![Power of data. Simplicity of design. Speed of innovation.
IBM Spark
spark.tc
spark.tc
Power of data. Simplicity of design. Speed of innovation.
IBM Spark
What about Average?
Overall AVG (
[3, 1]
((3 + 5) + (5 + 7))
20
[5, 1] == ----------------------- == --- == 5
[5, 1]
((1 + 2) + 1)
4
[7, 1]
)
11
value
count
Pairwise AVG
(3 + 5) (5 + 7) 8 12 20
------- + ------- == --- + --- == --- == 10 != 5
2
2
2 2
2
Divide, Add, Divide?
Not
Composable
Single Divide at the End?
Doesn’t need to be Composable!
AVG (3, 5, 5, 7) == 5
Add, Add, Add?
Composable!](https://image.slidesharecdn.com/5thathensbigdatameetup-pipelineioworkshop-181127115525/75/5th-Athens-Big-Data-Meetup-PipelineIO-Workshop-Real-Time-Training-and-Deploying-Spark-ML-Recommendations-With-Kafka-and-NetflixOSS-11-2048.jpg)
![Power of data. Simplicity of design. Speed of innovation.
IBM Spark
spark.tc
spark.tc
Power of data. Simplicity of design. Speed of innovation.
IBM Spark
Feature Engineering
Dimension Reduction
Reduce number of features in feature space
Principle Component Analysis (PCA)
Help find principle features that best describe variance in data
Peel the dimensional layers back until you describe the data
One-Hot Encoding
Convert nominal categorical feature values to 0’s, 1’s
Remove numerical relationship between the categories
Bears
-> 1
Bears ->
[1,0,0]
49’ers -> 2
-->
49’ers ->
[0,1,0]
Steelers-> 3
Steelers-> [0,0,1]
28
1 binary column
per category](https://image.slidesharecdn.com/5thathensbigdatameetup-pipelineioworkshop-181127115525/75/5th-Athens-Big-Data-Meetup-PipelineIO-Workshop-Real-Time-Training-and-Deploying-Spark-ML-Recommendations-With-Kafka-and-NetflixOSS-28-2048.jpg)
![Power of data. Simplicity of design. Speed of innovation.
IBM Spark
spark.tc
spark.tc
Power of data. Simplicity of design. Speed of innovation.
IBM Spark
Item-to-Item Collaborative Filtering
Made famous by Amazon Paper ~2003
Problem
As # of users grew, user-item collab filtering didn’t scale
Solution
Offline/Batch
Generate itemId -> List[userId] vectors
Online/Real-time
For each item in cart, recommend similar items from vector space
36](https://image.slidesharecdn.com/5thathensbigdatameetup-pipelineioworkshop-181127115525/75/5th-Athens-Big-Data-Meetup-PipelineIO-Workshop-Real-Time-Training-and-Deploying-Spark-ML-Recommendations-With-Kafka-and-NetflixOSS-36-2048.jpg)
![Power of data. Simplicity of design. Speed of innovation.
IBM Spark
spark.tc
spark.tc
Power of data. Simplicity of design. Speed of innovation.
IBM Spark
Spark Approximations
Spark Core
RDD.count*Approx()
Spark SQL
PartialResult
approxCountDistinct(column), HyperLogLogPlus
Spark ML
Stratified sampling
PairRDD.sampleByKey(fractions: Double[ ])
DIMSUM sampling
Probabilistic sampling reduces amount of comparison shuffle
RowMatrix.columnSimilarities(threshold)
Spark Streaming
A/B testing
StreamingTest.setTestMethod(“welch”).registerStream(dstream)
60](https://image.slidesharecdn.com/5thathensbigdatameetup-pipelineioworkshop-181127115525/75/5th-Athens-Big-Data-Meetup-PipelineIO-Workshop-Real-Time-Training-and-Deploying-Spark-ML-Recommendations-With-Kafka-and-NetflixOSS-60-2048.jpg)
Uploaded byAthens Big Data
5th Athens Big Data Meetup - PipelineIO Workshop - Real-Time Training and Deploying Spark ML Recommendations With Kafka and NetflixOSS
The document discusses the capabilities of IBM Spark, focusing on data processing techniques like streaming, machine learning, and recommendations. It highlights various meetups and events, as well as methodologies for scaling, similarity measurements, and personalized recommendations. Essential topics include feature engineering, clustering, and collaborative filtering related to big data applications.
