Download as PDF, PPTX
More Related Content
Similar to Apache Mahout Algorithms
Apache Mahout Algorithms
- 1.
- 2. Agenda - Introduction - CollaborativeFiltering - Map/Reduce - Clustering - Demo
- 3. What mahout means Elephantrider in Hindi
- 4. What Apache Mahoutis - Java, Hadoop - Collaborative Filtering - Mahout In Action - user@mahout.apache.org - 0.9 (1-Feb-2014)
- 5.
- 6. Mahout in ApacheFoundation
- 7. overstock.com saves $2ma year Judd Bagley Saum Noursalehi
- 8. Others - Weka (MachineLearning Library) - Lenskit (Grouplens) - EasyRec (RestAPI) - Write yourself:)
- 9. Need to knowML?
- 10. Need to knowML? hadoop.jar mahout-core-0.8-job.jar org.apache.mahout.cf.taste.hadoop.item. RecommenderJob -Dmapred.input.dir=input/input.txt -Dmapred.output.dir=output --usersFile input/users.txt --booleanData
- 11.
- 12.
- 13.
- 14.
- 15. Collaborative Filtering - Dataformat = userId, itemId, rating - Create Model + Predict
- 16. Item Based -Similarity Matrix (Item-Item)
- 17. Item Based -Predict - Weighted Sum: r^(3,1) = 2 * 0.91 + ...
- 18.
- 19. Item Based.. Whyin Mahout - Generic recommender like User Based - User Based similarity matrix is heavier
- 20.
- 21.
- 22.
- 23.
- 24.
- 25. m * n→ m * k + n * k 10M → 100K + 10K Lets say; m=10K n = 1K k=10 Singular Value Decomposition (SVD)
- 26. SVD k=3 λ=0.1a=40 c.a=1
- 27. SVD k=3 λ=0.1a=40 c.a=1
- 28. SVD k=3 λ=0.1a=40 c.a=10
- 29. SVD.. Why inMahout - Won Netflix Prize - Parallelizable by row, column
- 30. Map / ReduceMapper 1.txt 2.txt Hello Hello Hello
- 31. Map / ReduceMapper
- 32. Map / ReduceMapper Map1 Map2 Hello,1 Hello,1 Hello,1
- 33. Map / ReduceReducer
- 34. Map / ReduceReducer Hello,3
- 35. Map / ReduceItemBased
- 36. Map / ReduceItemBased hadoop.jar mahout-core-0.8-job.jar org.apache.mahout.cf.taste.hadoop.item. RecommenderJob -Dmapred.input.dir=input/input.txt -Dmapred.output.dir=output --usersFile input/users.txt --booleanData
- 37. Map / ReduceItemBased
- 38. Map / ReduceItemBased
- 39. Map / ReduceItemBased Map 1
- 40. Map / ReduceItemBased Reduce 1
- 41. Map / ReduceItemBased Reduce 1
- 42. Map / ReduceItemBased Map 2
- 43. Map / ReduceItemBased Reduce 2
- 44. Map / ReduceItemBased
- 45. Map / Reduce..Why in Mahout
- 46. Clustering - KMeans Clustering(SM,MR) - Fuzzy kMeans (SM,MR) - Canopy Clustering (SM,MR) - Dirichlet (SM,MR)
- 47.
- 48.
- 49.
- 50.
- 51.
- 52. Clustering.. Why inMahout - Sparsity - ~10m of 11m users registered 1 Sony product
- 53. Clustering.. Why inMahout - Group Recommendation - Cluster Based Recommendation
- 54. Create WishList Experience -Mahout (SVD) - Play - Heroku - MongoLab - Rest http://recommenderplaybbs.herokuapp.com/
- 55.