Diese Präsentation wurde erfolgreich gemeldet.
Wir verwenden Ihre LinkedIn Profilangaben und Informationen zu Ihren Aktivitäten, um Anzeigen zu personalisieren und Ihnen relevantere Inhalte anzuzeigen. Sie können Ihre Anzeigeneinstellungen jederzeit ändern.
Nächste SlideShare
×

# Datascience Introduction WebSci Summer School 2014

1.055 Aufrufe

Veröffentlicht am

http://www.summerschool.websci.net/
WebScience Summer School Southampton
Data Science 2014

Veröffentlicht in: Bildung
• Full Name
Comment goes here.

Are you sure you want to Yes No
• Claudia - nice overview of the building blocks and basic intuitions

Sind Sie sicher, dass Sie …  Ja  Nein
Ihre Nachricht erscheint hier

### Datascience Introduction WebSci Summer School 2014

1. 1. Dr. ClaudiaWagner http://claudiawagner.info/ Web Science Summer SchoolWS3 , Southampton, UK , 21th July 2014
3. 3.  Statistical computing is very central , but data science is more than statistics  Activities of data scientists:  collection and generation,  preparation,  analysis,  visualization,  management and preservation of large collections of data Jeffrey Stanton, Introduction to Data Science, free e-book 3
4. 4.  Ask interesting question  Why is it important?Which number answers your question?  Get or generate the data  Which data will help answering you question? How is the data generated? Are their any sampling biases? Ethical issues?  Analyze the data  Are there any anomalies or regularities?  Which hidden process has generated the data?  Fit a model to the data and validate it  Visualize and communicate results  What does 75% probability mean?  Preserve and share the data to make results reproducible 4
5. 5.  Data is a collection of facts  Facts can be numbers, words, measurements, observations or even just descriptions of things  Qualitative data (e.g., “it was great”)  Quantitative data  Discrete (e.g., 5)  Continuous (e.g., 3.723) 5
6. 6. 6 Stevens, S. S. (1946). "On theTheory of Scales of Measurement". Science 103 (2684): 677–680. Nominal (e.g., ethnic group, sex, nationality) Ordinal (e.g., status) Interval (e.g., temperature in Celsius) Ratio (e.g., weight) Observations are only named Observations can be ordered Distance is meaningful Absolute zero
7. 7. 7
8. 8.  Random sample of Twitter users  Random sample of tweets from the public timeline  More active users are more likely to be included  Friendship Paradox  Select a random sample of people and ask them to list the people they know. Contact a sample of the listed friends and repeat the survey.  Sampling bias: people with more friends are more likely to show up in the friend lists which we generate at the first stage 8
9. 9.  A study found that the profession with the lowest average age of death was student.  Being a student does not cause you to die at an early age. Being a student means you are young.This is what makes the average of those that die so low.  Amount of ice cream consumed per day is highly correlated with number of drownings per day  Both variables are correlated with the daily temperature 9 "Teaching Statistics:A Bag ofTricks," by Gelman and Nolan (2002)
10. 10.  A study found that only 1.5% of drivers in accidents reported that they were using a cell phone, whereas 10.9% reported that they were distracted by another occupant in the car.  Can we conclude that using a cell phone safer than speaking with another occupant?  P(cellphone | accident) != P(accident | cellphone)  Compare P(accident|cellphone) and P(accident|occupant)  We need to know the prevalence of cell phone use  It is likely that much more people talk to another occupant in the car while driving than talking on the cell phone 10 Jessica Utts, What Educated Citizens Should Know about Statistics and Probability,The American Statistician, Vol. 57, No. 2 (May, 2003), pp. 74-79
11. 11.  Ecological Fallacy  Illiteracy rate in each US state and the proportion of immigrants per state  Negative correlation of −0.53 ▪ The greater the proportion of immigrants in a state, the lower its average illiteracy.  When individuals are considered, the correlation was +0.12 — immigrants were on average more illiterate than native citizens. 11 Robinson, W.S. (1950). "Ecological Correlations and the Behavior of Individuals". American Sociological Review (American Sociological Review, Vol. 15, No. 3) 15 (3): 351–357.
