Igqie14 analytics and ethics 20141107
•Transferir como PPTX, PDF•
1 gostou•1,375 visualizações
a whistlestop tour through some of the ethical dilemmas and challenges that arise in this "Big Data Age" and the various approaches to considering them, if not solving them. In this 10 minute "lightning talk" delegates will get insights into some of the research agenda and issues being considered in this area, touching on Business Analytics, Data Quality, analytic risks, ethics and evidence-based decision-making culture
Recomendados
Recomendados
Mais conteúdo relacionado
Mais procurados
Mais procurados (20)
Destaque
Destaque (20)
Semelhante a Igqie14 analytics and ethics 20141107
Semelhante a Igqie14 analytics and ethics 20141107 (20)
Último
Último (20)
Igqie14 analytics and ethics 20141107
- 1. “Get your facts first, then you can distort them as you please.” Data, Analytics & Ethics: finding the creepy line Sponsored by Mark Twain Alan D. Duncan http://blogs.gartner.com/alan-duncan/ Tw:@Alan_D_Duncan E: alan.duncan@gartner.com Business Analytics | Information Strategy | Data Governance | Better Business Outcomes
- 2. A bit about me.... • NEW! Research Director, Gartner • Former advisory board member, QFire Software • 22 years Information Strategy, Data Governance, Analytics & Business Consulting – Director of Data Governance at UNSW – EDS, KPMG, CPW, Acuma, Pelion, SMS – Scottish Power, United Distillers, O2, Astra Zeneca, Carphone Warehouse, Vodafone, Riyad Bank – Commonwealth Bank, NSW Roads & Maritime Services, Centrelink, OATSIH, NSW Family & Community Services, CASA, AMSA, FaHCSIA, DAFF, Navy… • Information-Management.com “Top 12 on Twitter” • Best supporting Actor, 2005 Barnet Drama Festival • See me this week in “Boudicca” at Verulamium… Alan D. Duncan http://blogs.gartner.com/alan-duncan/ Tw:@Alan_D_Duncan E: alan.duncan@gartner.com Business Analytics | Information Strategy | Data Governance | Better Business Outcomes
- 3. Key issues • Analytic value & Data Quality • Analytic risks & The Creepy Line • Suggested actions & final thoughts Alan D. Duncan http://blogs.gartner.com/alan-duncan/ Tw:@Alan_D_Duncan E: alan.duncan@gartner.com Business Analytics | Information Strategy | Data Governance | Better Business Outcomes
- 4. “The value of an idea lies in the using of it.” Analytic Value & Data Quality, Sponsored by Thomas Edison Alan D. Duncan http://blogs.gartner.com/alan-duncan/ Tw:@Alan_D_Duncan E: alan.duncan@gartner.com Business Analytics | Information Strategy | Data Governance | Better Business Outcomes
- 5. Pay a visit to Whitefellah Burrows… • 34% of organisations derive greater business value from “Big Data” once unstructured data types are addressed.1 • Unstructured data represents 90% of all real-time data being created today.2 • 59% of organisations say data quality problems are the biggest barrier to successful analytics initiatives.3 • 80% of the effort involved in dealing with data is cleaning it.3 • 1 “What Works in Big Data”, The Data Warehouse Institute, 2014. • 2 “2011 CMO Study”, IBM Institute of Business Value, 2011. • 3 “2014 Analytics, BI and Information Management Survey”, Information Week, November 2013. • 4 “Planning for Big Data: A CIO’s Handbook”, O’Reilly Media, 2012. Alan D. Duncan http://blogs.gartner.com/alan-duncan/ Tw:@Alan_D_Duncan E: alan.duncan@gartner.com Business Analytics | Information Strategy | Data Governance | Better Business Outcomes
- 6. …and don’t fall in! • Do data professionals have a special ethical responsibility? • Is there such a thing as "good analytics" and "bad analytics?" • How can you organize a structured ethical debate? • What happens if you ignore ethical debate? • Where do you stand? Alan D. Duncan http://blogs.gartner.com/alan-duncan/ Tw:@Alan_D_Duncan E: alan.duncan@gartner.com Business Analytics | Information Strategy | Data Governance | Better Business Outcomes
