Workshop Data Manager
0 likes842 views
The document discusses the importance of extracting business value from big data, emphasizing that data alone is not valuable without actionable insights and the right questions. It covers topics such as teamwork, decision trees, data ethics, and the evolution of data infrastructure, while encouraging collaboration between data scientists and business teams. Additionally, it highlights the need for domain knowledge and the risks of overfitting in predictive modeling.
Workshop Data Manager
- 1. LutzFinger.com How to extract significant business value from big data
- 3. LutzFinger.com Disclaimer This presentation is solemnly my opinion and not necessarily the opinion of my employer Harvard, Linkedin or Cornell.
- 4. LutzFinger.com Agenda The right Ask Data is King Team-Work: Discover an Ask Lunch Decision Tree Team-Work: Your Model Pitfalls with Data Break Technology Team-Work: Become Data Driven? BI vs. Data Science Build A Team
- 5. LutzFinger.com Why is there such hype?
- 7. LutzFinger.com The ones who predict: image by Mike under Creative Commons
- 8. LutzFinger.com McK Study forecasted: 10 Times More Managers per Data Savvy Person
- 11. LutzFinger.com ASK the right Questions. MEASURE the right data – even if it is not Big data. Take Actions and LEARN from them. ?
- 12. LutzFinger.com
- 13. LutzFinger.com Google had the right Question is difficult to find
- 15. LutzFinger.com Already Known Asks byrgiesekingunderCreativeCommons(CCBY2.0) Who should get an E-Shot? Territory Planning for my Sales Force Budget Planning of Marketing Spent Online Product Recommendation Real Time Betting for Ad-spaces Customer Segmentation Social Media Influencers Call Center Routing based on Questions Capacity Forecasting …. and more
- 17. LutzFinger.com Data by itself is USELESS Information by itself is USELESS Only action counts!
- 18. LutzFinger.com Data by itself is USELESS Information by itself is USELESS Let’s connect...
- 20. LutzFinger.com A good ‘so what’? DatabyLinkedIn Example: Laboratory Manager?
- 21. LutzFinger.com A bad ‘so what’ 300+ Million Member at LinkedIn 60.000 with a Job Title that might fit 19.000 who switched after 3 to 8 years 24 who had the same career path
- 24. LutzFinger.com Recommendations: Your Focus People You May Know Groups You May Like Ads in Which You May Be Interested Companies You May Want to Follow Pulse Similar Profiles
- 26. LutzFinger.com Many Good Examples Benchmark Recommendations
- 27. LutzFinger.com Agenda The right Ask Data is King Team-Work: Discover an Ask Lunch Decision Tree Team-Work: Your Model Pitfalls with Data Technology Team-Work: Become Data Driven? BI vs. Data Science Build A Team
- 28. LutzFinger.com “Data is the new oil” - World Economic Forum
- 29. LutzFinger.com “DATA IS THE NEW OIL” Oil Mine the oil Use the oil Goal
- 30. LutzFinger.com V OF “BIG DATA” Data at scale (TB, PB … ) Data in many forms (Structured, unstructured ...) Speed (Streaming, real time, near time ..) Uncertainty (imprecise, not always up-to-date ..)
- 31. LutzFinger.com 1st. Round Monopoly Photo by William Warby under the Creative Commons (CC BY 2.0)
- 33. LutzFinger.com Prediction Photo by KOMUnews under the Creative Commons (CC BY 2.0) Boring could be the New Sexy!
- 34. LutzFinger.com Data Might (Not) Be A Barrier To Enter
- 35. LutzFinger.com Data Might (Not) Be A Barrier To Enter
- 36. LutzFinger.com Public Data is Not
- 37. LutzFinger.com DATA Categorical • Ordinal: Monday, Tuesday, Wednesday • Nominal: Man, Woman Quantitative: • Ratio: Kelvin, Height, Weight • Interval: Celsius, Fahrenheit Structure: • Structured • Unstructured • Semi-structured / Meta data Read more: “On the Theory of scales of measurement” S.Stevens 1946
- 38. LutzFinger.com Data Is King but not all data is equal.