12. 12. Data Collection Data Preprocessing DataAnalysis DataVisualization Data Preservation
13. 13.  Found data or observational data  Are observational data enough?  Are such data available?  Generate Data  Designs the data generation process ▪ E.g., via surveys, experiments, crowdsourcing 13
14. 14. 14http://www.intel.com/content/www/us/en/communications/internet-minute-infographic.html
15. 15. Two general types of traces: 15 Accretion - a build-up of physical traces Erosion - the wearing away of material Webb, Eugene J. et al. Unobtrusive Measures: nonreactive research in the social sciences. Chicago: Rand McNally, 1966
16. 16.  Bulk downloads  Wikipedia, IMDB, Million Song Database, etc.  API access  NYTimes,Twitter, Facebook, Foursquare, etc.  Web scraping  Tools e.g., http://scrapy.org/  What data is ok to scrap? ▪ Public, non-sensitive, anonymized, fully referenced information, Check terms of conditions! 16
17. 17.  Takes time to accumulate  Conservative estimate  Only what happened counts! Intentions, motivations or internal states don’t count.  Inferentially weak  Cannot answer “what-if” questions 17
18. 18.  Surveys  Simulations  Model behavior of users/agents on a micro-level  Simulate what happens under different conditions  Empirical validation  Experiments  Keep all variables constant and only manipulate one variable (e.g., emotions) 18
19. 19.  Simulations  Study of macro-phenomena  Difficult to validate empirically  Surveys and/or Experiments  We only get data from those who are accessible and willing to respond or participate  Responders provide answers that are in line with self- image and researcher’s expectations  Hawthorne effect, etc. 19
20. 20. Data Collection Data Preprocessing DataAnalysis DataVisualization Data Preservation
21. 21. 21  Data cleaning  Fill in missing values  Smooth noisy data  Identify or remove outliers  Resolve inconsistencies  Data integration  Integration of multiple databases, or files
22. 22. 22  Data transformation  Normalization: scaled to fall within a small, specified range  Standardization: how many standard deviations from the mean lies each data point  Discretization: divide the range of a continuous attribute into intervals  some algorithms require discrete attributes.  Data reduction  Dimensionality reduction (remove unimportant attributes via feature selection, group features into factors e.g. PCA, SVD)  Aggregation and clustering  Sampling
23. 23. Data Collection Data Preprocessing Data Mining DataAnalysis  Statistical Inference DataVisualization Machine Learning Data Preservation
24. 24.  Problem:  Given high dimensional space (e.g., fb-user which are described via various attributes such as locations they visited)  Find pairs of data points (𝒙, y) that are within some distance threshold 𝒅(𝒙, y) ≤ 𝒔  We first need to decide what „distance“ means 24
25. 25.  Distance Measures  Jaccard similarity between 2 sets of items I1, I2 sim(I1, I2) = |𝐼1 ∩ 𝐼2| |𝐼1 ∪ 𝐼2| dist(I1, I2) = 1- sim(I1, I2)  Euclidian distance, Hamming distance, Cosine Similarity, etc. 25
26. 26.  Goal: Given a set of items group the items into some number of clusters, so that  Members of a cluster are similar to each other  Members of different clusters are dissimilar 26Anand Rajaraman, Jeffrey Ullman, Jure Leskovec, Mining of Massive Datasets, Cambridge University Press
27. 27.  Not-Hierarchical / Point assignment:  Maintain a set of clusters  Point belong to “nearest” cluster  Hierarchical:  Agglomerative (bottom up): ▪ Initially, each point is a cluster ▪ Repeatedly combine the two “nearest” clusters into one  Divisive (top down): ▪ Start with one cluster and recursively split it 27Anand Rajaraman, Jeffrey Ullman, Jure Leskovec, Mining of Massive Datasets, Cambridge University Press
28. 28. 28
29. 29. 29
30. 30. 30
31. 31. 31
32. 32. 32
