![Page 1: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/1.jpg)
Advanced Computation Models for Rule Based Networks Alexander Gegov University of Portsmouth, UK [email protected]
![Page 2: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/2.jpg)
2
Presentation Outline Introduction Types of Rule Based Systems Formal Models for Rule Based Networks Basic Operations in Rule Based Networks Structural Properties of Basic Operations Advanced Operations in Rule Based Networks Feedforward Rule Based Networks Feedback Rule Based Networks Evaluation of Rule Based Networks Rule Based Network Toolbox Conclusion References
![Page 3: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/3.jpg)
3
Introduction
Modelling Aspects of Systemic Complexity • non-linearity (input-output functional relationships) • uncertainty (incomplete and imprecise data) • dimensionality (large number of inputs and outputs) • structure (interacting subsystems)
Complexity Management by Rule Based Systems
• model feasibility (achievable in the case of non-linearity) • model accuracy (achievable in the case of uncertainty) • model efficiency (problematic in the case of dimensionality) • model transparency (problematic in the case of structure)
![Page 4: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/4.jpg)
4
Types of Rule Based Systems Logical Connections
• disjunctive antecedents and conjunctive rules • conjunctive antecedents and conjunctive rules • disjunctive antecedents and disjunctive rules • conjunctive antecedents and disjunctive rules
Inputs and Outputs
• single-input-single-output • single-input-multiple-output • multiple-input-single-output • multiple-input-multiple-output
![Page 5: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/5.jpg)
5
Types of Rule Based Systems
Rule Base Properties • completeness • consistency • exhaustiveness • monotonicity
Rule Base Type
• single rule base (standard rule based system) • chained rule bases (hierarchical rule based system) • modular rule bases (rule based network)
![Page 6: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/6.jpg)
6
Types of Rule Based Systems
Figure 1: Standard rule based system (single rule base)
![Page 7: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/7.jpg)
7
Types of Rule Based Systems Figure 2: Hierarchical rule based system (chained rule bases)
![Page 8: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/8.jpg)
8
Types of Rule Based Systems Figure 3: Rule based network (modular rule bases)
![Page 9: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/9.jpg)
9
Formal Models for Rule Based Networks Node Modelling by If-then Rules Rule 1: If x is low, then y is small Rule 2: If x is average, then y is medium Rule 3: If x is high, then y is big Node Modelling by Integer Tables Linguistic terms for x Linguistic terms for y 1 (low) 1 (small) 2 (average) 2 (medium) 3 (high) 3 (big)
![Page 10: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/10.jpg)
10
