Slide 1
Slide 1 text
TREND ANALYSIS AND MACHINE
LEARNING IN PHP
SYMFONYCON CLUJ 2017
@MICHAELCULLUMUK
Slide 2
Slide 2 text
TREND ANALYSIS AND MACHINE
LEARNING IN PHP
SYMFONYCON CLUJ 2017
@MICHAELCULLUMUK
Slide 3
Slide 3 text
@MICHAELCULLUMUK
ME?
Slide 4
Slide 4 text
MICHAEL CULLUM
@MICHAELCULLUMUK
Slide 5
Slide 5 text
@MICHAELCULLUMUK
STATISTICS
Slide 6
Slide 6 text
@MICHAELCULLUMUK
ARTIFICIAL
INTELLIGENCE
Slide 7
Slide 7 text
@MICHAELCULLUMUK
MACHINE LEARNING
Slide 8
Slide 8 text
@MICHAELCULLUMUK
MACHINE LEARNING
Slide 9
Slide 9 text
@MICHAELCULLUMUK
LEARNING
Cause
&
Effect
Context
Process Knowledge
Slide 10
Slide 10 text
@MICHAELCULLUMUK
USING
Cause
Knowledge
Process
Prediction
of an
effect
Slide 11
Slide 11 text
@MICHAELCULLUMUK
Cause
Knowledge
Process
Process
Cause
Predicted
effect
Effect
Slide 12
Slide 12 text
@MICHAELCULLUMUK
EXAMPLE
1+1= 2
1+2= 3
1+3= 4
2+1= 3
3+1= 4
Cause Effect
Slide 13
Slide 13 text
@MICHAELCULLUMUK
EXAMPLE
1+1= 2
1+2= 3
1+3= 4
2+1= 3
3+1= 4
Knowledge
Slide 14
Slide 14 text
@MICHAELCULLUMUK
EXAMPLE
1+1= 2
1+2= 3
1+3= 4
2+1= 3
3+1= 4
Cause
3+1=
Predicted Effect
4
Knowledge
Slide 15
Slide 15 text
@MICHAELCULLUMUK
MACHINE LEARNING AS
A 4-STEP PROCESS
Slide 16
Slide 16 text
@MICHAELCULLUMUK
1. ACQUIRE DATA
Slide 17
Slide 17 text
@MICHAELCULLUMUK
2. TRAIN MODEL
Slide 18
Slide 18 text
@MICHAELCULLUMUK
3. ASK YOUR QUESTION
Slide 19
Slide 19 text
@MICHAELCULLUMUK
4. GET PREDICTED
ANSWER
Slide 20
Slide 20 text
@MICHAELCULLUMUK
1. ACQUIRE DATA
Slide 21
Slide 21 text
@MICHAELCULLUMUK
GOOD LUCK
Slide 22
Slide 22 text
@MICHAELCULLUMUK
2. TRAIN MODEL
Slide 23
Slide 23 text
@MICHAELCULLUMUK
SUPERVISED LEARNING
UNSUPERVISED LEARNING
Slide 24
Slide 24 text
@MICHAELCULLUMUK
SUPERVISED LEARNING
Slide 25
Slide 25 text
@MICHAELCULLUMUK
KNOWN OUTCOMES
Slide 26
Slide 26 text
@MICHAELCULLUMUK
QUALITATIVE - REGRESSION
QUANTATIVE - CLASSIFICATION
Slide 27
Slide 27 text
@MICHAELCULLUMUK
QUALITATIVE - REGRESSION
QUANTATIVE - CLASSIFICATION
Slide 28
Slide 28 text
@MICHAELCULLUMUK
CLASSIFICATION
Rating Conclusion
100 Good
25 Bad
50 Good
40 Bad
Slide 29
Slide 29 text
@MICHAELCULLUMUK
CLASSIFICATION
Rating Conclusion
100 Good
25 Bad
50 Good
40 Bad
Slide 30
Slide 30 text
@MICHAELCULLUMUK
LINEAR CLASSIFIER
Item Value
Black
PHP
Devs
