Download Data Ethics, Privacy, and Consent in Business Intelligence: A Comprehensive Guide and more Lecture notes Business in PDF only on Docsity!
BI
BUSINESS INTELLIGENCE
INFS 2036
Business Intelligence
We e k 7
P R I VA C Y, E T H I C S ,
L E G A L I S S U E S + T R U S T
Course Outline
The Why
of BI
The How of BI
Emerging
BI Concepts
Future Analytics
Week 1 Why data is important to business + The BI Process Week 10 Week 2 Front-End: Data Visualisation Week 3 Back-End: Organisation Information Systems Week 4 From Data to Intelligence Week 5 Lifecycle model and project management Week 6 Analytics - predicting the future & performance management Week 7 Privacy, Ethics, Legal Issues, Trust Week 8 Mining Technologies Week 9 Data Integration at an Enterprise and Cross-Organisation Level + Strategic value of information in the future You’re the CEO (Exam Review) INFS 2036 2
Week 7 Reading/Viewing Textbook
- Chapter 8: 8.5 Issues of Legality, Privacy and Access (pages 474-479)
- Chapter 4: 4.7. Data Mining Privacy Issues, Myths, and Blunders (pages 263-264) Additional reading/viewing – see Workshop Notes page (course site)
- McKinsey – Why Data Culture Matters: https://www.mckinsey.com/business-functions/mckinsey-analytics/our- insights/why-data-culture-matters
- IBM - Transparency and Trust in the Cognitive Era: https://www.ibm.com/blogs/think/2017/01/ibm-cognitive- principles/
D ATA P R I V A C Y, E T H I C S , L E G A L I S S U E S + T R U S T
INFS 2036 3
Focus On … Privacy + Ethics + International Legal Implications
- Putting it Into Practice … during the entire workshop + COVID-19 Mobility Tracker
- What’s coming next… Key Points + Workshop Activities
- Ethics, Privacy, Consent + Trust:
- Ethics Toolkit
- QR-Code Check-ins, RayBan x Facebook
- Cambridge Analytica
- The role of consent and trust in creating a
data-driven culture
- Legal: International Privacy Acts Topic Presentation + Practice W h a t w e ’ r e d o i n g t h i s w e e k Agenda – Week 7 THEORY PRACTICE This Week’s Case Study Ethics and privacy about tracking via your phone Accreditation Requirement INFS 2036 4
Why Ethics, Privacy, Consent and Trust? The number of trackers and ads blocked by my browser … in 4 months. INFS 2036^ - B. Chiera^7
Why Ethics, Privacy, Consent and Trust? If it’s for free then you are the product. INFS 2036 8
Why Ethics, Privacy, Consent and Trust? INFS 2036 9 The dark side of your data One prospective buyer claims to have been quoted $ 5 , 000 for the data of 1 million Facebook user accounts. The data provided contains the following personal information of Facebook users:
- Name
- Email
- Location
- Gender
- Phone number
- User ID https://www.privacyaffairs.com/dark-web- price-index-2021/
If it’s not for free then you are not the product. Why Ethics, Privacy, Consent and Trust? INFS 2036 10
- Data ethics also applies to what is collected and shared.
- A more ethical approach to data can lead to increased access to data. This happens when you can trust ethical codes and those who apply them.
- Applying good data ethics practices means you will have addressed people’s fears about how data will be collected, maintained and used. They are more likely to share data, which will support effective decision making.
- Relates to the collection and use of personal and non-personal data
- Complying with regulations e.g. Australian Privacy Laws, GDPR, is one part of data ethics.
- Data-related activity can be unethical but lawful.
- E.g. Facebook secretly manipulated the emotions of nearly 700,000 users.
- The research was not unlawful however people objected to the research on ethical grounds.
- Data ethics only concerns personal data
- Data ethics is solely concerned with data protection compliance Understanding Data Ethics https://missiondrive.typeform.com/to/xFIi4l
- Data ethics is only about how data is used.
- Data ethics is solely about restricting data access.
- Data ethics can help you make more informed decisions. INFS 2036 13
I n - C l a s s D i s c u s s i o n Ethics vs Privacy THEORY PRACTICE RayBan x Facebook smart glasses for videos, photos + music Five Eyes (+ 1) social media data sharing Police access to QR code check-in data Apple auto-scanning your iPhone and reporting harmful content Pros vs Cons Ethics vs Privacy? What would YOU do?
ACTIVITY
INFS 2036 14
W H AT W E C A N D O
Ethics Toolkit R o y a l C o l l e g e o f A r t ( U K ) “Automated decisions are part of many services we use everyday, but how they work is rarely explained or understood. This matters because automated decisions have an impact at scale. These decisions made by automated systems should be transparent, explainable and accountable.” RCA Design + Data Series Ethical Reasoning
- Checklist and/or a set of questions
- Aim: do no harm Ethics Framework
- A formalised document or process
- Achieve a holistic view of ethical practices Ethical Practices
- Intentionally exploiting users
- Being aware of the hidden traps INFS 2036 15
- Personal and organisational values a major factor in ethical decision making
- Ethical issues can be complex because of many different factors
- Different roles involved: employees, customers, managers
- People with different backgrounds: ethnic, religious etc.