- 39. LutzFinger.com The Tale of “Social Media” Data Source:‘AskMeasureLearn’byO’ReillyMedia
- 40. LutzFinger.com Structured Data Is Often Better New York Weather in April 2013 Source: ‘Ask Measure Learn’ by O’Reilly Media
- 41. LutzFinger.com Sometimes, it’s worth it. RE @dave_mcgregor: Publicly pledging to never fly @delta again. The worst airline ever. U have lost my patronage forever du to ur incompetence Completely unimpressed with @continental or @united. Poor communication, goofy reservations systems and all to turn my trip into a mess. @SouthWestAir I know you don't make the weather. But at least pretend I am not a bother when I ask if the delay will make miss my connection
- 42. LutzFinger.com But Data Is King This will give birth to devices (i.e., the Star Trek Tricorder) that allow you, the consumer, to self-diagnose, anytime, anywhere.
- 43. LutzFinger.com Agenda The right Ask Data is King Team-Work: Discover an Ask Lunch Decision Tree Team-Work: Your Model Pitfalls with Data Technology Team-Work: Become Data Driven? BI vs. Data Science Build A Team
- 45. LutzFinger.com The Media industry has changed! The retail industry has change! The Education sector is changing! Finance Industry and healthcare sector are under attack. Which industry will be next?
- 46. LutzFinger.com Team Work Photo by Creative Sustainability under the Creative Commons (CC BY 2.0) A. The Ask ○ is it actionable? “So What? ○ is it Benchmarking / is it Recommendation B. The Data ○ do only you have this data? ○ do you have a feedback loop?
- 48. LutzFinger.com Agenda The right Ask Data is King Team-Work: Discover an Ask Lunch Decision Tree Team-Work: Your Model Pitfalls with Data Technology Team-Work: Become Data Driven? BI vs. Data Science Build A Team
- 49. LutzFinger.com Are Retail Banks ‘dead’?
- 50. LutzFinger.com Decision Trees Step by Step by Maciej Lewandowski under Creative Commons (CC BY-SA 2.0)
- 51. LutzFinger.com Split Apples & Mandarins
- 52. LutzFinger.com What Is The Target Variable?
- 53. LutzFinger.com What Is The Features To Describe The Target?
- 54. LutzFinger.com What Is The Features To Describe The Target? • Weight: light, medium, heavy - or x gram • Size: round or not • Color:green, orange, red • Surface: flat or porous surface • …
- 55. LutzFinger.com Which Feature Works Best? ● The variable with the most important information about target variable. ● Which variable can split the group as homogeneous with respect to the target variable. (pure vs. inpure)
- 56. LutzFinger.com Color Red? Color Orange? Split on Color Red vs. Split on Color Orange Which One Is Better?
- 57. LutzFinger.com We Need A Way To Describe Chaos "ClaudeElwoodShannon(1916- 2001)"bySource.Licensedunder FairuseviaWikipedia
- 58. LutzFinger.com ENTROPY Entropy is a measure of disorder. Entropy only tells us how impure one individual subset is.
- 59. LutzFinger.com ENTROPY & PROBABILITY entropy = -p1 * log (p1) - p2 * log (p2) - ….
- 60. LutzFinger.com ● Highest Entropy Reduction ● Highest Information Gain
- 61. LutzFinger.com 1st. Entropy Without Split entropy = -p1 * log (p1) - p2 * log (p2) Apple: 8 out of 15 p(apple)= 8/15 Mandarines: 7 out of 15 p(mandarine)= 7/15 ENTROPY (Without Split): -p(apple)*log(p(apple)) -p(mandarins)*log(p (mandarines)) = 0.996791632 = 1 very impure
- 62. LutzFinger.com Color Red? Color Orange? entropy = -p1 * log (p1) - p2 * log (p2) ENTROPY (After Split on Red): = 8/15* ENTROPY (Split on Red=’no’) + 7/15* ENTROPY (Split on Red=’yes’) = 0.43 + 0.28 = 0.71 INFORMATION GAIN = Entropy (Before) - Entropy (After) = 1 - 0.71 = 0.29 ENTROPY (Split on Red=’ no’): = -6/8*(log2 (6/8))-2/8*(log2 (2/8)) = 0.81 ENTROPY (Split on Red=’yes’): = -6/7*(log2 (6/7)) -1/7*(log2 (1/7)) = 0.59 ENTROPY (Split on Orange=’yes’): = -6/6*(log2 (6/6)) = 0 ENTROPY (Split on Orange=’no’): = -8/9*(log2 (8/9))-1/9*(log2 (1/9)) = 0.50 ENTROPY (After Split on Orange): = 6/15* ENTROPY (Split on Orange=’no’) + 9/15* ENTROPY (Split on Orange=’yes’) = 0 + 0.23 = 0.23 INFORMATION GAIN = Entropy (Before) - Entropy (After) = 1 - 0.23 = 0.77
- 63. LutzFinger.com INFORMATION GAIN (IG) Information gain measures how much a given feature improves (decreases) entropy over the whole segmentation it creates. How important is this feature for the prediction?