White Sales
Sales Test
PHP Test
Slide 31
Slide 31 text
@MICHAELCULLUMUK
QUALITATIVE - REGRESSION
QUANTATIVE - CLASSIFICATION
Slide 32
Slide 32 text
@MICHAELCULLUMUK
REGRESSION
Price % alcohol
5.00 8.25
6.00 10
4.00 6.75
3.00 5
0
2.5
5
7.5
10
0 1.5 3 4.5 6
Slide 33
Slide 33 text
@MICHAELCULLUMUK
REGRESSION
Price % alcohol
5.00 8.25
6.00 10
4.00 6.75
3.00 5
0
2.5
5
7.5
10
0 1.5 3 4.5 6
Slide 34
Slide 34 text
@MICHAELCULLUMUK
UNSUPERVISED
LEARNING
Slide 35
Slide 35 text
@MICHAELCULLUMUK
DISCOVERY
Slide 36
Slide 36 text
@MICHAELCULLUMUK
ASSOCIATION
CLUSTERING
Slide 37
Slide 37 text
@MICHAELCULLUMUK
ASSOCIATION
CLUSTERING
Slide 38
Slide 38 text
@MICHAELCULLUMUK
ASSOCIATION
Item 1 Item 2
T-shirt Shorts
Shorts T-shirt
Suit Black Shoes
Socks Underwear
Item 1 Item 2
Black shoes Suit
Socks Underwear
Underwear Socks
T-shirt Socks
Slide 39
Slide 39 text
@MICHAELCULLUMUK
ASSOCIATION
Item 1 Item 2
T-shirt Shorts
Shorts T-shirt
Suit Black Shoes
Socks Underwear
Item 1 Item 2
Black shoes Suit
Socks Underwear
Underwear Socks
T-shirt Socks
People who buy socks,
also often buy underwear
People who who buy
underwear always buy
socks
Slide 40
Slide 40 text
@MICHAELCULLUMUK
ASSOCIATION
Item 1 Item 2
T-shirt Shorts
Shorts T-shirt
Suit Black Shoes
Socks Underwear
Item 1 Item 2
Black shoes Suit
Socks Underwear
Underwear Socks
T-shirt Socks
People who buy socks,
also often buy underwear
People who buy suits
always buy black shoes
People who who buy
underwear always buy
socks
People who buy black
shoes, always buy suits
Slide 41
Slide 41 text
@MICHAELCULLUMUK
ASSOCIATION
Item 1 Item 2
T-shirt Shorts
Shorts T-shirt
Suit Black Shoes
Socks Underwear
Item 1 Item 2
Black shoes Suit
Socks Underwear
Underwear Socks
T-shirt Socks
People who buy socks,
also often buy underwear
People who buy suits
always buy black shoes
People who who buy
underwear always buy
socks
People who buy black
shoes, always buy suits
People who buy shorts
always buy t-shirts
People who buy t-shirts,
also often buy shorts
Slide 42
Slide 42 text
@MICHAELCULLUMUK
ASSOCIATION
CLUSTERING
Slide 43
Slide 43 text
@MICHAELCULLUMUK
CLUSTER ANALYSIS IS THE TASK OF
GROUPING A SET OF OBJECTS IN SUCH A
WAY THAT OBJECTS IN THE SAME GROUP
ARE MORE SIMILAR TO EACH OTHER
THAN TO THOSE IN OTHER GROUPS.