More Related Content
Similar to 5th Athens Big Data Meetup - PipelineIO Workshop - Real-Time Training and Deploying Spark ML Recommendations With Kafka and NetflixOSS
5th Athens Big Data Meetup - PipelineIO Workshop - Real-Time Training and Deploying Spark ML Recommendations With Kafka and NetflixOSS
- 1. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Spark & Recommendations Spark, Streaming, Machine Learning, Graph Processing, Approximations, Probabilistic Data Structures, NLP Apache Spark Maryland Meetup Thanks to Tetra Concepts & Jailbreak Brewing Co!! Feb 22nd, 2016 Chris Fregly Principal Data Solutions Engineer We’re Hiring! (Only Nice People) advancedspark.com!
- 2. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Who Am I? 2 Streaming Data Engineer Netflix OSS Committer Data Solutions Engineer Apache Contributor Principal Data Solutions Engineer IBM Technology Center Meetup Organizer Advanced Apache Meetup Book Author Advanced . Due 2016
- 3. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Recent World Tour: Freg-a-Palooza! London Spark Meetup (Oct 12th) Scotland Data Science Meetup (Oct 13th) Dublin Spark Meetup (Oct 15th) Barcelona Spark Meetup (Oct 20th) Madrid Big Data Meetup (Oct 22nd) Paris Spark Meetup (Oct 26th) Amsterdam Spark Summit (Oct 27th) Brussels Spark Meetup (Oct 30th) Zurich Big Data Meetup (Nov 2nd) Geneva Spark Meetup (Nov 5th) 3 Oslo Big Data Hadoop Meetup (Nov 19th) Helsinki Spark Meetup (Nov 20th) Stockholm Spark Meetup (Nov 23rd) Copenhagen Spark Meetup (Nov 25th) Istanbul Spark Meetup (Nov 26th) Budapest Spark Meetup (Nov 28th) Singapore Spark Meetup (Dec 1st) Sydney Spark Meetup (Dec 8th) Melbourne Spark Meetup (Dec 9th) Toronto Spark Meetup (Dec 14th)
- 4. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Advanced Apache Spark Meetup http://advancedspark.com Meetup Metrics Top 5 Most-active Spark Meetup! 2600 Members in just 6 mos!! 2600 Docker downloads (demos) Meetup Mission Deep-dive into Spark and related open source projects Surface key patterns and idioms Focus on distributed systems, scale, and performance 4
- 5. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark spark.tc Live, Interactive Demo!! Audience Participation Required (cell phone or laptop) 5
- 6. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark demo.advancedspark.com End User -> ElasticSearch -> Spark ML -> Data Scientist -> 6 <- Kafka <- Spark Streaming <- Cassandra, Redis <- Zeppelin, iPython
- 7. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Presentation Outline Scaling with Parallelism and Composability Similarity and Recommendations When to Approximate Common Algorithms and Data Structures Common Libraries and Tools Netflix Recommendations and Data Pipeline 7
- 8. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Scaling with Parallelism 8 Peter O(log n) O(log n)
- 9. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Scaling with Composability Max (a max b max c max d) == (a max b) max (c max d) Set Union (a U b U c U d) == (a U b) U (c U d) Addition (a + b + c + d) == (a + b) + (c + d) Multiply (a * b * c * d) == (a * b) * (c * d) Division?? 9
- 10. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark What about Division? Division (a / b / c / d) != (a / b) / (c / d) (3 / 4 / 7 / 8) != (3 / 4) / (7 / 8) (((3 / 4) / 7) / 8) != ((3 * 8) / (4 * 7)) 0.134 != 0.857 10 What were the Egyptians thinking?! Not Composable “Divide like an Egyptian”
- 11. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark What about Average? Overall AVG ( [3, 1] ((3 + 5) + (5 + 7)) 20 [5, 1] == ----------------------- == --- == 5 [5, 1] ((1 + 2) + 1) 4 [7, 1] ) 11 value count Pairwise AVG (3 + 5) (5 + 7) 8 12 20 ------- + ------- == --- + --- == --- == 10 != 5 2 2 2 2 2 Divide, Add, Divide? Not Composable Single Divide at the End? Doesn’t need to be Composable! AVG (3, 5, 5, 7) == 5 Add, Add, Add? Composable!
