Machine Learning for the rescue

Gathering the data is not a problem today. The bigger challenge is to understand these informations and draw some conclusions. Fortunately, we can use some techniques like machine learning to „teach” computer how to learn from our data. Fast artificial neural networks, random forests, SVMs, classification, clustering - just to name a few concepts ready to use… We will apply all these solutions to PHP application to deliver automatic insights/predictions and create a real business value for a client. By the end of this session you will be familiar with Machine Learning ideas and prepared to solve unsolvable problems in PHP.

Transcript

1. RESCUE
   Machine Learning for a
   Mariusz Gil
   

   View Slide

2. View Slide

3. View Slide

4. CLIENT PROBLEM
   

   View Slide

5. 1M BACKLINKS
   CLASSIFY THEM
   

   View Slide

6. View Slide

7. OK
   NOT OK
   I DON’T CARE
   

   View Slide

8. OK
   NOT OK
   I DON’T CARE
   

   View Slide

9. OK
   NOT OK
   I DON’T CARE
   

   View Slide

10. View Slide

11. T(URL) → [1, 2, 3, …]
    

    View Slide

12. IF-OLOGY
    UGLY CODE
    FOR POC
    1ST APPROACH
    

    View Slide

13. I DON’ KNOW
    

    View Slide

14. NAIVE
    MACHINE LEARNING
    2ND APPROACH
    

    View Slide

15. NAIVE
    MACHINE LEARNING
    2ND APPROACH
    

    View Slide

16. DATA ML
    TASK
    SEND TO
    RESULTS
    CALCULATE
    

    View Slide

17. View Slide

18. RECIPE FOR A FAILURE
    DOING WITHOUT KNOWING
    

    View Slide

19. DATA ORIENTED
    MACHINE LEARNING
    WORKFLOW
    3RD APPROACH, FINAL
    

    View Slide

20. A COMPUTER PROGRAM
    IS SAID TO LEARN FROM EXPERIENCE E
    WITH RESPECT TO SOME CLASS OF TASKS T
    AND PERFORMANCE MEASURE P
    IF ITS PERFORMANCE AT TASKS IN T,
    AS MEASURED BY P,
    IMPROVES WITH EXPERIENCE E
    

    View Slide

21. DATA ML
    TASK
    PREPARED, INPUT FOR
    RESULTS
    WITH PERFORMANCE
    EXPERIENCE FEEDBACK LOOP
    LEARNING, VALIDATING
    

    View Slide

22. ML
    TASK
    CLASSIFICATION
    REGRESSION
    CLUSTERING
    DIMENSIONALITY REDUCTION
    ASSOCIATION RULES
    

    View Slide

23. EXAMPLE TIME :)
    

    View Slide

24. FAST ARTIFICIAL
    NEURAL NETWORK
    CLASSIFICATION
    

    View Slide

25. View Slide

26. 80 2 4
    2 1
    1 0 0 0
    1 9
    1 0 0 0
    1 8
    1 0 0 0
    9 8
    1 0 0 0
    4 3
    1 0 0 0
    5 8
    1 0 0 0
    5 1
    1 0 0 0
    9 10
    1 0 0 0
    4 7
    1 0 0 0
    5 9
    …
    

    View Slide

27. View Slide

28. $num_input = 2;
    $num_output = 4;
    $num_layers = 3;
    $num_neurons_hidden = 4;
    $desired_error = 0.001;
    $max_epochs = 500000;
    $epochs_between_reports = 1000;
    $ann = fann_create_standard($num_layers, $num_input, $num_neurons_hidden, $num_output);
    if ($ann) {
    fann_set_activation_function_hidden($ann, FANN_SIGMOID_SYMMETRIC);
    fann_set_activation_function_output($ann, FANN_SIGMOID_SYMMETRIC);
    $filename = dirname(__FILE__) . "/coordination_system.data";
    if (fann_train_on_file(
    $ann,
    $filename,
    $max_epochs,
    $epochs_between_reports,
    $desired_error
    )) {
    fann_save($ann, dirname(__FILE__) . "/coordination_system.net");
    }
    fann_destroy($ann);
    }
    

    View Slide

29. $train_file = (dirname(__FILE__) . "/coordination_system.net");
    $ann = fann_create_from_file($train_file);
    if ($ann) {
    $input = array($argv[1], $argv[2]);
    $calc_out = fann_run($ann, $input);
    fann_destroy($ann);
    }
    

    View Slide

30. SUPERVISED
    LEARNING
    

    View Slide

31. SUPPORT
    VECTOR MACHINES
    CLASSIFICATION
    

    View Slide

32. A SUPPORT VECTOR MACHINE
    PERFORMS CLASSIFICATION
    BY FINDING THE HYPERPLANE
    THAT MAXIMIZES THE MARGIN
    BETWEEN THE GIVEN CLASSES
    