Slide 44
Slide 44 text
@MICHAELCULLUMUK
CLUSTER ANALYSIS
Slide 45
Slide 45 text
@MICHAELCULLUMUK
3. QUESTION
4. PREDICTED ANSWER
Slide 46
Slide 46 text
@MICHAELCULLUMUK
ALGORITHMS
Slide 47
Slide 47 text
@MICHAELCULLUMUK
LEAST SQUARES
Slide 48
Slide 48 text
@MICHAELCULLUMUK
LEAST SQUARES REGRESSION LINE
Slide 49
Slide 49 text
@MICHAELCULLUMUK
MATHS
Slide 50
Slide 50 text
@MICHAELCULLUMUK
A+BX=Y
Slide 51
Slide 51 text
@MICHAELCULLUMUK
▸ Each point has an x and y value
▸ We need an equation of a line
▸ We move the line an infinite number of times
▸ Each time, we draw a box between every point, and the line, with one corner
on the line, and another on the point
▸ The correct line is the one where the sum of the area of all the squares is
smallest
PROCESS
Slide 52
Slide 52 text
@MICHAELCULLUMUK
DEMO
Slide 53
Slide 53 text
@MICHAELCULLUMUK
PHP
Slide 54
Slide 54 text
@MICHAELCULLUMUK
DEMO
Slide 55
Slide 55 text
@MICHAELCULLUMUK
DEMOS
php-ai/php-ml
Slide 56
Slide 56 text
@MICHAELCULLUMUK
CODE
$samples = [[60], [61], [62], [63], [65]];
$targets = [3.1, 3.6, 3.8, 4, 4.1];
$regression = new LeastSquares();
$regression->train($samples, $targets);
echo $regression->predict([64]);
Slide 57
Slide 57 text
@MICHAELCULLUMUK
NEAREST NEIGHBOUR
Slide 58
Slide 58 text
@MICHAELCULLUMUK
NEAREST NEIGHBOUR
Slide 59
Slide 59 text
@MICHAELCULLUMUK
3-NEAREST NEIGHBOUR
Slide 60
Slide 60 text
@MICHAELCULLUMUK
5-NEAREST NEIGHBOUR
Slide 61
Slide 61 text
@MICHAELCULLUMUK
MATHS
Slide 62
Slide 62 text
@MICHAELCULLUMUK
PHP
Slide 63
Slide 63 text
@MICHAELCULLUMUK
DEMO
Slide 64
Slide 64 text
@MICHAELCULLUMUK
DEMO
0
1
2
3
4
0 1 2 3 4
Slide 65
Slide 65 text
@MICHAELCULLUMUK
DEMO
0
1.25
2.5
3.75
5
0 1 2 3 4
Slide 66
Slide 66 text
@MICHAELCULLUMUK
DEMO
$samples = [[1, 3], [1, 4], [2, 4], [3, 1], [4, 1], [4, 2]];
$labels = ['a', 'a', 'a', 'b', 'b', 'b'];
$classifier = new KNearestNeighbors();
$classifier->train($samples, $labels);
echo $classifier->predict([3, 2]);
Slide 67
Slide 67 text
@MICHAELCULLUMUK
USES OF MACHINE
LEARNING
Slide 68
Slide 68 text
@MICHAELCULLUMUK
NUMERICAL ANALYSIS
Slide 69
Slide 69 text
@MICHAELCULLUMUK
EXCEPTIONS
Slide 70
Slide 70 text
@MICHAELCULLUMUK
E-COMMERCE
Slide 71
Slide 71 text
@MICHAELCULLUMUK
FAULT DETECTION
Slide 72
Slide 72 text
@MICHAELCULLUMUK
ROOT CAUSE ANALYSIS
Slide 73
Slide 73 text
@MICHAELCULLUMUK
CLASSIFICATION
Slide 74
Slide 74 text
@MICHAELCULLUMUK
NATURAL LANGUAGE
PROCESSING
Slide 75
Slide 75 text
@MICHAELCULLUMUK
ANALYSIS OF SUPPORT
QUERIES
Slide 76
Slide 76 text
@MICHAELCULLUMUK
ANALYSIS OF LARGE
NUMBERS OF DOCUMENTS
Slide 77
Slide 77 text
@MICHAELCULLUMUK
FUN
Slide 78
Slide 78 text
@MICHAELCULLUMUK
ANY QUESTIONS?
Slide 79
Slide 79 text
THANKS
@MICHAELCULLUMUK
Slide 80
Slide 80 text
TREND ANALYSIS AND MACHINE
LEARNING IN PHP
SYMFONYCON CLUJ 2017
@MICHAELCULLUMUK