- 64. LutzFinger.com Decision Tree Color Orange? ROOT NODE LEAFS
- 65. LutzFinger.com Decision Tree Color Orange? Decision Tree Structure
- 66. LutzFinger.com Which Feature Would Be Better?
- 69. LutzFinger.com Hyperplane (2 dimensions) Mandarines Red Green LightHeavy
- 70. LutzFinger.com Agenda The right Ask Data is King Team-Work: Discover an Ask Lunch Decision Tree Team-Work: Your Model Pitfalls with Data Technology Team-Work: Become Data Driven? BI vs. Data Science Build A Team
- 71. LutzFinger.com Back To The Lending Industry
- 72. LutzFinger.com BIG ML Competitors: ● Algorithms.io ● SnapAnalytx ● wise.io ● Predixion Software ● Google Prediction API
- 74. LutzFinger.com Build Database How Do You Deal With Categorical vs. Numeric Variables in Decision Trees? screenshot from bigML tool
- 75. LutzFinger.com Configure And Build Model Select The Objective Field - What To Train The Model On? That is the Row ID - surely no impact. screenshot from bigML tool
- 76. LutzFinger.com screenshot from bigML tool
- 77. LutzFinger.com screenshot from bigML tool
- 78. LutzFinger.com Highest Information Gain screenshot from bigML tool
- 79. LutzFinger.com Using The Model screenshot from bigML tool
- 80. LutzFinger.com Using The Model screenshot from bigML tool
- 81. LutzFinger.com Using The Model screenshot from bigML tool
- 82. LutzFinger.com Found 2,470 New Instances screenshot from bigML tool
- 83. LutzFinger.com How Can I Improve Now Quality?
- 84. LutzFinger.com Agenda The right Ask Data is King Team-Work: Discover an Ask Lunch Decision Tree Team-Work: Your Model Pitfalls with Data Technology Team-Work: Become Data Driven? BI vs. Data Science Build A Team
- 85. LutzFinger.com Pitfalls with Data Data & Ethics More Data & Overfitting Confidence & Cut-off Cause & Correlation …
- 86. LutzFinger.com How Did They Improve Scoring?
- 87. LutzFinger.com Social Network Info Could Social Network improve the quality of our prediction? Who is more credit-worthy? a. Tim whose friends are all very credit worthy b. Tom whose friends are not creditworthy
- 89. LutzFinger.com Nobel Worthy! Muhammad Yunus Photo by University of Salford under Creative Commons CC BY 2.0
- 90. LutzFinger.com In the EU insurers will no longer be allowed to take the gender of their customers into account for insurance premiums: ● young men's premiums will fall by up to 10% ● young women's premiums will rise by up to 30% by: BBC News: http://www.bbc.com/news/business-12608777 Not Everything That Is Possible Is Legal
- 91. LutzFinger.com Pitfalls with Data Data & Ethics More Data & Overfitting Confidence & Cut-off Cause & Correlation …
- 92. LutzFinger.com The Tale of Big Data
- 93. LutzFinger.com Overfitting To tailor a model to training data at the expense of being generalizable for previously unseen data points. The model becomes perfect in describing noise and spurious correlations. TRADE OFF Complexity of a Model & Overfitting Likelihood
- 94. LutzFinger.com How Trustworthy Is This Prediction? • 45 instances • 59% confidence screenshot from bigML tool
- 95. LutzFinger.com The Need for Domain Knowledge
- 96. LutzFinger.com Pitfalls with Data Data & Ethics More Data & Overfitting Confidence & Cut-off Cause & Correlation …
- 97. LutzFinger.com Give Credit or Not? 49% Confidence screenshot from bigML tool
- 98. LutzFinger.com CONFUSION MATRIX Pregnant (60) Not pregnant (940) Pregnant (A) true positive (B) false positive Not pregnant (C) false negative (D) true negative Classifier Reality
- 99. LutzFinger.com Agenda The right Ask Data is King Team-Work: Discover an Ask Lunch Decision Tree Team-Work: Your Model Pitfalls with Data Technology Team-Work: Become Data Driven? BI vs. Data Science Build A Team
- 100. LutzFinger.com How Invented Big Data Infrastructure?
