview Lecture/Conference Thermal radiation Entropie <Informationstheorie>
Web page Trail Mobile app Context awareness Service (economics) Software developer Projective plane Interactive television Analytic set Expected value Web 2.0 Facebook Voting Lecture/Conference Business model Online game Table (information) Service-oriented architecture Traffic reporting Computing platform Resultant
Statistics Divisor Endliche Modelltheorie
Computer animation Agreeableness Endliche Modelltheorie Open set Dimensional analysis
System call Numerical analysis Attractor
Message passing Theory of relativity Chemical equation 1 (number) Correlation and dependence Metadata
Arithmetic mean Message passing Different (Kate Ryan album) Internetworking Agreeableness Gradient Videoconferencing Musical ensemble Open set System call
Predictability Wave Lecture/Conference Smartphone Endliche Modelltheorie Login
Facebook Lecture/Conference Profil (magazine) Online help
Predictability Slide rule Presentation of a group Root Lecture/Conference Radio-frequency identification Video game Analytic set Open set Web browser Leak Metadata
Data mining Sign (mathematics) Theory of relativity Lecture/Conference State of matter Multiplication sign Bit System call Theory Row (database) Attribute grammar
Category of being Mathematics Algorithm Machine learning Cross-correlation Meeting/Interview Mathematical analysis Virtual machine
Web page Service (economics) Multiplication sign Coma Berenices Web browser Web 2.0 Mathematics Meeting/Interview Profil (magazine) Calculation Formal grammar Website Software testing Extension (kinesiology) Resultant Compilation album
Web 2.0 Message passing Context awareness Different (Kate Ryan album) Internet service provider Software developer Electronic mailing list System call Power (physics) Neuroinformatik Product (business)
Mobile Web Mobile app Service (economics) Information Software developer Orientation (vector space) Analytic set Attribute grammar Formal language Tablet computer Category of being Lecture/Conference Smartphone Computing platform Physical system
Predictability Category of being Facebook Algorithm Observational study Gender Orientation (vector space) View (database) Website Resultant Descriptive statistics Product (business)
Point (geometry) Lecture/Conference Smartphone Disk read-and-write head Family
Trail State of matter Weight Sheaf (mathematics) Fitness function Insertion loss Food energy Twitter Arithmetic mean Hypermedia Single-precision floating-point format Formal verification Video game Website Traffic reporting Row (database)
Pay television Lecture/Conference Flow separation Discounts and allowances
Computer programming Lecture/Conference Internetworking Business model Fitness function 1 (number) Smartphone Internet der Dinge Object (grammar) Thermische Zustandsgleichung Connected space
Latent heat Expert system Device driver Internet der Dinge Spacetime
Computer programming Pay television Lecture/Conference Discounts and allowances
Purchasing Information Range (statistics) Similarity (geometry) Database Mass Mereology Data mining Videoconferencing YouTube Physical system Asynchronous Transfer Mode Social class
Lecture/Conference Confidence interval Decision theory Website Analytic set Likelihood function
Complex (psychology) Pulse (signal processing) Algorithm Presentation of a group Digitizing Decision theory Chemical equation Field (computer science) Product (business) Vector space Software Lecture/Conference Hypermedia Object (grammar)
Facebook Goodness of fit Presentation of a group Lecture/Conference Internetworking Office suite Client (computing) Service-oriented architecture Axiom Connected space Twitter
Purchasing Identifiability Matching (graph theory) Computer file Multiplication sign Information privacy Facebook Process (computing) Lecture/Conference Video game Smartphone Data logger HTTP cookie Computing platform
Purchasing Predictability Dataflow Implementation Key (cryptography) Robot Source code Interactive television Combinational logic Database Mass Mereology Numerical analysis Product (business) Facebook Software Customer relationship management Video game Service-oriented architecture Address space Computing platform Identity management Physical system
Axiom of choice Coma Berenices Product (business) Connected space Facebook Mathematics Lecture/Conference Different (Kate Ryan album) Personal digital assistant Internetworking Calculation Gastropod shell Lie group
Axiom of choice Predictability Profil (magazine) Analogy Decision theory Customer relationship management Analytic set Pattern language Software testing Cartesian coordinate system
Laptop Area Predictability Multiplication Information Multiplication sign Moment (mathematics) Projective plane Content (media) Mass Frequency Mathematics Latent heat Process (computing) Bit rate Computer configuration Internetworking Term (mathematics) Cuboid Lipschitz-Stetigkeit Data conversion Traffic reporting
Game controller Group action Open source Regulator gene Connectivity (graph theory) Moment (mathematics) Information privacy Goodness of fit Mathematics Internetworking Software framework Metropolitan area network Informationsgesellschaft
Computer animation Meeting/Interview
while I'm in and
necks lot hybrid by these by name is what it is as you said and this is again see and a density about how thousands of companies are attracting more and more aspects of our everyday lives largely without our knowledge but it's not only governments was spying on us today we're constantly getting analyzed radiates by deal well in entropy of online