- 12. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Presentation Outline Scaling with Parallelism and Composability Similarity and Recommendations When to Approximate Common Algorithms and Data Structures Common Libraries and Tools Netflix Recommendations and Data Pipeline 12
- 13. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark spark.tc Similarity 13
- 14. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Euclidean Similarity Exists in Euclidean, flat space Based on Euclidean distance Linear measure Bias towards magnitude 14
- 15. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Cosine Similarity Angular measure Adjusts for Euclidean magnitude bias 15 Normalizes to unit vectors
- 16. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Jaccard Similarity Set similarity measurement Set intersection / set union -> Based on Jaccard distance Bias towards popularity 16
- 17. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Log Likelihood Similarity Adjusts for popularity bias Netflix “Shawshank” problem 17
- 18. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Word Similarity Based on edit distance Calculate char differences between words Deletes, transposes, replaces, inserts 18
- 19. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Document Similarity TD/IDF Term Freq / Inverse Document Freq Used by most search engines Word2Vec Words embedded in vector space nearby similars 19
- 20. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Similarity Pathway ie. Closest recommendations between 2 people 20
- 21. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Calculating Similarity Exact Brute-Force “All-pairs similarity” aka “Pair-wise similarity”, “Similarity join” Cartesian O(n^2) shuffle and comparison Approximate Sampling Bucketing (aka “Partitioning”, “Clustering”) Remove data with low probability of similarity Reduce shuffle and comparisons 21
- 22. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Bonus: Document Summary Text Rank aka “Sentence Rank” TF/IDF + Similarity Graph + PageRank Intuition Surface summary sentences (abstract) Most similar to all others (TF/IDF + Similarity Graph) Most influential sentences (PageRank) 22
- 23. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Similarity Graph Vertex is movie, tag, actor, plot summary, etc. Edges are relationships and weights 23
- 24. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Topic-Sensitive PageRank Graph diffusion algorithm Pre-process graph, add vector of probabilities to each vertex Probability of landing at this vertex from every other vertex 24
- 25. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark spark.tc Recommendations 25
- 26. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Basic Terminology User: User seeking recommendations Item: Item being recommended Explicit User Feedback: like, rating, movie view, profile read, search Implicit User Feedback: click, hover, scroll, navigation Instances: Rows of user feedback/input data Overfitting: Training a model too closely to the training data & hyperparameters Hold Out Split: Holding out some of the instances to avoid overfitting Features: Columns of instance rows (of feedback/input data) Cold Start Problem: Not enough data to personalize (new) Hyperparameter: Model-specific config knobs for tuning (tree depth, iterations) Model Evaluation: Compare predictions to actual values of hold out split Feature Engineering: Modify, reduce, combine features 26
- 27. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Features Binary: True or False Numeric Discrete: Integers Numeric: Real Values Binning: Convert Continuous into Discrete (Time of Day->Morning, Afternoon) Categorical Ordinal: Size (Small->Medium->Large), Ratings (1->5) Categorical Nominal: Independent, Favorite Sports Teams, Dating Spots Temporal: Time-based, Time of Day, Binge Viewing Text: Movie Titles, Genres, Tags, Reviews (Tokenize, Stop Words, Stemming) Media: Images, Audio, Video Geographic: (Longitude, Latitude), Geohash Latent: Hidden Features within Data (Collaborative Filtering) Derived: Age of Movie, Duration of User Subscription 27
- 28. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Feature Engineering Dimension Reduction Reduce number of features in feature space Principle Component Analysis (PCA) Help find principle features that best describe variance in data Peel the dimensional layers back until you describe the data One-Hot Encoding Convert nominal categorical feature values to 0’s, 1’s Remove numerical relationship between the categories Bears -> 1 Bears -> [1,0,0] 49’ers -> 2 --> 49’ers -> [0,1,0] Steelers-> 3 Steelers-> [0,0,1] 28 1 binary column per category
