Attributor/attribution_test.go at main · Crynge/Attributor · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
package main

import (
	"math"
	"testing"

	"github.com/Crynge/Attributor/internal/attribution"
)

func TestFirstTouch(t *testing.T) {
	model := &attribution.FirstTouchModel{}
	paths := [][]string{
		{"email", "search", "social"},
		{"search", "display"},
		{"social", "email", "search"},
	}
	conv := []bool{true, true, false}
	results := model.Attribute(paths, conv)
	credits := mapResults(results)
	if credits["email"] != 1 {
		t.Errorf("expected email credit 1, got %f", credits["email"])
	}
	if credits["search"] != 1 {
		t.Errorf("expected search credit 1, got %f", credits["search"])
	}
	if credits["social"] != 0 {
		t.Errorf("expected social credit 0, got %f", credits["social"])
	}
}

func TestLastTouch(t *testing.T) {
	model := &attribution.LastTouchModel{}
	paths := [][]string{
		{"email", "search", "social"},
		{"search", "display"},
		{"social", "email", "search"},
	}
	conv := []bool{true, true, false}
	results := model.Attribute(paths, conv)
	credits := mapResults(results)
	if credits["social"] != 1 {
		t.Errorf("expected social credit 1, got %f", credits["social"])
	}
	if credits["display"] != 1 {
		t.Errorf("expected display credit 1, got %f", credits["display"])
	}
}

func TestLinear(t *testing.T) {
	model := &attribution.LinearModel{}
	paths := [][]string{
		{"a", "b"},
		{"a", "b", "c"},
	}
	conv := []bool{true, true}
	results := model.Attribute(paths, conv)
	credits := mapResults(results)
	if math.Abs(credits["a"]-0.5-1.0/3.0) > 1e-6 {
		t.Errorf("unexpected a credit: %f", credits["a"])
	}
}

func TestTimeDecay(t *testing.T) {
	model := &attribution.TimeDecayModel{}
	paths := [][]string{
		{"email", "search"},
	}
	conv := []bool{true}
	results := model.Attribute(paths, conv)
	credits := mapResults(results)
	if credits["search"] <= credits["email"] {
		t.Errorf("expected last touch (search) to have higher credit than first (email): search=%f email=%f", credits["search"], credits["email"])
	}
}

func TestShapleyValue(t *testing.T) {
	model := &attribution.ShapleyValueModel{}
	paths := [][]string{
		{"a", "b"},
		{"a"},
		{"b"},
	}
	conv := []bool{true, true, false}
	results := model.Attribute(paths, conv)
	credits := mapResults(results)
	total := credits["a"] + credits["b"]
	if math.Abs(total-1) > 1e-6 {
		t.Errorf("expected total credit 1, got %f", total)
	}
}

func TestShapleySymmetric(t *testing.T) {
	model := &attribution.ShapleyValueModel{}
	paths := [][]string{
		{"a", "b"},
		{"b", "a"},
	}
	conv := []bool{true, true}
	results := model.Attribute(paths, conv)
	credits := mapResults(results)
	if math.Abs(credits["a"]-credits["b"]) > 1e-6 {
		t.Errorf("expected symmetric credits: a=%f b=%f", credits["a"], credits["b"])
	}
}

func TestMarkovChain(t *testing.T) {
	model := &attribution.MarkovChainModel{}
	paths := [][]string{
		{"a", "b"},
		{"a"},
		{"b"},
	}
	conv := []bool{true, true, false}
	results := model.Attribute(paths, conv)
	if len(results) == 0 {
		t.Fatal("expected non-empty results")
	}
	total := 0.0
	for _, r := range results {
		total += r.Share
	}
	if math.Abs(total-1) > 1e-6 {
		t.Errorf("expected shares sum to 1, got %f", total)
	}
}

func TestShapleyRemoveEffect(t *testing.T) {
	conversions := map[int]float64{
		0: 0,
		1: 5,
		2: 3,
		3: 10,
	}
	effects := attribution.RemoveEffect(conversions, 10, 2)
	if len(effects) != 2 {
		t.Fatalf("expected 2 effects, got %d", len(effects))
	}
}

func mapResults(results []attribution.AttributionResult) map[string]float64 {
	m := make(map[string]float64)
	for _, r := range results {
		m[r.Channel] = r.Credit
	}
	return m
}