33. 33.  Try different k, looking at the change in the average distance to centroid as k increases  Average falls rapidly until right k, then changes little 33 Average Diameter k best k
34. 34.  Aim: Find hidden concepts/groups in a matrix  Method: SingularValue Decomposition (SVD) 34 Lescovec et al., Mining of Massive Datasets, p. 418
35. 35.  Rank = 2  Rank denotes the information content of the matrix.  For instance, a rank-1 matrix can be written as a product of one column and one vector 35
36. 36. 36
37. 37. 37 Lescovec et al., Mining of Massive Datasets, p. 418 Relates users and concepts Relates movies to concepts Strength of concepts
38. 38. Data Collection Data Preprocessing Data Mining DataAnalysis  Statistical Inference DataVisualization Machine Learning Data Preservation
39. 39.  Estimate population parameter from sample statistics  Sampling Distribution of statistic:  Draw a finite set of samples of size n from the population  Computing the statistic on the sample  Repeat this process  The mean of the sampling distribution is the expected value of the statistic in the true population  SD of the sampling distribution is the standard error 39
40. 40. 40
41. 41.  Some descriptive statistics such as mean or median are unbiased estimators of central tendency  Expected value of the statistic is the true population parameter  Expected value of dispersion in a sample is an underestimate of the true population value 41
42. 42.  True population size is N  Sample size n < N (e.g., n=100)  Correction factor : 𝑛 𝑛−1  For n=100 the correction factor is ~ 1.01  For n=100.000 our correction factor is ~1.00001  Estimate PopulationVar: ( 𝑛 𝑛−1 ) ∗ (𝑥 𝑖−𝜇𝑛 𝑖=1 ) 𝑛 42
43. 43.  Specify the range of values that have a high probability of containing the true population parameter  Confidence level α: the probability that confidence interval contains true population parameter 43
44. 44.  CI = sample statistic + MOE  MOE = SE * Critical value  MOE = 𝜎 𝑛 ∗ 𝑧 𝛼/2  CriticalValue: how far away from the mean must a point lie in order to be considered as “extreme” or “unexpected”? 44 n … sample size σ … standard deviation z α/2 … confidence coefficient
45. 45. 45
46. 46. Area under the curve is 0.475 What’s the z- score? 46
47. 47.  Select 1000 fb-user randomly  Average number of bar visits per year X = 78  Standard Deviation: (𝑥 𝑖−𝜇𝑛 𝑖=1 ) 2 𝑛 = 30  Confidence level is 95%  divide 0.95 by 2 to get 0.475  Check out the z table  z = 1.98  MOE = 𝜎 𝑛 ∗ 𝑧 𝛼/2 = 30 1000 ∗ 1.98= 1.88  78 +/- 1.88 CI: [76.12 ; 79.88] 47
48. 48.  Exact CI can only be computed when the sampling distribution and SD of sampling distribution (i.e., SE) are known  Otherwise we have to estimate the Standard Error (SE)  Bootstrap 48
49. 49.  Sampling with replacement  Population is unknown  But we observe one sample from the population of size n=4: {2, 3, 8, 8}  We use this sample to generate a large number of bootstrap samples of size n: ▪ 8, 8, 8, 3 ▪ 3, 3, 8, 2 ▪ …  Compute statistic (e.g. ,mean) for each bootstrap sample  Estimate SE from the bootstrap distribution 49
50. 50. 50 Population Sample Bootstrap Sample Bootstrap Sample Bootstrap Sample Bootstrap Sample Calculate statistic for each bootstrap sample Statistic +/- MOE MOE for 95% CI = 2 * SE Bootstrap Distribution Standard Error (SE): SD of bootstrap distribution
51. 51.  Randomly selected sample of fb-user  Have they ever checked in at a nightclub?  Democrats: 100/1000 yes  Republican: 90/1000 yes  Do the nightlife preferences differ significantly across political parties?  Give 95% CI for difference in proportions 51
52. 52.  dems = rep( c(0,1), c(1000-100, 100) )  repubs = rep( c(0,1), c(1000-90, 90) )  mean(dems) #0.1  mean(repubs) #0.09  del.p = mean(dems) - mean(repubs) #0.01 (point estimate)  reps = replicate( 1000, { ds = sample( dems, 1000, replace=TRUE ) rs = sample( repubs, 1000, replace=TRUE ) mean( ds ) - mean( rs ) } )  SE = sd( reps ) # 0.0131  c( del.p - 2*SE, del.p + 2*SE ) #-0.0162 0.0362 (interval estimate) 52