- 29. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Normalize and Standardize Features Goal Scale features to standard size Required by many ML algos Normalize Features Calculate L1 (or L2, etc) norm Divide elements by norm org.apache.spark.ml.feature.Normalizer Standardize Features Apply standard normal transformation Mean == 0 StdDev == 1 org.apache.spark.ml.feature.StandardScaler 29
- 30. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark spark.tc Non-Personalized Recommendations 30
- 31. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Cold Start Problem “Cold Start” problem New user, don’t know their preference, must show something! Movies with highest-rated actors Top K Aggregations Most desirable singles PageRank of likes and dislikes Facebook social graph Friend-based recommendations 31
- 32. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark spark.tc Personalized Recommendations 32
- 33. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Clustering (aka. Nearest Neighbors) User-to-User Clustering (User Behavior) Similar items viewed or rated Similar viewing pattern (ie. binge or casual) Item-to-Item Clustering (Item Description) Similar item tags/metadata (Jaccard Similiarity, Locality Sensitive Hash) Similar profile text and categories (TF/IDF, Word2Vec, NLP) Similar images/facial structures (Convolutional Neural Nets, Eigenfaces) 33 http://crockpotveggies.com/2015/02/09/automating-tinder-with-eigenfaces.htmMy OKCupid Profile My Hinge Profile Dating Site ->
- 34. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Bonus: NLP Conversation Bot 34 “If your responses to my generic opening lines are positive, I may read your profile.” Spark ML and Stanford CoreNLP: TF/IDF, DecisionTrees, Sentiment Analysis
- 35. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark User-to-Item Collaborative Filtering Matrix Factorization ① Factor the large matrix (left) into 2 smaller matrices (right) ② Smaller matrices, when multiplied, approximate original ③ Fill in the missing values with in the large matrix ④ Surface latent features from within user-item interaction 35
- 36. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Item-to-Item Collaborative Filtering Made famous by Amazon Paper ~2003 Problem As # of users grew, user-item collab filtering didn’t scale Solution Offline/Batch Generate itemId -> List[userId] vectors Online/Real-time For each item in cart, recommend similar items from vector space 36
- 37. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Presentation Outline Scaling with Parallelism and Composability Similarity and Recommendations When to Approximate Common Algorithms and Data Structures Common Libraries and Tools Netflix Recommendations and Data Pipeline 37
- 38. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark When to Approximate? Memory or time constrained queries Relative vs. exact counts are OK (# errors between then and now) Using machine learning or graph algos Inherently probabilistic and approximate Finding topics in documents (LDA) Finding similar pairs of users, items, words at scale (LSH) Finding top influencers (PageRank) Streaming aggregations Inherently sloppy collection (exactly once?) 38 Approximate as much as you can get away with! Ask for forgiveness later !!
- 39. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark When NOT to Approximate? If you’ve ever heard the term… “Sarbanes-Oxley” …at the office after 2002. 39
- 40. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Presentation Outline Scaling with Parallelism and Composability Similarity and Recommendations When to Approximate Common Algorithms and Data Structures Common Libraries and Tools Netflix Recommendations and Data Pipeline 40
- 41. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark spark.tc A Few Good Algorithms 41 You can’t handle the approximate!
- 42. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Common to These Algos & Data Structs Low, fixed size in memory Known error bounds Store large amount of data Less memory than Java/Scala collections Tunable tradeoff between size and error Rely on multiple hash functions or operations Size of hash range defines error 42
- 43. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark spark.tc Bloom Filter Set.contains(key): Boolean “Hash Multiple Times and Flip the Bits Wherever You Land” 43
- 44. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Bloom Filter Approximate set membership for key False positive: expect contains(), actual !contains() True negative: expect !contains(), actual !contains() Elements are only added, never removed 44
- 45. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Bloom Filter in Action 45 set(key) contains(key): Boolean Images by @avibryant TRUE -> maybe contains FALSE -> definitely does not contain.