platforms and service app developers analytics companies data brokers and many more whose business models are based on the expectation of our personal data uh during the last few years I tried to find out how to beat this challenge my 1st try looked like that I was asking worried about a policy for getting it turn the tables and final daughter dirty details about your friends neighbors in the rest of the votes I co-created data dealer and online games even a Facebook going about collecting and selling personal data but maybe somebody heard of because still played online then lots of research but this should tracking data brokers in today's puzzle
data ecosystem of end of 2014 I published an extensive 100 pages research report about corporate so awareness individually age of recently I've been contributing to the do Not Track projects in interactive Web document very serious about dishes tracking in today asking about some of my research results of 1st of all
I'm cues who in the audience has ever been to psychotherapy mental illness and so on come on don't be shy you've got nothing to hide according this to the
statistics you should be more with no problem will get that done ever heard of 2 big 5 the big 5 for the 5 factor model it is 1 of the most popular possibility Molly models in psychology according to this every human being can be described using
5 main dimensions Neuroticism neurotic neurotic people
or nervous emotionally unstable afraid of everything extroversion openness to experience and conscientiousness of and agreeableness is about cooperation and social harmony the five-factor model is well established and has been used in research and more than 3 thousand academic papers as as many years now a research
team from Switzerland experience experiment we asked 83rd of 3 persons to fill out this so
way and we use the classical empirical approach to calculate the values of each of these 5 personality traits for example like that in a 2nd step they asked all of them to instill in at on the phone which collected some data about the phone usage it recorded the number of users of each of these steps during 8 months of the attractor number and duration of incoming outgoing in these calls uh the
same for short messages metadata not content and then they analyze all this data and try to find a balance the tried to calculate the big 5 personality traits out of this phone log data
and there was successful the phone statistical correlations not very strong ones but still relations for example they found out people schools used off
Kalinda apps more often are other neurotic let's open for new experiences and when it comes to the agreeableness also grade bad heavy Internet users on the fold less extroverted and US open watching videos and listening to music not very consciousness but that doesn't mean that it's always like that but it's a higher probability people receiving many incoming calls less neurotic and they're probably extroverts uh the longer the duration of somebody's incoming calls the more extroverted opened agreeable but many missed calls lack of openness and finally when somebody's calling lots of different people but it seems like this person could be super social pavement is somebody professor sent messages instead of calling sounds like both at agree ability bad conscientiousness
OK now you could tell me yes we already knew most of that but that
that's not what this is about uh is about Big Data whoever is able to access this phone logs can do this predictions for millions of phone users finally when they try to be predict someone's personality just basal smartphone
mere data for example somebody lower higher in Europe it is if you would guess you get a prediction accuracy 50 % of but if the use of the prediction model based phone data that did quite a better on the but a wave I'm totally open agreeable extroverted and neurotic uh well just joking but then finally found out the
truth about me with the help of 5 let's they're analyzing a Facebook postings to
predict your personality and the personality of your friends check it out if you want to look at your own risk of the also calculated Big Five personality profiles for Symbian pieces that marked abrupt 82
% Nordicism know barbecue
pit to me whereas only 39 % of openness yeah shocking but OK it's based on just 150 roots but there are also other parties interested in the Big Five this is a present this presentation
slide from the British intelligence agency from the stone leaks it seems that here they're thinking about how the big 5 personality traits correlated in web browsers used for example they found out the Firefox users could be more neurotic what it did now lies in this in the research paper mentioned is who is calling whom uh who do you know where your friends social network analytics also offers powerful ways to make predictions about people this is also just metadata metadata is powerful or like the former CIA Director said we kill people based on metadata but or listen to his home in from the NSA he said metadata absolutely tells you everything about somebody's life if you have enough metadata you don't really need constant ok this is not so much about corporate so Wayland's
so as to make it a bit less dramatic is this is signi theorist signifier of and yes started calculating credit scores from mobile phone call records this
states coarser created from behave real
time related attributes no
traditional payment date data requirements and 4 weeks of calling history is not so what is this all about it's all about data mining using methods like cluster