Formal Models for Rule Based Networks Node Modelling by Boolean Matrices x / y 1 (small) 2 (medium) 3 (big) 1 (low) 1 0 0 2 (average) 0 1 0 3 (high) 0 0 1 Node Modelling by Binary Relations {(1, 1), (2, 2), (3, 3)}
![Page 11: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/11.jpg)
11
Formal Models for Rule Based Networks Network Modelling by Grid Structures Layer 1 Level 1 N11(x, y)
Network Modelling by Interconnection Structures Layer 1 Level 1 y
![Page 12: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/12.jpg)
12
Formal Models for Rule Based Networks Network Modelling by Block Schemes
x N11 y
Network Modelling by Topological Expressions [N11] (x | y)
![Page 13: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/13.jpg)
13
Basic Operations in Rule Based Networks Horizontal Merging of Nodes
[N11] (x11 | z11,12) * [N 12] (z11,12
| y12) = [N11*12] (x11 | y12)
* symbol for horizontal merging
N11 : z11,12 1 2 3 x11 1 1 0 0 2 0 0 1 3 0 1 0
![Page 14: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/14.jpg)
14
Basic Operations in Rule Based Networks N12 : y12 1 2 3 z11,12 1 0 1 0 2 0 0 1 3 1 0 0 N11*12 : y12 1 2 3 x11 1 0 1 0 2 1 0 0 3 0 0 1
![Page 15: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/15.jpg)
15
Basic Operations in Rule Based Networks
Horizontal Splitting of Nodes
[N11/12] (x11 | y12) = [N11] (x11
| z11,12) / [N12] (z11,12 | y12)
/ symbol for horizontal splitting
N11/12 : y12 1 2 3 x11 1 1 0 0 2 0 0 1 3 0 1 0
![Page 16: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/16.jpg)
16
Basic Operations in Rule Based Networks
N11 : z11,12 1 2 3
x11 1 0 1 0 2 1 0 0 3 0 0 1 N12 : y12 1 2 3 z11,12 1 0 0 1 2 1 0 0 3 0 1 0
![Page 17: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/17.jpg)
17
Basic Operations in Rule Based Networks
Vertical Merging of Nodes
[N11] (x11 | y11) + [N21] (x21
| y21) = [N11+21] (x11, x21 | y11, y21)
+ symbol for vertical merging
![Page 18: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/18.jpg)
18
Basic Operations in Rule Based Networks
N11 : y11 1 2 3 x11 1 1 0 0 2 0 0 1 3 0 1 0 N21 : y21 1 2 3 x21 1 0 1 0 2 1 0 0 3 0 0 1
![Page 19: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/19.jpg)
19
Basic Operations in Rule Based Networks N11+21 : y11, y21 11 12 13 21 22 23 31 32 33 x11, x21 11 0 1 0 0 0 0 0 0 0 12 1 0 0 0 0 0 0 0 0 13 0 0 1 0 0 0 0 0 0 21 0 0 0 0 0 0 0 1 0 22 0 0 0 0 0 0 1 0 0 23 0 0 0 0 0 0 0 0 1 31 0 0 0 0 1 0 0 0 0 32 0 0 0 1 0 0 0 0 0 33 0 0 0 0 0 1 0 0 0
![Page 20: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/20.jpg)
20
Basic Operations in Rule Based Networks
Vertical Splitting of Nodes
[N11–21] (x11, x21 | y11, y21) = [N11] (x11
| y11) – [N21] (x21 | y21)
– symbol for vertical splitting
![Page 21: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/21.jpg)
21
Basic Operations in Rule Based Networks
N11+21 : y11, y21 11 12 13 21 22 23 31 32 33 x11, x21 11 0 0 0 0 0 1 0 0 0 12 0 0 0 1 0 0 0 0 0 13 0 0 0 0 1 0 0 0 0 21 0 0 0 0 0 0 0 0 1 22 0 0 0 0 0 0 1 0 0 23 0 0 0 0 0 0 0 1 0 31 0 0 1 0 0 0 0 0 0 32 1 0 0 0 0 0 0 0 0 33 0 1 0 0 0 0 0 0 0