    View Slide

33. SUPERVISED
    LEARNING
    

    View Slide

34. K-MEANS
    CLUSTERING
    

    View Slide

35. IRIS DATASET
    1936, RONALD FISHER
    

    View Slide

36. View Slide

37. View Slide

38. View Slide

39. require 'vendor/autoload.php';
    use Phpml\Dataset\Demo\Iris;
    use Phpml\Clustering\KMeans;
    $dataset = new Iris();
    $kmeans = new KMeans(3);
    echo 'Dataset size: ' . count($dataset->getSamples()) . PHP_EOL;
    $clusters = $kmeans->cluster($dataset->getSamples());
    foreach ($clusters as $i => $cluster) {
    echo 'Cluster #' . $i . ' :' . count(($cluster)) . PHP_EOL;
    }
    

    View Slide

40. $ php -f ./iris-clustering.php
    Dataset size: 150
    Cluster #0 :39
    Cluster #1 :50
    Cluster #2 :61
    $ php -f ./iris-clustering.php
    Dataset size: 150
    Cluster #0 :38
    Cluster #1 :50
    Cluster #2 :62
    $ php -f ./iris-clustering.php
    Dataset size: 150
    Cluster #0 :39
    Cluster #1 :50
    Cluster #2 :61
    $ php -f ./iris-clustering.php
    Dataset size: 150
    Cluster #0 :96
    Cluster #1 :24
    Cluster #2 :30
    

    View Slide

41. RESULTS STABILITY
    

    View Slide

42. UNSUPERVISED
    LEARNING
    

    View Slide

43. RECIPE FOR A FAILURE
    DON’T YOU KNOW YOUR DATA?
    

    View Slide

44. PREDICTING VALUES
    REGRESSION
    

    View Slide

45. HOW MANY BRITISH POUNDS… EURO
    I SHOULD EARN AS DEVELOPER
    ACCORDING TO MY SKILLSET?
    

    View Slide

46. | age | linkedin_php | salary |
    |-----|--------------|--------|
    | 20 | 0 | 2000 |
    | 26 | 8 | 3975 |
    | 30 | 10 | 4000 |
    

    View Slide

47. YEARS →
    LINKEDIN PHP →
    

    View Slide

48. require 'vendor/autoload.php';
    use Phpml\Dataset\ArrayDataset;
    use Phpml\Regression\LeastSquares;
    $dataset = new ArrayDataset(
    [
    [20, 0],
    [26, 8],
    [30, 10],
    ],
    [
    2000,
    3975,
    4000,
    ]
    );
    $regression = new LeastSquares();
    $regression->train($dataset->getSamples(), $dataset->getTargets());
    echo $regression->predict(array_slice($argv, 1)) . PHP_EOL;
    

    View Slide

49. | age | city_size | linkedin_php | salary |
    |-----|-----------|--------------|--------|
    | 20 | 900000 | 0 | 2000 |
    | 20 | 400000 | 0 | 1800 |
    | 25 | 450000 | 8 | 3700 |
    | 26 | 900000 | 8 | 3975 |
    | 30 | 100000 | 10 | 4000 |
    | 30 | 500000 | 10 | 3500 |
    

    View Slide

50. SUPERVISED
    LEARNING
    

    View Slide

51. TECHNOLOGY
    

    View Slide

52. View Slide

53. View Slide

54. View Slide

55. View Slide

56. View Slide

57. …JVM, PYTHON
    

    View Slide

58. View Slide

59. ML IS NOT
    A SINGLE RUN
    OF ALGORITHM
    

    View Slide

60. IT’S A PROCESS
    

    View Slide

61. ML
    PROCESS
    DEFINE A PROBLEM
    ANALYZE YOUR DATA
    UNDERSTAND YOUR DATA
    PREPARE DATA FOR ML
    SELECT & RUN ALGO(S)
    TUNE ALGO(S) PARAMETERS
    SELECT FINAL MODEL
    VALIDATE FINAL MODEL
    

    View Slide

62. ML
    PROCESS
    DEFINE A PROBLEM
    ANALYZE YOUR DATA
    UNDERSTAND YOUR DATA
    PREPARE DATA FOR ML
    SELECT & RUN ALGO(S)
    TUNE ALGO(S) PARAMETERS
    SELECT FINAL MODEL
    VALIDATE FINAL MODEL
    

    View Slide

63. | age | city_size | linkedin_php | salary |
    |-----|-----------|--------------|--------|
    | 20 | 900000 | 0 | 2000 |
    | 20 | 400000 | 0 | 1800 |
    | 25 | 450000 | 8 | 3700 |
    | 26 | 900000 | 8 | 3975 |
    | 30 | 100000 | 10 | 4000 |
    | 30 | 500000 | 10 | 3500 |
    

    View Slide

64. | age | city_size | linkedin_php | salary | currency |
    |-----|-----------|--------------|--------|----------|
    | 20 | 900000 | 0 | 2000 | EUR |
    | 20 | 400000 | 0 | 1800 | USD |
    | 25 | 450000 | 8 | 3700 | USD |
    | 26 | 900000 | 8 | 3975 | USD |
    | 30 | 100000 | 10 | 4000 | USD |
    | 30 | 500000 | 10 | 3500 | USD |
    

    View Slide

65. ONE MORE THING…
    

    View Slide

66. PHPCON CFP WILL BE
    CLOSED TOMORROW!
    http://phpcon.pl/2016/en/cfp
    

    View Slide

67. THANKS
    mariuszgil
    HAPPY LEARNING YOUR MACHINES!
    

    View Slide