analysis and so on involving methods from mathematics is that this 6 to machine learning machine learning algorithms that try to learn from existing data that gets trained to find correlations in large datasets and they're able to find where human beings with their traditional statistical approaches really give up
for example there's this company founded by former
boolean pretty called test finances again
about credit scoring as they say this is the math we all learned to the page was important in this search results for what
was on it but also for how good a grammar wars or which foster used or when it was edited the created everything now just finances is doing the same for credit scoring they're combining 70 thousand signals to calculate personal credit scores and they say all data is Craig data we just don't know how to use it yet all data just lots of data about us out there but some of you probably know this is like being a browser extension uh which shows who is watching us when we surfing the web so here was it it's the 5 websites and you've times Caribbean BuzzFeed was to dictionary . com in the background these 5 Sec those sites connected to 104 other third-party services and told them about my visit this is how they are able to compile profiles about online behavior as there are tens of
thousands of tracking companies doing this stuff but it's not just surfing the web does all also that special URI somebody ready mentioned are smart
phone is a powerful small computer with plenty of different sizes under the hood even if it's a big problem like this 1 but
also contains lists of context of friends of our calls of messages to track some movements lots of very product data now who is able to access the sale more with four-letter expect from providers like the and that developers
but there are also many hidden third-party services included in many apps for example flaring mobile analytics platform they're basically helping app
developers of during money and additionally there monitoring the behavior of
users in more than 500 thousand apps on 1 . 4 billion smartphones and tablets the excessive attributes that generates language kind of the wise subtracting system and so on and they're also sorting users into pedigrees like of gamers for national peace new mothers or even categories based on their sexual orientation but wait house very able to offer information about the sexual orientation of people did tracking maybe
they using similar algorithms than the researchers of descriptive study uh they try to predict private at tributes of users just based on the Facebook likes and these were the results they were able to successfully predict gender or sexual orientation political and religious views but just based on about 100 70 Facebook likes per user and are like a special category of data very similar to the web sites we visited the products we bought and so on reaction we use
of by the way in 2014 uh they established a partnership with a market
research and consumer data company and now they are offering an additional 300 offline data points about family background income and so on OK
smartphones and then some people started uh this special kind of the movements and years ago they called it the quantified self this guy seems to be 1 of the early
adopters of not everybody in lost has got this kind of tactic is on the head of others of for example
using BodyMedia my GP has to have the life of the really fancy slogans for advertising I love it that these devices the tracking rectivity years steps the calories heartbeats and so on and this is an
example report very important you can enter target for example 10 thousand steps and then you and try to reach that verification so it is see this person is weight loss trend great but only 4 hours of sleep on the top so good however that wonderful at from body media I love that had records everything from sunrise to sunset and beyond uh OK this absolutely beautiful with school maybe uh because they're from 2012 to the there using really state of the art eds but this is a no means indicat acquired patrol bomb bone of 1 of 2 market leaders and if you the fitness tracker single market leaders Fitbit very stylish uh but if you moved to their corporate when a section on the website you can see the
fault that already the partnerships with several big companies for example with that will accompany pp the employees control choose to join the Fitbit corporate when this program that if they reached 1 million steps in the year they cut the
discount on their health insurance premium of PP self interest so there and please don't have any other health insurance according to Bloomberg 1 of the employees cut the discount of 1 thousand
200 dollars on a standard insurance in 2013 when he reaches 1 million steps I mean he didn't have to join the program but it's a 1 thousand 200 dollars but this is not just a tiny little incentive for many people uh this amount would make it mandatory to participate and this is maybe but the famous Internet grew to really meant to uh when he was talking about you know the way that advertising turned out to be the native business model of the and the free internet I think that insurance of which is going to be the native business model for the Internet of Things and yes it's not only smartphones of fitness trackers in the upcoming Internet of Things
more and more objects with sensors a bit smaller ones than this but of the senses uh and after connection from the book readers that smart to be connected cars smart meters thermostats and much more the subject will be everywhere at