- Tension between getting the job done with less and/or correct data versus taking the time to do it correctly, but causing delays (short-term vs long-term trade off)
T H E F O U R F U N D A M E N TA L Q U E S T I O N S
Applying Ethical Reasoning
- Who is/are the person(s) involved?
- What action was actually taken or is being contemplated?
- What are the results or consequences of the act?
- Is the result fair or just for all stakeholders? Ethical Reasoning
- Checklist and/or a set of questions
- Aim: do no harm INFS 2036 16
Target can buy data points about Your ethnicity Your job history Magazines you read If you’ve ever declared bankruptcy If you’re divorced The year you bought/lost your house Which Uni you go to if at all What topics you talk about online Brand preference Political leanings Reading habits Charitable giving Cars you own Websites you visit
- Big Data analytics can bring huge benefits
- Personalised products and services on a massive scale – offers that are tailored for you/your lifestyle.
- Fuels new services and business models – Airbnb hosts offering virtual classes (cooking, meditation, Q + A)
- Can help mitigate business risks
- There are real concerns + risks
- Increased reputational risk if (inadvertently) overstepping certain legal or social boundaries.
- Accurate customer profiling based on a variety of data sources, including social media and mobile phone data and knowing so much that customers experience a ‘creep factor’.
- E.g. Target knew a teenager was expecting before she told anyone – her parents found out from a coupon Target sent to her home.
B E N E F I T S A N D R I S K S O F U S I N G B I G D ATA
Why We Apply Ethical Reasoning INFS 2036 19
Ethics Framework Ensuring Ethical Data Practices Ethics Framework
- A formalised document or process
- Achieve a holistic view of ethical practices INFS 2036 20
Sample Ethics Framework A Data Ethics Canvas can be used to help identify and manage ethical issues at the start of a project that uses data, and throughout. It encourages you to ask important questions about projects that use data. Data ethics should be addressed at all stages of a project:
- Stewarding data
- Creating information from that data
- Decision making.
Application of a data Ethics Framework INFS 2036 22
Behavioural Exploitation RISKS USING BIG DATA
- E.g. how could supermarkets use your data?
- Your shopping history linked to your social media data can be used for dynamic pricing.
- Impulse and full-price buyers can be charged different amounts for the same item.
- Smart targeting (providing vouchers only to the ‘right’ customers)
- Sharing/selling your personal purchasing history with health and life insurance companies.
- This is already happening.
WHAT BUSINESSES CAN DO
R I S K S v s A C T I O N S T O TA K E
- Ethical debate for and against certain actions
- Analysing own initiatives or real world case studies
- Develop a code of conduct and/or a framework to describe the ethical checks and balances
- Legal implications: whether the intended use of the data matches how it is actually being used
- Be comfortable if the results became public
- Make use of organisation-wide dashboards monitoring ethical practices and breaches.
- Keep ethics at the forefront of decision-making Ethical Practices
- Intentionally exploiting users
- Being aware of the hidden traps INFS 2036 25
C O N F I R M AT I O N B I A S C R E AT E S F L A W E D D E C I S I O N - M A K I N G
A business is considering launching a new product. The CEO is using market research as a sham to confirm preconceived beliefs about a product idea and is not letting the data do the talking. The team is launching into the product development process knowing what their boss wants. As a result, the questions they craft will likely be biased to give the answers the CEO wants. The CEO has an idea for the “next big thing”. The marketing team conducts research to explore its feasibility through surveys, focus groups, and competitive analyses with this in mind. The CEO makes decisions based on this research. The product is launched and fails OR the product is not launched and a competitor successfully launches a similar product. Confirmation bias Ethical Practices Ethical Practices
- Intentionally exploiting users
- Being aware of the hidden traps INFS 2036 26
- Confirmation bias: is the tendency to use and/or interpret data as confirmation of an existing belief or theory while discounting the impact of contradictory evidence.
C O N F I R M AT I O N B I A S C R E AT E S F L A W E D D E C I S I O N - M A K I N G
Ethical Practices Ethical Practices
- Intentionally exploiting users
- Being aware of the hidden traps Confirmation bias can lead to statistical errors, as it influences the way people gather information and interpret it. Incorrect data can cause incorrect conclusions.
- Decision making is shaped by individual personality and behavioural characteristics.
- Most people don’t want new information, they want validating information. This leads to flawed decision-making. To combat this: statistical training to avoid bias in surveys and using the correct statistical tools use what-if scenarios to take unpopular perspectives into account ask a job candidate/colleague why they are not the right person for the project, to gain an objective perspective of their abilities Ask Neutral Questions Debate the Decision(s) Rethink your team INFS 2036 27
- Facebook users were shown an abnormally low number of either positive or negative posts.
- The secret experiment aimed to determine whether Facebook could alter the emotional state of its users.
- Facebook published the results in the Proceedings of the National Academy of Sciences.