- 101. LutzFinger.com How Invented Big Data Infrastructure?
- 102. LutzFinger.com Issue Of Yahoo CENTRALIZED SYSTEMS ARE EXPENSIVE • diminishing returns in power (overhead issue) • exponential cost to scale. • slow to transport (ETL) the data Scan 1000 TB Datasets on a 1000 node cluster: • Remote Storage @ 10 MB/s = 165 min • Local Storage @ 200 MB's = 8 min MAKE SYSTEMS FAULT TOLERANT 1000 nodes - a machine a day will break
- 103. LutzFinger.com The Vision CHEAP Systems • can run on commodity hardware Computation are done DECENTRAL • ability to ‘dispatch’ a task • parallelize work-streams Fault TOLERANT no matter where and when break is not an issue
- 104. LutzFinger.com
- 105. LutzFinger.com How To Access HDFS Hadoop Storage (HDFS / HBase / Solr) Map Reduce
- 106. LutzFinger.com Via The Normal Languages Hadoop Storage (HDFS / HBase / Solr) Map Reduce MapReduce Hive Pig/Casscading Giraph Mahout SQL Like Scripting Like Graph Oriented ML Engine
- 107. LutzFinger.com Pro & Con Hadoop Storage (HDFS / HBase / Solr) Map Reduce MapReduce Hive Pig/Casscading Giraph Mahout SQL Like Scripting Like Graph Oriented ML Engine Store ETL: Extract / Transform / Load DB / Key Value Store Visualize Pro: way better than traditional BI Con: Heavy tech involvement. 12-18 month for non-tech company to implement a schema
- 108. LutzFinger.com Pro & Con Hadoop Storage (HDFS / HBase / Solr) Map Reduce MapReduce Hive Pig/Casscading Giraph Mahout SQL Like Scripting Like Graph Oriented ML Engine DB / Key Value Store Visualize New Approaches: ● Spark ● Tez ● Flink
- 109. LutzFinger.com Agenda The right Ask Data is King Team-Work: Discover an Ask Lunch Decision Tree Team-Work: Your Model Pitfalls with Data Technology Team-Work: Become Data Driven? BI vs. Data Science Build A Team
- 110. LutzFinger.com Why Is It So Hard To Become Data Driven
- 111. LutzFinger.com Ingredients of Data Products The question? Ask The need? The Why? Measure The Data? The features? Team All of them are necessary - None of them is sufficient! The algorithms? The right Skills? Collaboration 111
- 112. LutzFinger.com How To Ingest Ideas Hack - Days & Incubator Internal Process External Competition Close Collaboration between Business & Data Scientists“All we do is Data” - Jeff Weiner 112
- 113. LutzFinger.com What Would You Need To Do To Be A Leader In Data
- 114. LutzFinger.com Agenda The right Ask Data is King Team-Work: Discover an Ask Lunch Decision Tree Team-Work: Your Model Pitfalls with Data Technology Team-Work: Become Data Driven? BI vs. Data Science Build A Team
- 115. LutzFinger.com Old vs. New Old School Today / Big data Data Amount Gigabytes & Terabytes Petabytes & Exabytes IT Infrastructure Centralized Decentralized / Parallelized Data Types Structured Structured & unstructured Schema Stable schema Schema on the fly When and How is the ASK formulated? Set ask Ad-hoc ask
- 116. LutzFinger.com How to build a Data Team
- 119. LutzFinger.com New Ways To Automate
- 120. LutzFinger.com Data Scientist BI Analyst Engineer Product Manager Communication Skills Domain Knowledge
- 121. LutzFinger.com You Learned image by Mike under Creative Commons • The Ask is the most Important part - you need Domain Knowledge • Data Science is NO Rocket Science • Data is King & There is Monopoly Game happening • Data Can be misleading • Data is a Team Sport
- 123. LutzFinger.com What to MEASURE? • Error • Correlation • Cost & • Privacy Workbook “Measure” at LutzFinger.com