home adverse and in public spaces so it won't just be people tracking them south they would also have tracking other people for example the children or their employees of who could that carry some of the wisest with them or stay at places where sensors are installed many experts expect that the major driver of the Internet of Things will be incentives to get people to change their behavior maybe to purchase something or
to act in how the safe manner or married to verify in a specific way for example more quickly but during the last few
years many car insurance companies already started to offer policies which are based on tracking the driving behavior of people of they're constantly matched measuring the cost the patient and then they're MPEG Lydia score this is an example that if you drive carefully and you get a discount on your insurance premium and most of these programs will lower your score of if you're accelerating fast or breaking hired but wait no harsh breaking the
really but I think this shows that 1 of
many problems of incentive systems like this but of course is of only phone love Scott driving behavior health data being collected today virtually everything we do is recorded monitored detract in some way all kinds of collected data
ends up in beach classes of databases data mining technologies have to find the relevant information in these massive amounts of data but this 1 US company was probably collected the largest amounts of personal information about consumers on the planet is can't actually collected up to 3 thousand at tributes on 700 million people of for example they've got credit history driving history criminal history of purchase behavior mode party had rest and much more they run a U-tube channel full of really nice videos 2nd persist to show a short part pay attention how beautiful they use their acts at all so there's no sound but I use my microphone to its will they use the X yes the excellent is only 1 of a couple of major companies in because the data brokering another 1 would be lexisnexis Risk solutions also very nice name of it but the similar amount of data about consumers and a really impressive range of offers but they're not only setting the date on problem
printers there are also offering employment screening so make quicker confident pre and post hiring decisions that
they've also but offers for health care here this a social networks analytics reveal hidden relationships the even right somewhere on the site we help predict the likelihood that the consumer will become then increment in the next 18 months
object for example social intelligence they're promising 2 also to drive intelligent decisions based on data from social media the offering products in a freezer insurance employment and much more to more fancy examples in the field of the military digital complexes of vector politicist balance here a silicon Willie data mining company so there often ing offering products for companies in the fields of health care insurance and finance this software is based on paper pulse fraud detection algorithms and they've got partnerships with SAP uh the US Department of Defense and with the
CA or recorded future piles of left that name of they offering smart the businesses this should decisions with with intelligence and they're working for the US military and intelligence agency as well as for B corporations so this is the horror show of present data usage I would say but
also lexisnexis office data to governmental agencies that look at them as black helicopter the x is not so much into governmental clients just a tiny bit of but anyway originally axiom is in kind of always good data broker they started 40 years ago with sending personalized letters of but they're trying hard to connect to the digital built a
mini-exome presentation from 2012 have found on the internet they ask what if we could social data like to the desert in 2015 15 they've got partnerships with kobe Facebook and Twitter and many more like in many other data brokers the anomaly focusing on how to connect data from companies to on nonprofits or how to connect online and offline data about consumers the so-called data on boarding we want to make it possible to match
in-store purchases in the supermarket or and other shops to people serving the by using the smartphones or more specific to the devices that using 2 cookie identifier's or to user accounts on platforms like Facebook this is where they
bought another company called life rampant 2014 The for 300 million dollars but don't worry everything is happening totally secure animals they promise to use probe process that protects consumer privacy believe me but at the same time but 1 of the key features is 1 to 1 exact matching on your file which find the customer data records 5 that so on the 1 hand so that there is
our on the behavior across the wise and cross-platform on the other hand have the company has its customer database and its purchase data and noted trying to combine it how are they able to do its by matching linking and synchronizing our immune addresses phone numbers device ideas which could get ideas and user accounts for example or a Facebook user account of course they also use