- 46. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark spark.tc CountMin Sketch Frequency Count and TopK “Hash Multiple Times and Add 1 Wherever You Land” 46
- 47. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark CountMin Sketch (CMS) Approximate frequency count and TopK for key ie. “Heavy Hitters” on Twitter 47 Matei Zaharia Martin Odersky Donald Trump
- 48. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark CountMin Sketch In Action (TopK, Count) 48 Images derived from @avibryant Find minimum of all rows … … Can overestimate, but never underestimate Multiple hash functions (1 hash function per row) Binary hash output (1 element per column) x 2 occurrences of “Top Gun” for slightly additional complexity Top Gun Top Gun Top Gun (x 2) A Few Good Men Taps Top Gun (x 2) add(Top Gun, 2) getCount(Top Gun): Long Use Case: TopK movies using total views add(A Few Good Men, 1) add(Taps, 1) A Few Good Men Taps … … Overlap Top Gun Overlap A Few Good Men
- 49. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark spark.tc HyperLogLog Count Distinct “Hash Multiple Times and Uniformly Distribute Where You Land” 49
- 50. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark HyperLogLog (HLL) Approximate count distinct Slight twist Special hash function creates uniform distribution Error estimate 14 bits for size of range m = 2^14 = 16,384 hash slots error = 1.04/(sqrt(16,384)) = .81% 50 Not many of these
- 51. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark HyperLogLog In Action (Count Distinct) Use Case: Number of distinct users who view a movie 51 0 32 Top Gun: Hour 2 user 2001 user 4009 user 3002 user 7002 user 1005 user 6001 User 8001 User 8002 user 1001 user 2009 user 3005 user 3003 Top Gun: Hour 1 user 3001 user 7009 0 16 Uniform Distribution: Estimate distinct # of users by inspecting just the beginning 0 32 Top Gun: Hour 1 + 2 user 2001 user 4009 user 3002 user 7002 user 1005 user 6001 User 8001 User 8002 Combine across different scales user 7009 user 1001 user 2009 user 3005 user 3003 user 3001
- 52. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark spark.tc Locality Sensitive Hashing Set Similarity “Pre-process Items into Buckets, Compare Within Buckets” 52
- 53. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Locality Sensitive Hashing (LSH) Approximate set similarity Hash designed to cluster similar items Avoids cartesian all-pairs comparison Pre-process m rows into b buckets b << m Hash items multiple times Similar items hash to overlapping buckets Compare just contents of buckets Much smaller cartesian … and parallel !! 53
- 54. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark spark.tc DIMSUM Set Similarity “Pre-process and ignore data that is unlikely to be similar.” 54
- 55. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark DIMSUM “Dimension Independent Matrix Square Using MR” Remove vectors with low probability of similarity RowMatrix.columnSimiliarites(threshold) Twitter DIMSUM Case Study 40% efficiency gain over bruce-force Cosine Sim 55
- 56. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Presentation Outline Scaling with Parallelism and Composability Similarity and Recommendations When to Approximate Common Algorithms and Data Structures Common Libraries and Tools Netflix Recommendations and Data Pipeline 56
- 57. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Common Tools to Approximate Twitter Algebird Redis Apache Spark 57 Composable Library Distributed Cache Big Data Processing
- 58. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Twitter Algebird Rooted in Algebraic Fundamentals! Parallel Associative Composable Examples Min, Max, Avg BloomFilter (Set.contains(key)) HyperLogLog (Count Distinct) CountMin Sketch (TopK Count) 58
- 59. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Redis Implementation of HyperLogLog (Count Distinct) 12KB per item count 2^64 max # of items 0.81% error (Tunable) Add user views for given movie PFADD TopGun_HLL user1001 user2009 user3005 PFADD TopGun_HLL user3003 user1001 Get distinct count (cardinality) of set PFCOUNT TopGun_HLL Returns: 4 (distinct users viewed this movie) 59 ignore duplicates Tunable Union 2 HyperLogLog Data Structures PFMERGE TopGun_HLL Taps_HLL
- 60. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Spark Approximations Spark Core RDD.count*Approx() Spark SQL PartialResult approxCountDistinct(column), HyperLogLogPlus Spark ML Stratified sampling PairRDD.sampleByKey(fractions: Double[ ]) DIMSUM sampling Probabilistic sampling reduces amount of comparison shuffle RowMatrix.columnSimilarities(threshold) Spark Streaming A/B testing StreamingTest.setTestMethod(“welch”).registerStream(dstream) 60
- 61. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark spark.tc Demos! 61
- 62. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark spark.tc Counting Exact Count vs. Approx HyperLogLog, CountMin Sketch 62