- Facebook says the huge psychological experiment it secretly conducted on its users should have been “done differently” and announced a new set of guidelines for how it will approach future research studies.
- Why was it conducted? (To create better persuasion techniques using social science)
EMOTIONAL CONTAGION EXPERIMENT
Unbeknownst to users Facebook tampered with the news feeds of nearly 700,000 people. Focus on Week 7 W H Y E T H I C S , P R I VA C Y, C O N S E N T A N D T R U S T? INFS 2036 28
W H AT ’ S N E X T
Privacy Apple was also revealed to be listening in to your queries via the OG virtual assistant, Siri. Alexa has been eavesdropping on you this whole time When Alexa runs your home, Amazon tracks you in more ways than you might want. Google too admitted it records your voice using Google assistant — (not) OK Google! Facebook has confirmed it has been paying third party contractors since 2015 to transcribe user conversations they recorded in secret. Review the advertised privacy assurance regarding RayBan x Facebook glasses – does it go far enough?
ACTIVITY
THEORY PRACTICE INFS 2036 31
O V E R V I E W
Privacy vs Access
- Dynamic tension between needing to know more vs. making do with
what you have
- Increased costs of finding out and maintaining that data
- Possible challenges around consent
- Increased privacy reporting: inform individuals how their data is
to be used (e.g. cookies on a website)
- Who owns the data: informs what can be done with it and the
obligation to keep it up to date.
- Data monetisation: when data is seen as an asset organisations
consider whether they can generate revenue from it
INFS 2036 32
O V E R V I E W
Business Risks vs Personal Costs
- New ways of accessing and using data with inherent power used for the good of society. - Can e.g. eliminate or reduce fraud, crime, government mismanagement, tax evasion, welfare cheating etc.
- E.g. what price must the individual pay in terms of loss of privacy so the government can better apprehend criminals? - The same holds for corporations - Private information about employees may aid in better decision making for their good and the company - However employees’ privacy may be affected - Similar issues with customers – how to better serve customers and meet expectations without overstepping boundaries (‘creep factor’). INFS 2036 33
Privacy Policies are Important 7 R E A S O N S To build trust It is the law! Because many third parties require it To evade costly and expensive legal battles To make more money To keep what you earn Because there is too INFS 2036^ much to risk…^34
F O R E I G N L E G I S L AT I O N O N D ATA P R I V A C Y
GDPR + CCPA
- Imposes restrictions on the transfer and processing of personal data both within and outside of the EU
- Protects the fundamental human right of privacy
- Australian businesses are subject to GDPR if they:
- Have a presence in the EU;
- Offer products or services to EU residents; or
- Monitor the behavior of EU residents (e.g. analytics on your website)
- Australian business may be required to comply with GDPR by their customers. - Local residents: - Know what personal data is being collected; - Know whether their personal data is sold or disclosed and to whom; - Can say no to the sale of personal data; - Can access their personal data; - Request a business delete their personal data; - Not be discriminated for exercising their privacy rights. - Australian business may be required to comply with CCPA by their customers. - Unsuccessfully fought by Facebook, Google and INFS 2036^ Uber technologies^ to soften the law.^37
D ATA L O C AT I O N M AT T E R S
Legal International Challenges
- German consumer goods company signs contract with Irish subsidiary of U.S. cloud provider and the backup of the data physically stored in a data centre in India. - Legal location of provider = Ireland - Political location = U.S - Physical location = India
- Priority: the organisation wants all data in Germany under protection of strict EU privacy laws (GDPR). INFS 2036 38
D ATA L O C AT I O N M AT T E R S
Legal International Challenges
- Data in transit (Germany via Ireland/U.S. to India) and data at rest (India) are encrypted.
- Encryption keys in Germany.
- IT administrators unable to access unencrypted data just administer servers + database
- All non-EU entities must follow EU law:
- German employees only can view and change data
- Employees trained in EU privacy
- Bound by German employment contract, enforceable in a German court - U.S. entity cannot hand over unencrypted data to the NSA - Increase in cost and complexity - Reduction of usability e.g. - Preview and search - Delays in accessing data - Legal vs. Physical location - Legal action in which country? - Physical location still matters - Legal location matters more than ever too (e.g. Privacy Act, GDPR, CCPA). INFS 2036 39
S E L E C T E D G O O D P R A C T I C E S
Data Privacy, Ethics, Trust + Consent
- Trust transparency will become is now a competitive advantage.
- Participate in privacy certification programs (e.g. TRUSTe).
- Ensure product planners, software architects, coders, product testers and auditors are familiar with privacy principles and alert for potential violations.
- Be transparent — have clear information about corporate data protection practices. - Don’t know how consumers or citizens want their personal data to be protected? Ask them! - Digital trust is very difficult for businesses to build with customers but very easy to lose. - Necessary to gain customer consent. - Achieve trust through data ethics — focus on data ethics throughout the data supply chain from collection, aggregation, sharing, and analysis to monetisation, storage, and disposal. INFS 2036 40