third-party of consumer data from data brokers and this way businesses in all industries are able to identify and profile customers also based on the on their behavior and this is the the uh I believe where things get really creepy but because businesses in all industries starts it just use the data from marketing they also use it for the whole customer relationship management ripped flow their calculating things that customer values and scoring present in today's corporate wrote companies also love to implement risk management systems and the more data they get the better most consumers have no idea that many of the everyday life interactions are affected about how they are categorized and rated by companies take for example talent part to the so called predictive marketing platform of this slogan is turn our science into your says of the offering it had a bot identity network which incorporates massive amounts of data from both online and offline sources to create a tell about identity key for each and every shop and then they create a custom of score for every shopper and product combination as a result so that some Councillors get personalized off 1st but based on the
online and even often behavior and by the way a few days ago Twitterbots tell apart for about 500 million dollar of lies pricing could soon be everywhere because well on the shelves of already showed differently prize products of for different users so even the same products at different prices based on the on the behavior that a patient data or other devices the use in research paper from 2012 shells the proselytes prices differed up to 100 is 66 % Mikey 38 uh and US pop priors advocated the father of Reputation . com even said the rich the ready at different Internet than the poor based on the personal data so in some cases but we still have the choice for example and below the slogan is change alive today they use data from Facebook accounts to calculate scores but users uh and to ensure you are who you say you are uh and they're also
analyzing the connections to your friends so be sure to have value difference so you can
use lender to analog loans online shopping and employment opportunities and there's and then there's this nice faced of European patterns In the case of
Rolando you have 2 choices to participate or not participate maybe except also if there is no other loan you can get or if it landed looking is required
for that specific top applications they're all using such wonderful in the sentence scoring risk management predictive analytics make better decisions but what they really mean is so it's persuasive and omnipresence away lands based dishes tracking and our profiles In a 2nd step they offering us incentives followed up testing
different options again and again which is an incentive is more successful they call
increasing engagement improving conversion rates optimizing business processes multi waiting period change but what defeat if I use a very old fashioned term and colleagues manipulation so today it's just about as a experience entirely on the advertising of 1 person is waiting longer on the phone hotline because the be venue prediction for this person is too crappy or that specific projects and on shoppers out of stock but only for you because you were predicted to be 2 proper to less profitable too risky in future similar things could happen in much more areas of our lives so was a hard and painful choppers some 30 years ago of but times are changing fast I think really really really have to be careful in a future society where additional monitoring is omnipresent could have massive impact on people's lives options and behaviors we really have to be careful that information age doesn't end up in a kind of Digital technocracy that's why working on these issues and that's where published his report last November unfortunate reason in German only but I hope that I will be able to fund an English Russian soon you can download it cracked lips of the park but that also and there's also at least a document contained in containing the contents and references in English so what has to be done 1 option could look like that this is not my laptop no I've been using the Internet since the early 19 nineties so I know that at the same time information technology is great box in the moment it's many companies and governments who were excess in data it's not balance and all most of this is happening in a largely non-transparent way uh is a quick summary of basic policy recommendations concerning corporate data collection which resulted from a research report I think the most urgent challenge is
to make corporate data collection more transparent of after that's a never ending story of 2 European Data Protection reform I think that we needed goods so to be good and we need soon but even if we get the quality data protection regulation refractors this won't be enough in the moments companies are only in control of our personal data they're even trying to shape our future information society and they've got billions and billions of dollars but I think we need much more support of decentralized privacy where technology a Maven EVA become even a completely new industrial policy and billions for open source components of frameworks supporting another kind of innovation which is respecting our privacy and then there is this large group group of people who committed to do this kind of another Internet of I think it's proved crucial to make digital uh serious society much stronger of answer we need much more digital literacy uh and uh so this is the end of of the big chill man Eric Schmidt said so you have to fight for your privacy so you know you would use its I was asking friendly advisors are a serious threat I don't know they in England Denver
1 question will the changes in the and they like to so I finished too fast now
could have 1 minute is there anybody and OK