![Page 22: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/22.jpg)
22
Basic Operations in Rule Based Networks
N11 : y11 1 2 3 x11 1 0 1 0 2 0 0 1 3 1 0 0 N21 : y21 1 2 3 x21 1 0 0 1 2 1 0 0 3 0 1 0
![Page 23: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/23.jpg)
23
Basic Operations in Rule Based Networks
Output Merging of Nodes
[N11] (x11,21 | y11) ; [N21] (x11,21
| y21) = [N11;21] (x11,21 | y11, y21)
; symbol for output merging
N11 : y11 1 2 3 x11,21 1 1 0 0 2 0 0 1 3 0 1 0
![Page 24: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/24.jpg)
24
Basic Operations in Rule Based Networks N21 : y21 1 2 3 x11,21 1 0 1 0 2 1 0 0 3 0 0 1
N11;21 : y11, y21 11 12 13 21 22 23 31 32 33 x11,21 1 0 1 0 0 0 0 0 0 0 2 0 0 0 0 0 0 1 0 0 3 0 0 0 0 0 1 0 0 0
![Page 25: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/25.jpg)
25
Basic Operations in Rule Based Networks
Output Splitting of Nodes
[N11:21] (x11, 21 | y11, y21) = [N11] (x11,21
| y11) : [N21] (x11,21 | y21)
: symbol for output splitting N11:21 : y11, y21 11 12 13 21 22 23 31 32 33 x11,21 1 0 0 0 0 0 1 0 0 0 2 0 0 0 0 0 0 1 0 0 3 0 1 0 0 0 0 0 0 0
![Page 26: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/26.jpg)
26
Basic Operations in Rule Based Networks N11 : y11 1 2 3 x11,21 1 0 1 0 2 0 0 1 3 1 0 0 N21 : y21 1 2 3 x11,21 1 0 0 1 2 1 0 0 3 0 1 0
![Page 27: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/27.jpg)
27
Structural Properties of Basic Operations Associativity of Horizontal Merging [N11] (x11
| z11,12) * [N 12] (z11,12 | z12,13) * [N 13] (z12,13
| y13) = [N11*12] (x11
| z12,13) * [N 13] (z12,13 | y13) =
[N11] (x11
| z11,12) * [N 12*13] (z11,12 | y13) =
[N11*12*13] (x11
| y13)
![Page 28: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/28.jpg)
28
Structural Properties of Basic Operations N11 : z11,12 1 2 3 x11 1 1 0 0 2 0 0 1 3 0 1 0 N12 : z12,13 1 2 3 z11,12 1 0 1 0 2 0 0 1 3 1 0 0
![Page 29: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/29.jpg)
29
Structural Properties of Basic Operations N13 : y13 1 2 3 z12,13 1 0 0 1 2 0 1 0 3 1 0 0 N(11*12)*13 = N11*(12*13) : y13 1 2 3 x11 1 0 1 0 2 0 0 1 3 1 0 0
![Page 30: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/30.jpg)
30
Structural Properties of Basic Operations Variability of Horizontal Splitting [N11/12/13] (x11
| y13) = [N11] (x11
| z11,12) / [N12/13] (z11,12 | y13) =
[N11/12] (x11
| z12,13) / [N13] (z12,13 | y13) =
[N11] (x11
| z11,12) / [N12] (z11,12 | z12,13) / [N13] (z12,13
| y13)
![Page 31: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/31.jpg)
31
Structural Properties of Basic Operations N11/(12/13) = N(11/21)/31 : y13 1 2 3 x11 1 0 0 1 2 1 0 0 3 0 1 0 N11 : z11,12
1 2 3 x11 1 0 1 0 2 1 0 0 3 0 0 1
![Page 32: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/32.jpg)
32
