Search by 
ifaqih / iffuzzy
Fuzzy Logic library for PHP based on the Mamdani, Sugeno or Tsukamoto methods.
1.0.0
2023-02-04 06:50 UTC
Requires
- php: >= 8.0
README
Fuzzy logic is a form of many-valued logic in which the truth value of variables may be any real number between 0 and 1. It is employed to handle the concept of partial truth, where the truth value may range between completely true and completely false. By contrast, in Boolean logic.
| PHP Version: 8.0 or above |
|---|
Installation
With Composer:
composer require ifaqih/iffuzzy
Use The Library
use IFaqih\AIMethods\Fuzzy;
Constants:
| Constant Name | Value | Description |
|---|---|---|
FUZZY_METHOD_MAMDANI |
"MAMDANI" |
Using the Mamdani method |
FUZZY_METHOD_SUGENO |
"SUGENO" |
Using the Sugeno method |
FUZZY_METHOD_TSUKAMOTO |
"TSUKAMOTO" |
Using the Tsukamoto method |
FUZZY_MEMBERSHIP_LINEAR_UP |
10 |
Using ascending linear membership |
FUZZY_MEMBERSHIP_LINEAR_DOWN |
11 |
Using descending linear membership |
FUZZY_MEMBERSHIP_TRIANGLE |
12 |
Using triangular membership |
FUZZY_MEMBERSHIP_TRAPEZOID |
13 |
Using trapezoidal membership |
FUZZY_MEMBERSHIP_SIGMOID_UP |
14 |
Using ascending sigmoid membership |
FUZZY_MEMBERSHIP_SIGMOID_DOWN |
15 |
Using descending sigmoid membership |
FUZZY_MEMBERSHIP_PI |
16 |
Using pi membership |
Set Method
Setting the fuzzy method to be used.
class::method()
- Type: static
- Parameter data type:
string $method_name - Return data type:
object
Prototype:
IFFuzzy::method(FUZZY_METHOD_MAMDANI)
Set Attribute
Set one attribute context.
class::attribute()
- Type: static
- Parameter data type:
string $context_name, array $attributes, ?array $domain = null - Return data type:
object
Prototype 1:
IFFuzzy::attribute("many", [ "little" => FUZZY_MEMBERSHIP_LINEAR_DOWN, "much" => FUZZY_MEMBERSHIP_LINEAR_UP ], [40, 80] )
Prototype 2:
IFFuzzy::attribute("level", [ "low" => [ "membership" => FUZZY_MEMBERSHIP_LINEAR_DOWN, "domain" => [40, 50] ], "medium" => [ "membership" => FUZZY_MEMBERSHIP_TRIANGLE, "domain" => [40, 60] ], "high" => [ "membership" => FUZZY_MEMBERSHIP_LINEAR_UP, "domain" => [50, 60] ] ] )
Set Attributes
Set multiple context attributes.
class::attributes()
- Type: static
- Parameter data type:
array ...$attribute_each_context - Return data type:
object
Prototype:
IFFuzzy::attributes( [ "many" => [ "little" => FUZZY_MEMBERSHIP_LINEAR_DOWN, "much" => FUZZY_MEMBERSHIP_LINEAR_UP ], [40, 80] ], [ "level" => [ "low" => [ "membership" => FUZZY_MEMBERSHIP_LINEAR_DOWN, "domain" => [40, 50] ], "medium" => [ "membership" => FUZZY_MEMBERSHIP_TRIANGLE, "domain" => [40, 60] ], "high" => [ "membership" => FUZZY_MEMBERSHIP_LINEAR_UP, "domain" => [50, 60] ] ] ], [ "speed" => [ "slow" => FUZZY_MEMBERSHIP_LINEAR_DOWN, "fast" => FUZZY_MEMBERSHIP_LINEAR_UP ], [500, 1200] ] )
Set Rule
Set one rule.
class::rule()
- Type: static
- Parameter data type:
string $result, array $attribute - Return data type:
object
Prototype:
IFFuzzy::rule("fast", ["many" => "much", "level" => "medium"])
Set Rules
Set multiple rule.