53. 53.  H1: political party affects the nightlife-preferences  H0: political party does not affects the nightlife- preferences  Proportion of users who visited nightclubs not matter which party they belong to: 190/2000 = 0.095  If political affinities have no effect, we would expect the following frequencies: 53 Democrats Republicans yes 100 90 190 no 900 910 1810 Democrats Republicans yes 95 95 190 no 905 905 1810
54. 54.  χ2= 𝑜−𝑒 2 𝑒 = 0.5815  DF = (number of rows – 1) x (number of columns – 1) = 1  Critical value of χ2 at 5% significance and 1 DF is 3.84  Our χ2 does not exceed the critical value  We cannot reject H0 54 Democrats Republicans yes 100 90 190 no 900 910 1810
55. 55.  If α=0.05 then 95% of all values fall in this interval  Two-tail test:  2.5% of values in the upper tail and 2.5% of the lower tail are considered as so extreme that we reject H0 if we observe them 55
56. 56.  Test if democrats on fb, on average, have more than 60 bar visits per year  H1: µ > 60  H0: µ <= 60  Random sample of 20 democratic fb-user:  {65 73 51 67 48 80 69 53 59 62 71 67 64 78 65 490 80 60 51 70}  Sample mean 𝜇=64.1  Assume we know SD in population = 10  𝑧 = 𝜇− 𝜇 𝑆𝐸 𝑆𝐸 = 𝑆𝐷 𝑛 𝑧 = 64.1−60 10/ 20 = 1.8336 56
57. 57.  Would we expect that? How extreme is this observation?  If H0 is true (mean<=60)  in which area around the mean do 95% of all points lie  Pick alpha level α=0.05  that’s the maximum probability where you reject the null hypothesis if the null hypothesis is true  Right-tail test: find our critical value for 0.45 using the z-distribution  If the z-score of our observed data exceed this value we have to reject H0 57 1.8336 > 1.645  reject the null hypothesis
58. 58.  Large Effects, Small Samples:  In small samples it is easy to overestimate an effect which might have happened by chance  Small Effects, Large Samples:  The smaller the effect you want to measure the larger the sample size you need to prove it significant!  Example: Assume a coin is biased: 10% head and 90% tail  Tossing the coin 10 times should be enough to convince people that the coin is biased.  Example: Assume a coin is biased: 51% head and 49% tail  Minimum sample size increases with decreasing effect size which one wants to demonstrate 58
59. 59.  The more we analyze, the more we find by chance!  If you calculate correlation between 10 variables (i.e., 44 different correlation coefficients) you should expect that at least 2 correlations are significant with p < 0.05 by chance (one in every 20)  Corrections or adjustments for the total number of comparison are needed! 59
60. 60.  Many tests such as z-test, t-test, ANOVA make the normality assumption.  If true population is very skewed (e.g. power law) the sampling distribution of the statistic will not be normal  Nonparametric methods like sign-test use e.g. median rather than the mean  Hypothesis about the median of the true population (e.g. H1: median < 100, H0: median = 100)  Count number of measurements that favor the null hypothesis  If H0 is true half of the measurement should fall on each side. 60
61. 61. Data Collection Data Preprocessing Data Mining DataAnalysis  Statistical Inference DataVisualization Machine Learning Data Preservation
62. 62.  Aim  Find a function that describes the relation between X (e.g. bar visits) andY (e.g. new friends)  Given X predictY  Problem  Infinite number of ways X andY could be related  Idea  Reduce space of possible function and start with the simplest one (linear relation)  Y= 𝑏0 + 𝑏1 𝑋 62
63. 63.  Y = 2 + 0.5 X 63 6 4 2 0 Y X 0 2 4 6 8
64. 64.  Use Gradient Descent to minimize Cost function C 𝑏0, 𝑏1  C 𝑏0, 𝑏1 = 1 2𝑁 (𝑌𝑖−𝑌𝑖)2𝑁 𝑖=1  C 𝑏0, 𝑏1 = 1 2𝑁 (𝑌𝑖 − 𝑏0 − 𝑏1 𝑋)2𝑁 𝑖=1  Start with some guess for 𝑏0, 𝑏1  Keep changing 𝑏0, 𝑏1 to reduce C 𝑏0, 𝑏1 until we hopefully end up at a minimum 64
65. 65. 𝑏0 ≔ 𝑏0 − 𝛼 𝜕 𝜕𝑏 𝑗 C 𝑏0, 𝑏1  𝑏1 ≔ 𝑏1 − 𝛼 𝜕 𝜕𝑏 𝑗 C 𝑏0, 𝑏1  Simultaneous updates of b0 and b1 65 Derivative of cost function informs us about the slope of the cost function Learning rate
66. 66. 66 C(b) b
67. 67.  Residuals: deviation between the observed and the predicted values  Residual sum of squares: 67 Is this a good measure? No it depends on the number of observations N What if we multiply it with 1/N?