- 7. “Experts often possess more data than judgement.” The “Creepy Line”, Sponsored by Colin Powell Alan D. Duncan http://blogs.gartner.com/alan-duncan/ Tw:@Alan_D_Duncan E: alan.duncan@gartner.com Business Analytics | Information Strategy | Data Governance | Better Business Outcomes
- 8. The Creepy line • The creepy line is real • The creepy line is fuzzy • The creepy line is near • There is no such thing as an “absolute right to privacy” Alan D. Duncan http://blogs.gartner.com/alan-duncan/ Tw:@Alan_D_Duncan E: alan.duncan@gartner.com Business Analytics | Information Strategy | Data Governance | Better Business Outcomes
- 9. Avoiding “creepy” ain’t so easy • Risk 1: Anonymization and Data Masking Could be Impossible • Risk 2: Protecting People From Themselves • Risk 3: It's Easy to Mistake Patterns for Reality • Risk 4: The Data Becomes Reality Itself • Risk 5: Don't Worry About Bad People; Worry About the Ignorant Ones Alan D. Duncan http://blogs.gartner.com/alan-duncan/ Tw:@Alan_D_Duncan E: alan.duncan@gartner.com Business Analytics | Information Strategy | Data Governance | Better Business Outcomes
- 10. “Data! Data! Data! I can’t make bricks without clay!” (Sherlock Holmes, The Adventure of the Copper Beeches) Digital & Analytic Ethics, Sponsored by Sir Arthur Conan Doyle Alan D. Duncan http://blogs.gartner.com/alan-duncan/ Tw:@Alan_D_Duncan E: alan.duncan@gartner.com Business Analytics | Information Strategy | Data Governance | Better Business Outcomes
- 11. Suggested Actions • First, Get the Basics Right • Information Governance Needs To Grow Up • Connect Business Value & Consumer Value • Be Sensitive to Culture & Values • Understand the Origin of Data • Realize That There Is no Such Thing as "Being Objective" in Analytics Alan D. Duncan http://blogs.gartner.com/alan-duncan/ Tw:@Alan_D_Duncan E: alan.duncan@gartner.com Business Analytics | Information Strategy | Data Governance | Better Business Outcomes
- 12. “Government-aku Law nyiringka ngarapai. Ananguku Law katangka munu kurunta ngarapai. Nganana Tjukurpa nyiringka tjunkupai wiya. Tjukurpa panaya tjamulu, mamalu, ngunytjulu nganananya ungu, kurunta munu katangka kanyintjaku.” “Government law is written on paper. Anangu carry our law in our heads and in our souls. We don't put our Law onto paper. It was given to us by our grandfathers and grandmothers, out fathers and mothers, to hold into in our heads and in our hearts.” Analytic Ethics, sponsored by the Anangu people of Uluru Kata Tjuta Alan D. Duncan http://blogs.gartner.com/alan-duncan/ Tw:@Alan_D_Duncan E: alan.duncan@gartner.com Business Analytics | Information Strategy | Data Governance | Better Business Outcomes
- 13. Suggested Resources • Maverick* Research: Ethics Are at the Center of the Nexus of Forces (G00256392) • Privacy and Ethical Concerns Can Make Big Data Analytics a Big Risk Too (G00249134) • Big Data Analytics Requires An Ethical Code of Conduct (G0026399) • Digital Ethics, or How to Not Mess Up With Technology (G00269636) • Also: • http://informationaction.blogspot.co.uk/p/favourite-articles.html Alan D. Duncan http://blogs.gartner.com/alan-duncan/ Tw:@Alan_D_Duncan E: alan.duncan@gartner.com Business Analytics | Information Strategy | Data Governance | Better Business Outcomes
- 14. Intellectual curiosity Skeptical scrutiny Critical thinking http://www.informationaction.blogspot.com.au/ http://blogs.gartner.com/alan-duncan/ @Alan_D_Duncan http://www.linkedin.com/in/alandduncan Alan D. Duncan http://blogs.gartner.com/alan-duncan/ Tw:@Alan_D_Duncan E: alan.duncan@gartner.com Business Analytics | Information Strategy | Data Governance | Better Business Outcomes
Notas do Editor
- Or; Just because we can doesn’t mean we should,
- For those of you who don’t know me. the potted history is that I’ve been doing Information & Data Management for over 20 years, mostly in Consulting/Service provider and advisory roles to commercial businesses. After 6 years in Australia. I came back to UK in July & joined Gartner about 2 months ago. This isn’t the only thing I do…
- We’ve got 10 minutes, so I’m going fly through at a heck of a speed – sorry!