- 63. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark HashSet vs. HyperLogLog (Memory) 63
- 64. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark HashSet vs. CountMin Sketch (Memory) 64
- 65. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark spark.tc Set Similarity Bruce Force vs. Locality Sensitive Hashing Similarity 65
- 66. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Brute Force Cartesian All Pair Similarity 66 47 seconds
- 67. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Locality Sensitive Hash All Pair Similarity 67 6 seconds
- 68. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark spark.tc Many More Demos! or Download Docker Clone Github 68 http://advancedspark.com
- 69. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Presentation Outline Scaling with Parallelism and Composability Similarity and Recommendations When to Approximate Common Algorithms and Data Structures Common Libraries and Tools Netflix Recommendations and Data Pipeline 69
- 70. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark spark.tc Netflix Recommendation & Data Pipeline From 5 Stars to Trending Now 70
- 71. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Netflix Has a Lot of Data Netflix has a lot of data about a lot of users and a lot of movies. Netflix can use this data to buy new movies. Netflix is global. Netflix can use this data to choose original programming. Netflix knows that a lot of people like politics and Kevin Spacey. 71 The UK doesn’t have White Castle. Renamed my favourite movie to: “Harold and Kumar Get the Munchies” My favorite movie: “Harold and Kumar Go to White Castle” Summary: Buy NFLX Stock! This broke my unit tests!
- 72. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark $1 Million Netflix Prize (2006-2009) Goal Improve movie predictions by 10% (RMSE) Dataset (userId, movieId, rating, timestamp) Test data withheld to calculate RMSE upon submission Winning algorithm 10.06% improvement (RMSE) Ensemble of 500+ ML combined with GBDT’s Computationally impractical 72
- 73. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Secrets to the Winning Algorithms Adjust for the following human bias… ① Alice Effect: rate lower than average user ② Inception Effect: rated higher than average movie ③ Overall mean rating of a movie ④ Number of people who have rated a movie ⑤ Mood, time of day, day of week, season, weather ⑥ Number of days since user’s first rating ⑦ Number of days since movie’s first rating 73
- 74. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Netflix Data Pipeline - Then 74 v1.0! v2.0!
- 75. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Netflix Data Pipeline - Now 75 v3.0! 8 million events per second
- 76. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Netflix Recommendation Pipeline 76 Throw away batch-generated user factors (U)
- 77. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Netflix Common ML Algorithms Logistic Regression Linear Regression Gradient Boosted Decision Trees Random Forest Matrix Factorization SVD Restricted Boltzmann Machines Deep Neural Nets Markov Models LDA Clustering 77 Ensembles
- 78. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Netflix Trending Now Time of day Personalized to user (viewing history, past ratings) Personalized to events (Valentine’s Day) 78 “VHS” Number of Plays Number of Impressions Calculate Take Rate
- 79. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Bonus: Pandora Time of Day Recs Work Days Play familiar music User is less likely accept new music Evenings and Weekends Play new music More like to accept new music 79
- 80. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Netflix Social Integration Post to Facebook after movie start (5 mins) Recommend without needing viewing history Helps with Cold Start problem 80
- 81. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Netflix Search No results? No problem… Show similar results! Empty searches are good! Explicit feedback for future recommendations Content to buy and produce! 81
- 82. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Bonus: Netflix in 2004 Netflix noticed people started to rate movies higher!? Why? Significant UI improvements made around that time Recommendation improvements (Cinematch) 82
- 83. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark Thank You!! Chris Fregly @cfregly IBM Spark Tech Center http://spark.tc San Francisco, California, USA http://advancedspark.com Sign up for the Meetup and Book Contribute to Github Repo Run all Demos using Docker Find me: LinkedIn, Twitter, Github, Email, Fax 83 Image derived from http://www.duchess-france.org/
- 84. Power of data.Simplicity of design. Speed of innovation. IBM Spark spark.tc Power of data. Simplicity of design. Speed of innovation. IBM Spark advancedspark.com @cfregly