68. 68.  𝑦𝑖… observed value  𝑦 … value predicted by the model  𝑦 … mean of observed data 68 Total variability in the outcome that needs to be explained Unexplained variability! Residuals: difference between the observed value and the estimated value Proportion of the total variability unexplained by the model
69. 69.  Independent variable is binary (e.g., went to nightclub or not)  We can group users by number of new friends year (20-25, 25-30, 30-35, etc.) and compute the proportion of people with high “nightclub-probability” 69
70. 70.  Logistic Regression:  Maximum Likelihood Estimator  Estimate unknown coefficients by maximizing the log likelihood function  Coefficient is interpreted as the rate of change in the "log odds" as X changes 70 ln 𝑃(𝑌 = 1) 1 − 𝑃(𝑌 = 1) = 𝑏0 + 𝑏1X + ϵ
71. 71. Simple Example: You have a coin that you know is biased towards heads and you want to know what the probability of heads (p) is. We want to estimate the unknown parameter p! 71
72. 72. You flip the coin 10 times and the coin comes up head 7 times. What’s your best guess for p? 72
73. 73. 3737 )1( !3!7 !10 )1( 7 10 )heads7( ppppP          Find the value for p that makes our data most likely! The probability of observing 7 times head when tossing a coin 10 times is given by this binomial distribution: 73
74. 74. )1log(3log7 !3!7 !10 loglog ppLikelihood    Set the derivative equal to 0 and solve for p. Derivative with respect to p. pp Likelihood dp d    1 37 0log 10 7 107377 3)1(70 )1( 3)1(7 0 1 37         p ppp pp pp pp pp *derivative of a constant is 0 *derivative 7f(x)=7f '(x) *derivative of log x is 1/x 3737 )1( !3!7 !10 )1( 7 10 ppppLikelihood          74 web.stanford.edu/~kcobb/hrp261/lecture4.ppt
75. 75. 267.)3(.)7(.120)3(.)7(. 7 10 LikelihoodtheofValue 3737        Likelihood of observing 7 times head when tossing a biased coin with p(head) = 0.7 and p(tail)=0.3 10 times is: 75
76. 76.  Linear Regression (R-squared)  Logistic Regression (pseudo R-squared) 76
77. 77. you can “prove” anything with graphics Data Collection Data Preprocessing DataAnalysis DataVisualization Data Preservation
78. 78. 78
79. 79. 79 http://www.motherjones.com/kevin-drum/2012/01/lying-charts-global-warming-edition
80. 80. 80 http://www.motherjones.com/kevin-drum/2012/01/lying-charts-global-warming-edition
81. 81.  Be careful when drawing conclusions from graphs  Size of effect shown in graphic != Size of effect in sample data != Size of the effect in the true population  Scale Disorting (e.g., bar charts not starting with zero)  Snapshot  … 81
82. 82. Data Collection Data Preprocessing DataAnalysis DataVisualization Data Preservation
83. 83.  GESIS Data Archives & Data Centers  Preserve research data and make them accessible for reuse.  Competencies and infrastructure ▪ e.g. https://datorium.gesis.org/xmlui/  CESSDA:  umbrella organisation for the European national data archives (http://www.cessda.net/)  Re3data  browse data archives by topic: http://www.re3data.org/ 83 DPC Digital Preservation Handbook: http://www.dpconline.org/advice/preservationhandbook
84. 84.  Legal and regulatory framework  including open access and licenses  Incentives to share data  Credentials? Citation principles under development (see e.g. http://www.datacite.org/).  Long term preservation strategies  software and hardware changes, documentation, metadata and retrieval/access Data preservation starts at an individual level Reasons for data loss often on an individual level, e.g. broken hardware, researchers leaving a group. 84
85. 85. http://claudiawagner.info/teaching/WebSciSS2014/
86. 86.  Vasant Dhar. Data Science and Prediction. In: Communications of the ACM, December 2013,Vol. 56, No. 12, pp. 64-73  Anand Rajaraman, Jeffrey Ullman, Jure Leskovec, Mining of Massive Datasets, Cambridge University Press (free download)  Jeffrey Stanton, Introduction to Data Science (free download)  Steffen Staab, Data Science Course University Koblenz-Landau, https://www.uni-koblenz-landau.de/campus- koblenz/fb4/west/teaching/ss14/data-science/data-science1  Serious Stats,Thom Baguley 86