1 / 10

Big Data and its Malcontents

Big Data and its Malcontents. Stuart Anderson. McKinsey on Big Data. What’s the population?. Darpa Network Challenge: Recruited 4000 volunteers. Identify 10 red weather baloons released across continental US Prize $40K…. Suppose the prize was $1m?. Professionals “gate” membership.

obelia
Download Presentation

Big Data and its Malcontents

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Big Data and its Malcontents Stuart Anderson

  2. McKinsey on Big Data

  3. What’s the population? • Darpa Network Challenge: • Recruited 4000 volunteers. • Identify 10 red weather baloons released across continental US • Prize $40K…

  4. Suppose the prize was $1m?

  5. Professionals “gate” membership • Renal Patient View • 20,000 Renal patients • Large number of rare conditions • Sub populations – membership warranted by consultants • Valuable access to almost complete UK populations • COPD • Self-reporting • No stable risk stratification • Chronic spectrum of conditions • High variability in symptoms/measures

  6. Context Dependence • Mammography • Attempting to move screening data to training context. • Many cues from practice are relevant in the training context. • No clear notion of how to “abstract” data • Cues from practice reveal provenance. • COPD • Simple coding scheme • Large numbers of false positives • Failure adequately to take account of context • People, environment, history • Problematic to implement effective telemonitoring

  7. Models S.E.C. Chairwoman Mary Schapiro testified that "stub quotes" may have played a role in certain stocks that traded for 1 cent a share The absurd result of valuable stocks being executed for a penny likely was attributable to the use of a practice called “stub quoting.” When a market order is submitted for a stock, if available liquidity has already been taken out, the market order will seek the next available liquidity, regardless of price. When a market maker’s liquidity has been exhausted, or if it is unwilling to provide liquidity, it may at that time submit what is called a stub quote – for example, an offer to buy a given stock at a penny. A stub quote is essentially a place holder quote because that quote would never – it is thought – be reached. When a market order is seeking liquidity and the only liquidity available is a penny-priced stub quote, the market order, by its terms, will execute against the stub quote. In this respect, automated trading systems will follow their coded logic regardless of outcome, while human involvement likely would have prevented these orders from executing at absurd prices. As noted below, we are reviewing the practice of displaying stub quotes that are never intended to be executed.

  8. Vizualisation

  9. Social Computation • Use heterogeneous populations to undertake “computing” tasks. • Stratify by expertise, accuracy, … • Use incentives to shape the population response. • Computer mediated social sensemaking – developing area of work – account for context – e.g. Telemonitoring

  10. Summary • Big Data is sensitive to issues of population choice both “accidental” and “deliberate” • Data interpretation is not particularly stable even in seemingly similar contexts • Context is “defined” by the data, and is unrecorded usually • Data is only comprehensible via a model • Vizualisationchoices strongly bias decision taking.

More Related