Structural Properties of Basic Operations N12 : z12,13 1 2 3 z11,12 1 0 0 1 2 1 0 0 3 0 1 0 N13 : y13 1 2 3 z12,13 1 0 0 1 2 0 1 0 3 1 0 0
![Page 33: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/33.jpg)
33
Structural Properties of Basic Operations Associativity of Vertical Merging [N11] (x11
| y11) + [N21] (x21 | y21) + [N31] (x31
| y31) = [N11+21] (x11,
x21 | y11, y21) + [N31] (x31
| y31) = [N11] (x11
| y11) + [N21+31] (x21, x31 | y21,
y31) = [N11+21+31] (x11,
x21, x31 | y11,
y21, y31)
![Page 34: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/34.jpg)
34
Structural Properties of Basic Operations N11 : y11 1 2 3 x11 1 1 0 0 2 0 0 1 3 0 1 0 N21 : y21 1 2 3 x21 1 0 1 0 2 1 0 0 3 0 0 1
![Page 35: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/35.jpg)
35
Structural Properties of Basic Operations N31 : y13 1 2 3 x31 1 0 0 1 2 0 1 0 3 1 0 0
![Page 36: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/36.jpg)
36
Structural Properties of Basic Operations N(11+21)+31 = N11+(21+31) : y11, y21, y31 A B C D E F G H I x11, x21, x31 A (111-113) 03 13 03 03 03 03 03 03 03 B (121-123) 13 03 03 03 03 03 03 03 03 C (131-133) 03 03 13 03 03 03 03 03 03 D (211-213) 03 03 03 03 03 03 03 13 03 E (221-223) 03 03 03 03 03 03 13 03 03 F (231-133) 03 03 03 03 03 03 03 03 13 G (311-313) 03 03 03 03 13 03 03 03 03 H (321-323) 03 03 03 13 03 03 03 03 03 I (331-333) 03 03 03 03 03 13 03 03 03 03 = ‘3 by 3’ null matrix, 13 = ‘3 by 3’ anti-diagonal matrix
![Page 37: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/37.jpg)
37
Structural Properties of Basic Operations Variability of Vertical Splitting [N11–21–31] (x11,
x21, x31 | y11,
y21, y31) =
[N11] (x11
| y11) – [N21–31] (x21, x31 | y21,
y31) = [N11–21] (x11,
x21 | y11, y21) – [N31] (x31
| y31) = [N11] (x11
| y11) – [N21] (x21 | y21) – [N31] (x31
| y31)
![Page 38: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/38.jpg)
38
Structural Properties of Basic Operations N11–(21–31) = N (11–21)–31 : y11, y21, y31 A B C D E F G H I x11, x21, x31 A (111-113) 03 03 03 03 03 13 03 03 03 B (121-123) 03 03 03 13 03 03 03 03 03 C (131-133) 03 03 03 03 13 03 03 03 03 D (211-213) 03 03 03 03 03 03 03 03 13 E (221-223) 03 03 03 03 03 03 13 03 03 F (231-133) 03 03 03 03 03 03 03 13 03 G (311-313) 03 03 13 03 03 03 03 03 03 H (321-323) 13 03 03 03 03 03 03 03 03 I (331-333) 03 13 03 03 03 03 03 03 03 03 = ‘3 by 3’ null matrix, 13 = ‘3 by 3’ anti-diagonal matrix
![Page 39: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/39.jpg)
39
Structural Properties of Basic Operations
N11 : y11 1 2 3 x11 1 0 1 0 2 0 0 1 3 1 0 0 N21 : y21 1 2 3 x21 1 0 0 1 2 1 0 0 3 0 1 0
![Page 40: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/40.jpg)
40
Structural Properties of Basic Operations
N31 : y13 1 2 3 x31 1 0 0 1 2 0 1 0 3 1 0 0
![Page 41: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/41.jpg)
41
Structural Properties of Basic Operations
Associativity of Output Merging [N11] (x11,21,31
| y11) ; [N21] (x11,21,31 | y21) ; [N31] (x11,21,31