class::rules()
- Type: static
- Parameter data type:
array ...$rules - Return data type:
object
Prototype:
IFFuzzy::rules( ["rules" => ["many" => "little", "level" => "low"], "result" => "slow"], ["rules" => ["many" => "little", "level" => "medium"], "result" => "slow"], ["rules" => ["many" => "little", "level" => "high"], "result" => "fast"], ["rules" => ["many" => "much", "level" => "low"], "result" => "slow"], ["rules" => ["many" => "much", "level" => "medium"], "result" => "fast"], ["rules" => ["many" => "much", "level" => "high"], "result" => "fast"] )
Builder of Rules
In the set_rules() method you can use builder
class::build()
- Type: static
- Parameter data type: no needed
- Return data type:
object
Object of build method
Set Rule Name
class::build()->rule()
- Type: static
- Parameter data
string|array $data - Return data type:
object
Set Rule Name
class::build()->rule()->result()
- Type: static
- Parameter data
string $data, bool $is_formula = false - Return data type:
array
Prototype:
IFFuzzy::build()->rule(["many" => "much", "level" => "medium"])->result('fast')
Set Value
Set value of one context attribute.
class::set_value()
- Type: static
- Parameter data type:
string $context_name, int|float $value - Return data type:
object
Prototype:
IFFuzzy::set_value("much", 50)
Set Values
Set value of multiple context attribute.
class::set_values()
- Type: static
- Parameter data type:
string $context_name, int|float $value - Return data type:
object
Prototype:
IFFuzzy::set_values( [ "much" => 50, "level" => 58 ] )
Constant Flags
| Constant Name | Value | Description |
|---|---|---|
FUZZY_EXEC_HOLD |
1 |
Not reset all data after executed |
FUZZY_EXEC_GET_ALL |
2 |
Getting all data has been processed of library (in object values) |
FUZZY_EXEC_AS_ARRAY |
3 |
Getting all data has been processed of library in array values |
Executing Fuzzy Logic
Execute datas has been added from all of method.
class::execute()
- Type: static
- Parameter data type:
array|int $flags = 0 - Return data type:
int|float|object|array|null
Prototype:
IFFuzzy::execute(FUZZY_EXEC_HOLD)
Getting Value
Getting all data has been processed of library.
class::get_all_values()
- Type: static
- Parameter data type:
int $flags = 0 - Return data type:
array|object
Prototype:
IFFuzzy::get_all_values(FUZZY_EXEC_AS_ARRAY)
Clear Data
Clearing datas has been added or result of processed.
class::clear()
- Type: static
- Parameter data type: no needed
- Return data type:
void
Prototype:
IFFuzzy::clear()
Example
Example 1:
IFFuzzy::method(FUZZY_METHOD_MAMDANI) ->attributes( [ "many" => [ 'little' => FUZZY_MEMBERSHIP_LINEAR_DOWN, 'much' => FUZZY_MEMBERSHIP_LINEAR_UP ], [40, 80] ], [ "level" => [ 'low' => [ 'membership' => FUZZY_MEMBERSHIP_LINEAR_DOWN, 'domain' => [40, 50] ], 'medium' => [ 'membership' => FUZZY_MEMBERSHIP_TRIANGLE, 'domain' => [40, 60] ], 'high' => [ 'membership' => FUZZY_MEMBERSHIP_LINEAR_UP, 'domain' => [50, 60] ] ] ], [ "speed" => [ 'slow' => FUZZY_MEMBERSHIP_LINEAR_DOWN, 'fast' => FUZZY_MEMBERSHIP_LINEAR_UP ], [500, 1200] ] ) ->rules( ['rules' => ["many" => "little", "level" => "low"], 'result' => 'slow'], ['rules' => ["many" => "little", "level" => "medium"], 'result' => 'slow'], ['rules' => ["many" => "little", "level" => "high"], 'result' => 'fast'], ['rules' => ["many" => "much", "level" => "low"], 'result' => 'slow'], ['rules' => ["many" => "much", "level" => "medium"], 'result' => 'fast'], ['rules' => ["many" => "much", "level" => "high"], 'result' => 'fast'] ) ->set_values([ 'many' => 50, 'level' => 58 ]) ->execute();
Result:
float(782.6128545848649)
Example 2:
IFFuzzy::method(FUZZY_METHOD_MAMDANI); IFFuzzy::attribute("many", [ 'little' => FUZZY_MEMBERSHIP_LINEAR_DOWN, 'much' => FUZZY_MEMBERSHIP_LINEAR_UP ], [40, 80]); IFFuzzy::attribute("level", [ 'low' => [ 'membership' => FUZZY_MEMBERSHIP_LINEAR_DOWN, 'domain' => [40, 50] ], 'medium' => [ 'membership' => FUZZY_MEMBERSHIP_TRIANGLE, 'domain' => [40, 60] ], 'high' => [ 'membership' => FUZZY_MEMBERSHIP_LINEAR_UP, 'domain' => [50, 60] ] ]); IFFuzzy::attribute("speed", [ 'slow' => FUZZY_MEMBERSHIP_LINEAR_DOWN, 'fast' => FUZZY_MEMBERSHIP_LINEAR_UP ], [500, 1200]); IFFuzzy::rules([ IFFuzzy::build()->rule(["many" => "little", "level" => "low"])->result('slow'), IFFuzzy::build()->rule(["many" => "little", "level" => "medium"])->result('slow'), IFFuzzy::build()->rule(["many" => "little", "level" => "high"])->result('fast'), IFFuzzy::build()->rule(["many" => "much", "level" => "low"])->result('slow'), IFFuzzy::build()->rule(["many" => "much", "level" => "medium"])->result('fast'), IFFuzzy::build()->rule(["many" => "much", "level" => "high"])->result('fast') ]); IFFuzzy::set_values([ 'many' => 50, 'level' => 58 ]); IFFuzzy::execute(FUZZY_EXEC_AS_ARRAY);
Result:
array(6) { ["method"]=> string(7) "MAMDANI" ["attributes"]=> array(3) { ["many"]=> array(2) { ["little"]=> array(3) { ["membership"]=> string(11) "Linear Down" ["domain"]=> array(2) { [0]=> int(40) [1]=> int(80) } ["fuzzification"]=> float(0.75) } ["much"]=> array(3) { ["membership"]=> string(9) "Linear Up" ["domain"]=> array(2) { [0]=> int(40) [1]=> int(80) } ["fuzzification"]=> float(0.25) } } ["level"]=> array(3) { ["low"]=> array(3) { ["membership"]=> string(11) "Linear Down" ["domain"]=> array(2) { [0]=> int(40) [1]=> int(50) } ["fuzzification"]=> int(0) } ["medium"]=> array(3) { ["membership"]=> string(8) "Triangle" ["domain"]=> array(2) { [0]=> int(40) [1]=> int(60) } ["fuzzification"]=> float(0.2) } ["high"]=> array(3) { ["membership"]=> string(9) "Linear Up" ["domain"]=> array(2) { [0]=> int(50) [1]=> int(60) } ["fuzzification"]=> float(0.8) } } ["speed"]=> array(2) { ["slow"]=> array(2) { ["membership"]=> int(11) ["domain"]=> array(2) { [0]=> int(500) [1]=> int(1200) } } ["fast"]=> array(2) { ["membership"]=> int(10) ["domain"]=> array(2) { [0]=> int(500) [1]=> int(1200) } } } } ["rules"]=> array(6) { [0]=> array(2) { ["rules"]=> array(2) { ["many"]=> string(6) "little" ["level"]=> string(3) "low" } ["result"]=> string(6) "slow" } [1]=> array(2) { ["rules"]=> array(2) { ["many"]=> string(6) "little" ["level"]=> string(6) "medium" } ["result"]=> string(6) "slow" } [2]=> array(2) { ["rules"]=> array(2) { ["many"]=> string(6) "little" ["level"]=> string(4) "high" } ["result"]=> string(5) "fast" } [3]=> array(2) { ["rules"]=> array(2) { ["many"]=> string(4) "much" ["level"]=> string(3) "low" } ["result"]=> string(6) "slow" } [4]=> array(2) { ["rules"]=> array(2) { ["many"]=> string(4) "much" ["level"]=> string(6) "medium" } ["result"]=> string(5) "fast" } [5]=> array(2) { ["rules"]=> array(2) { ["many"]=> string(4) "much" ["level"]=> string(4) "high" } ["result"]=> string(5) "fast" } } ["values"]=> array(2) { ["many"]=> int(50) ["level"]=> int(58) } ["inference"]=> array(2) { ["alpha_predicate"]=> array(2) { ["slow"]=> array(3) { [0]=> int(0) [1]=> float(0.2) [3]=> int(0) } ["fast"]=> array(3) { [2]=> float(0.75) [4]=> float(0.2) [5]=> float(0.25) } } ["z"]=> array(2) { ["momentum"]=> array(3) { [0]=> float(40960) [1]=> float(159037.08333333337) [2]=> float(146015.625) } ["area"]=> array(3) { [0]=> float(128) [1]=> float(182.875) [2]=> float(131.25) } } } ["result"]=> float(782.6128545848649) }