- MILESTONE CHECKPOINT = @30MIN What do you think about “need to know”? Control, or publish and be damned?! Value from data = shared knowledge. Who gets to decide on “control”… I think the key is transparency and accountability…
- Analytic usage drives better Data Quality: Information > Action > Outcome. VIRTUOUS CIRCLE, Three Vs that matter Variability: Within any given data set, is the structure of that data regular and dependable, or is subject to unpredictable change? If so, how can we understand the nature of the “unstructured” text data content (or sound, or video) and interpret it in a way that becomes meaningful for the required business analytic-ready output? Veracity: How do we know that the data is actually correct and fit for purpose? Can we test the data against a set of defined criteria that establish the degree of confidence and trustworthiness? What are the business rules that enable the data to be tested and profiled? If there are issues with the data, what actions can be taken to clean and correct the data before any analysis is carried out? Value: What is the business purpose or outcome that we are trying to meet? What questions are we seeking to answer, and what actions do we expect to take as a result? What benefits do we expect to achieve from collecting and analysing the data? Has the data been aligned with the desired outcome? All three of these additional characteristics require a clear understanding of the business context, which then is used to frame the meaning and purpose of the data content. “Variability”, “Veracity” and “Value” all express different aspects of the fitness-for-purpose of the data sets in question, all of which need to be addressed in order to solve a business problem in business terms.
- cf. Maverick* Research: Ethics Are at the Center of the Nexus of Forces Questions to be asking…
- MILESTONE CHECKPOINT = @30MIN What do you think about “need to know”? Control, or publish and be damned?! Value from data = shared knowledge. Who gets to decide on “control”… I think the key is transparency and accountability…
- cf. Big Data Analytics Requires An Ethical Code of Conduct The creepy line is real . There have been many reported incidents resulting in public outcry over the analytical capabilities and action of governments and commercial enterprise (see "Privacy and Ethical Concerns Can Make Big Data Analytics a Big Risk Too" ). The creepy line is fuzzy . What gets accepted from one organization becomes a scandal elsewhere. Internet giants such as Google and Facebook seem to have created much more latitude than, for instance, financial services institutions. Governance agencies set the right example with open data but, at the same time, create the most impactful scandals through their intelligence agencies. The creepy line is near . With privacy becoming such a sensitive topic and consumers being cynical, it is quite easy to step over the creepy line, even with the most straightforward of business initiatives, such as reselling aggregated and anonymized data. GOOGLE STREETVIEW, TARGET, TOMTOM examples
- “Nazar” amulets to ward off the “Evil eye” f. Privacy and Ethical Concerns Can Make Big Data Analytics a Big Risk Too Risk 1: Anonymization and Data Masking Could be Impossible Large datasets are often subjected to an "anonymization" process to enable the data to be used for marketing or scientific research, without the potential of leaking information about the individuals. However, no useful database can ever be perfectly anonymous. 1 Furthermore, for several decades, the information security research community has recognized that bodies of low sensitivity data, when they can be correlated, can often result in a set of data that has much higher significance that any of the original datasets. When done with malicious intent, this is referred to as an inference attack, or slightly the more neutral term "reidentification" (see "Tossing Your Cookies: The Privacy Implications of Context-Aware Agents "). The "triple identifier" of birthday, gender and zip code is all that someone needs to uniquely identify at least 87% of U.S. citizens in publicly available databases. 2 The individuals who might have given permission to have their data used in what they believe to have been an anonymous fashion might have no idea that reidentification is even possible. This can lead to harmful results, revealing information on medical history, personal habits, financial situation and family relations that most people would classify as private. Risk 2: Protecting People From Themselves Not everyone cares enough about their own privacy. Many consumers use social media or Internet-based services carelessly, allowing others to make use of information in unintended ways. Consider the following examples: Publicizing on Twitter that you are on vacation or "checked in" somewhere with the whole family shows you are not at home (see Note 3). Consumers almost never read the "terms and conditions." 