| y31) = [N11;21] (x11,21,31 | y11,
y21) ; [N31] (x11,21,31 | y31) =
[N11] (x11,21,31
| y11) ; [N21;31] (x11,21,31 | y21, y31) =
[N11;21;31] (x11,21,31 | y11,
y21, y31)
![Page 42: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/42.jpg)
42
Structural Properties of Basic Operations N11 : y11 1 2 3 x11,21,31 1 1 0 0 2 0 0 1 3 0 1 0 N21 : y21 1 2 3 x11,21,31 1 0 1 0 2 1 0 0 3 0 0 1
![Page 43: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/43.jpg)
43
Structural Properties of Basic Operations N31 : y13 1 2 3 x11,21,31 1 0 0 1 2 0 1 0 3 1 0 0 N(11;21);31 = N11;(21;31) : y11, y21, y31 A B C D E F G H I x11,21,31 1 03 13 03 03 03 03 03 03 03 2 03 03 03 03 03 03 12 03 03 3 03 03 03 03 03 11 03 03 03 03 = 0 0 0, 13 = 0 0 1, 12 = 0 1 0, 11 = 1 0 0
![Page 44: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/44.jpg)
44
Structural Properties of Basic Operations Variability of Output Splitting [N11:21:31] (x11,21,31 | y11,
y21, y31) =
[N11] (x11,21,31
| y11) : [N21:31] (x11,21,31 | y21, y31) =
[N11:21] (x11,21,31 | y11,
y21) : [N31] (x11,21,31 | y31) =
[N11] (x11,21,31
| y11) : [N21] (x11,21,31 | y21) : [N31] (x11,21,31
| y31)
![Page 45: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/45.jpg)
45
Structural Properties of Basic Operations N11:(21:31) = N(11: 21):31 : y11, y21, y31 A B C D E F G H I x11,21,31 1 03 03 03 03 13 03 03 03 03 2 03 03 03 03 03 03 12 03 03 3 03 11 03 03 03 03 03 03 03 03 = 0 0 0, 13 = 0 0 1, 12 = 0 1 0, 11 = 1 0 0 N11 : y11 1 2 3 x11,12,13 1 0 1 0 2 0 0 1 3 1 0 0
![Page 46: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/46.jpg)
46
Structural Properties of Basic Operations N21 : y21 1 2 3 x11,21,31 1 0 0 1 2 1 0 0 3 0 1 0 N31 : y31 1 2 3 x11,21,31 1 0 0 1 2 0 1 0 3 1 0 0
![Page 47: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/47.jpg)
47
Advanced Operations in Rule Based Networks Node Transformation in Input Augmentation
[N] (x | y) ⇒ [NAI] (x , xAI | y)
N : y 1 2 3 x 1 0 1 0 2 1 0 0 3 0 0 1
![Page 48: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/48.jpg)
48
Advanced Operations in Rule Based Networks NAI
: y 1 2 3 x, xAI 11 0 1 0 12 0 1 0 13 0 1 0 21 1 0 0 22 1 0 0 23 1 0 0 31 0 0 1 32 0 0 1 33 0 0 1
![Page 49: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/49.jpg)
49
Advanced Operations in Rule Based Networks Node Transformation in Output Permutation [N] (x | y1, y2) ⇒ [NPO] (x | y2, y1) N : y1, y2 11 12 13 21 22 23 31 32 33 x 1 0 0 0 0 0 1 0 0 0 2 0 0 0 0 0 0 1 0 0 3 0 1 0 0 0 0 0 0 0
![Page 50: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/50.jpg)
50
Advanced Operations in Rule Based Networks NPO
: y2, y1 11 12 13 21 22 23 31 32 33 x 1 0 0 0 0 0 0 0 1 0 2 0 0 1 0 0 0 0 0 0 3 0 0 0 1 0 0 0 0 0
![Page 51: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/51.jpg)
51
Advanced Operations in Rule Based Networks Node Transformation in Feedback Equivalence [N] (x, z | y, z) ⇒ [NEF] (x, xEF | y, yEF) N : y, z 11 12 21 22 x, z 11 ? ? ? ? 12 ? ? ? ? 21 ? ? ? ? 22 ? ? ? ?
