Machine Learning for the rescue

Slide 1

Slide 1 text

RESCUE Machine Learning for a Mariusz Gil

Slide 2

Slide 2 text

No content

Slide 3

Slide 3 text

No content

Slide 4

Slide 4 text

CLIENT PROBLEM

Slide 5

Slide 5 text

1M BACKLINKS CLASSIFY THEM

Slide 6

Slide 6 text

No content

Slide 7

Slide 7 text

OK NOT OK I DON’T CARE

Slide 8

Slide 8 text

OK NOT OK I DON’T CARE

Slide 9

Slide 9 text

OK NOT OK I DON’T CARE

Slide 10

Slide 10 text

No content

Slide 11

Slide 11 text

T(URL) → [1, 2, 3, …]

Slide 12

Slide 12 text

IF-OLOGY UGLY CODE FOR POC 1ST APPROACH

Slide 13

Slide 13 text

I DON’ KNOW

Slide 14

Slide 14 text

NAIVE MACHINE LEARNING 2ND APPROACH

Slide 15

Slide 15 text

NAIVE MACHINE LEARNING 2ND APPROACH

Slide 16

Slide 16 text

DATA ML TASK SEND TO RESULTS CALCULATE

Slide 17

Slide 17 text

No content

Slide 18

Slide 18 text

RECIPE FOR A FAILURE DOING WITHOUT KNOWING

Slide 19

Slide 19 text

DATA ORIENTED MACHINE LEARNING WORKFLOW 3RD APPROACH, FINAL

Slide 20

Slide 20 text

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

Slide 21

Slide 21 text

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

Slide 22

Slide 22 text

ML TASK CLASSIFICATION REGRESSION CLUSTERING DIMENSIONALITY REDUCTION ASSOCIATION RULES

Slide 23

Slide 23 text

EXAMPLE TIME :)

Slide 24

Slide 24 text

FAST ARTIFICIAL NEURAL NETWORK CLASSIFICATION

Slide 25

Slide 25 text

No content

Slide 26

Slide 26 text

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 …

Slide 27

Slide 27 text

No content

Slide 28

Slide 28 text

Slide 29

Slide 29 text

Slide 30

Slide 30 text

SUPERVISED LEARNING

Slide 31

Slide 31 text

SUPPORT VECTOR MACHINES CLASSIFICATION

Slide 32

Slide 32 text

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

Slide 33

Slide 33 text

SUPERVISED LEARNING

Slide 34

Slide 34 text

K-MEANS CLUSTERING

Slide 35

Slide 35 text

IRIS DATASET 1936, RONALD FISHER

Slide 36

Slide 36 text

No content

Slide 37

Slide 37 text

No content

Slide 38

Slide 38 text

No content

Slide 39

Slide 39 text

getSamples()) . PHP_EOL; $clusters = $kmeans->cluster($dataset->getSamples()); foreach ($clusters as $i => $cluster) { echo 'Cluster #' . $i . ' :' . count(($cluster)) . PHP_EOL; }

Slide 40

Slide 40 text

$ 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

Slide 41

Slide 41 text

RESULTS STABILITY

Slide 42

Slide 42 text

UNSUPERVISED LEARNING

Slide 43

Slide 43 text

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

Slide 44

Slide 44 text

PREDICTING VALUES REGRESSION

Slide 45

Slide 45 text

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

Slide 46

Slide 46 text

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

Slide 47

Slide 47 text

YEARS → LINKEDIN PHP →

Slide 48

Slide 48 text

train($dataset->getSamples(), $dataset->getTargets()); echo $regression->predict(array_slice($argv, 1)) . PHP_EOL;

Slide 49

Slide 49 text

| 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 |

Slide 50

Slide 50 text

SUPERVISED LEARNING

Slide 51

Slide 51 text

TECHNOLOGY

Slide 52

Slide 52 text

No content

Slide 53

Slide 53 text

No content

Slide 54

Slide 54 text

No content

Slide 55

Slide 55 text

No content

Slide 56

Slide 56 text

No content

Slide 57

Slide 57 text

…JVM, PYTHON

Slide 58

Slide 58 text

No content

Slide 59

Slide 59 text

ML IS NOT A SINGLE RUN OF ALGORITHM

Slide 60

Slide 60 text

IT’S A PROCESS

Slide 61

Slide 61 text

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

Slide 62

Slide 62 text

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

Slide 63

Slide 63 text

Slide 64

Slide 64 text

| 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 |