3 (3 http://www.foxnews.com/tech/2010/04/15/online-shoppers-unknowingly-sold-souls/) To receive a promotion, consumers often need to provide some personal information. Even though people are expected to know what they are doing, and there may be no legal issues after consumers consent to providing information, there is reputational risk to companies if consumers feel their trust and confidence was breached. What consumers trust you to do (or not do) doesn't necessarily equal what is legally allowed to do. Risk 3: It's Easy to Mistake Patterns for Reality Mass shootings, for example, in the U.S., have generated interest in attempting to determine which individuals are likely to act out on violent impulses. These clues are believed to be available in Facebook and other social media. Institutionalizing this type of activity could result in a sort of "Minority Report" phenomenon (see Note 4). Governments are already conducting data mining of cash transactions to infer the activities of terrorists and other organized criminals. Police forces use advanced predictive analytics to predict a higher chance of crime rates in certain areas on certain days or times in the day. Surveillance cameras in streets are connected to analytical software that is engineered to detect behavioral patterns indicating trouble. This may easily lead to "fishing expeditions," where authorities conduct mass analytic exercises, in which any person fitting a certain pattern becomes a suspect. For crime prevention purposes, there is a direct issue with the constitutional presumption of innocence. In business, a pattern doesn't necessarily equate to behavior. Risk 4: The Data Becomes Reality Itself In business, unintended behavioral influence happens as well. Based on advanced analytics, retailers provide customers with personalized offers. Confronted with perceived "endless" choice in online and street retail and a lack of ability to compare to other offerings, customers are likely to welcome such offers. The acceptance of the offer refines the profile, leading to an even more targeted offer, leading to higher conversion rates again. Through this closed loop, the profiling and associated prescriptive analytics start driving customer behavior, rather than the other way around. This is commercially interesting, but ethically debatable. Risk 5: Don't Worry About Bad People; Worry About the Ignorant Ones Big data analytics distinguishes itself through the use of automated discovery techniques, presenting potentially interesting clusters and combinations in data. This is a powerful tool when dealing with high volume and high velocity data with a high degree of variation, but also potentially dangerous. Customer segmentation and profiling can easily lead to discrimination based on age, gender, ethnic background, health condition, social background, and so on. These are limitations known to analysts, but not to technology. This is particularly an issue when dealing with automatic discovery tools that expose all the data it can find to a set of data mining algorithms and presents the results. Automated discovery tools effectively answer questions that were never asked, leading to potentially embarrassing or even illegal results. Knowledge, once gained, cannot be undone. Even deciding not to do anything with the knowledge is a decision with consequences already. In addition, the way many big data analytical processes work doesn't automatically provide a set of checks and balances. Small teams experiment with datasets in a highly associative and iterative way. This doesn't encourage discussion about what is right or wrong in analytics. To guide big data analytics, it makes sense to also consider what data and analytics you would like to have, and equally important, what not. One bank, for instance, removed face recognition algorithms from its set of analytics, because it didn't even want to be seen as being able to use it. Organizations need to evaluate the value of knowing the answers to specific information-driven questions, analysis and models before they develop the model. Intent becomes the precursor to big data analytics. "Why do you want to know it" becomes the gateway before "what do you want to know."
- Cf. Big Data Analytics Requires An Ethical Code of Conduct First, Get the Basics Right (Don’t break the law, be professional, take care of security) Information Governance Needs To Grow Up (freedom to explore vs freedom to act. Data masking. NDAs, Information Asset Management) Connect Business Value & Consumer Value (Profiling vs prejudice; informed consent vs Legal Ts&Cs; building trusted relationship) Be Sensitive to Culture & Values (industry, geographical, societal norms) Understand the Origin of Data (where it came from, why it was collected. Primary purpose vs secordary uses eg TomTom data collection used by Police) Realize That There Is no Such Thing as "Being Objective" in Analytics (Confirmation biases)
- Final thoughts – the Anangu people don’t write their laws down. That’s whitefellah stuff. They LIVE them.