![Page 52: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/52.jpg)
52
Advanced Operations in Rule Based Networks NEF
: y, yEF 11 12 21 22
x, xEF
11 1 0 1 0 12 0 1 0 1 21 1 0 1 0 22 0 1 0 1
![Page 53: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/53.jpg)
53
Advanced Operations in Rule Based Networks Node Identification in Horizontal Merging A * U = C A, C – known nodes, U – non-unique unknown node A : zA,U 1 2 3 x1A, x2A 11 1 0 0 12 1 0 0 21 0 0 1 22 0 0 1
![Page 54: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/54.jpg)
54
Advanced Operations in Rule Based Networks C : y1U, y2U 11 12 21 22 x1A, x2A 11 0 0 0 1 12 0 0 0 1 21 1 0 0 0 22 1 0 0 0 U: y1U, y2U 11 12 21 22 zA,U 1 0 0 0 1 2 0 0 0 0 3 1 0 0 0
![Page 55: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/55.jpg)
55
Advanced Operations in Rule Based Networks U * B = C B, C – known nodes, U – non-unique unknown node B : y1B, y2B 11 12 21 22 zU,B 1 1 0 0 0 2 0 0 0 0 3 0 0 0 1
![Page 56: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/56.jpg)
56
Advanced Operations in Rule Based Networks C : y1B, y2B 11 12 21 22 x1U, x2U 11 0 0 0 1 12 0 0 0 1 21 1 0 0 0 22 1 0 0 0 U : zU,B 1 2 3 x1U, x2U 11 0 0 1 12 0 0 1 21 1 0 0 22 1 0 0
![Page 57: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/57.jpg)
57
Advanced Operations in Rule Based Networks A * U * B = C A, B, C – known nodes, U – non-unique unknown node A * U = D D * B = C D – non-unique unknown node U * B = E A * E = C E – non-unique unknown node
![Page 58: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/58.jpg)
58
Advanced Operations in Rule Based Networks Node Identification in Vertical Merging A + U = C A, C – known nodes, U – unique unknown node A : yA 1 2 3 xA 1 0 1 0 2 1 0 0 3 0 0 1
![Page 59: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/59.jpg)
59
Advanced Operations in Rule Based Networks C : yA, yU 11 12 13 21 22 23 31 32 33 xA, xU 11 0 0 0 1 0 0 0 0 0 12 0 0 0 0 0 1 0 0 0 13 0 0 0 0 1 0 0 0 0 21 1 0 0 0 0 0 0 0 0 22 0 0 1 0 0 0 0 0 0 23 0 1 0 0 0 0 0 0 0 31 0 0 0 0 0 0 1 0 0 32 0 0 0 0 0 0 0 0 1 33 0 0 0 0 0 0 0 1 0
![Page 60: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/60.jpg)
60
Advanced Operations in Rule Based Networks U : yU 1 2 3 xU 1 1 0 0 2 0 0 1 3 0 1 0
![Page 61: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/61.jpg)
61
Advanced Operations in Rule Based Networks U + B = C B, C – known nodes, U – unique unknown node B : yB 1 2 3 xB 1 0 1 0 2 0 0 1 3 1 0 0
![Page 62: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/62.jpg)
62
Advanced Operations in Rule Based Networks C : yU, yB 11 12 13 21 22 23 31 32 33 xU, xB 11 0 0 0 0 0 0 0 1 0 12 0 0 0 0 0 0 0 0 1 13 0 0 0 0 0 0 1 0 0 21 0 1 0 0 0 0 0 0 0 22 0 0 1 0 0 0 0 0 0 23 1 0 0 0 0 0 0 0 0 31 0 0 0 0 1 0 0 0 0 32 0 0 0 0 0 1 0 0 0 33 0 0 0 1 0 0 0 0 0
![Page 63: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/63.jpg)
63
Advanced Operations in Rule Based Networks U : yU 1 2 3 xU 1 0 0 1 2 1 0 0 3 0 1 0
![Page 64: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/64.jpg)
64
Advanced Operations in Rule Based Networks A + U + B = C A, B, C – known nodes, U – unique unknown node A + U = D D + B = C D – unique unknown node
U + B = E A + E = C E – unique unknown node
![Page 65: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/65.jpg)
65
Advanced Operations in Rule Based Networks Node Identification in Output Merging A ; U = C A, C – known nodes, U – unique unknown node A : yA 1 2 3 xA,U 1 0 1 0 2 1 0 0 3 0 0 1
![Page 66: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/66.jpg)
66
Advanced Operations in Rule Based Networks C : yA, yU 11 12 13 21 22 23 31 32 33 xA,U 1 0 0 0 1 0 0 0 0 0 2 0 0 1 0 0 0 0 0 0 3 0 0 0 0 0 0 0 1 0 U : yU 1 2 3 xA,U 1 1 0 0 2 0 0 1 3 0 1 0
![Page 67: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/67.jpg)
67
Advanced Operations in Rule Based Networks U ; B = C B, C – known nodes, U – unique unknown node B : yB 1 2 3 xU,B 1 0 1 0 2 0 0 1 3 1 0 0
![Page 68: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/68.jpg)
68
Advanced Operations in Rule Based Networks C : yU, yB 11 12 13 21 22 23 31 32 33 xU,B 1 0 0 0 0 0 0 0 1 0 2 0 0 1 0 0 0 0 0 0 3 0 0 0 1 0 0 0 0 0 U : yU 1 2 3 xU,B 1 0 0 1 2 1 0 0 3 0 1 0
![Page 69: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/69.jpg)
69
Advanced Operations in Rule Based Networks A ; U ; B = C A, B, C – known nodes, U – unique unknown node A ; U = D D ; B = C D – unique unknown node
U ; B = E A ; E = C E – unique unknown node
![Page 70: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/70.jpg)
70
Feedforward Rule Based Networks Network with Single Level and Single Layer
• rule based system 1,1 [N11] (x11 | y11)
![Page 71: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/71.jpg)
71
Feedforward Rule Based Networks Network with Single Level and Multiple Layers
• queue of ‘n’ rule based systems 1,1 … 1,n Example 1 (Network Analysis) [N11] (x11
| z11,121,2, z11,12
2,1) * [N 12] (z11,122,1, z11,12
1,2 | y12) ⇒ [N11
PO] (x11 | z11,12
2,1, z11,121,2) * [N 12] (z11,12
2,1, z11,121,2 | y12) ⇒
[N11
PO * N12] (x11 | y12) ⇒
![Page 72: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/72.jpg)
72
Feedforward Rule Based Networks NE = N11
PO * N12 Algorithm 1.1 (Network Design) 1. Define NE and N12. 2. Derive N11
PO, if possible. 3. Find N11 by inverse output permutation of N11
PO. Algorithm 1.2 (Network Design) 1. Define NE and N11. 2. Find N11
PO by output permutation of N11. 3. Derive N12, if possible.
![Page 73: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/73.jpg)
73
Feedforward Rule Based Networks Network with Multiple Levels and Single Layer
• stack of ‘m’ rule based systems 1,1 … m,1 Example 2 (Network Analysis) [N11] (x11,21
1,1, x112 | y11) ; [N21] (x11,21
1,1| y21) ⇒ [N11] (x11,21
1,1, x112 | y11) ; [N21
AI] (x11,211,1, x11
2 | y21) ⇒
![Page 74: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/74.jpg)
74
Feedforward Rule Based Networks [N11 ; N21
AI] (x11,211,1, x11
2 | y11, y21) ⇒ NE = N11 ; N21
AI Algorithm 2.1 (Network Design) 1. Define NE and N21. 2. Find N21
AI by input augmentation of N21. 3. Derive N11, if possible. Algorithm 2.2 (Network Design) 1. Define NE and N11. 2. Derive N21
AI, if possible. 3. Find N21 by inverse input augmentation of N21
AI.
![Page 75: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/75.jpg)
75
Feedforward Rule Based Networks Network with Multiple Levels and Multiple Layers
• grid of ‘m by n’ rule based systems 1,1 … 1,n … … … m,1 … m,n Example 3 (Network Analysis) {[N 11] (x11
| z11,121,1, z11,22
2,1) + [N21] (x21 | z21,12
1,2)} * {[N 12] (z11,12
1,1, z21,121,2 | y12) + [N22] (z11,22
2,1 | y22)} ⇒
![Page 76: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/76.jpg)
76
Feedforward Rule Based Networks [N11 + N21] (x11, x21
| z11,121,1, z11,22
2,1, z21,121,2) *
[N12 + N22] (z11,12
1,1, z21,121,2, z11,22
2,1 | y12, y22) ⇒ [(N11
+ N21)PO] (x11, x21
| z11,121,1, z21,12
1,2, z11,222,1) *
[N12 + N22] (z11,12
1,1, z21,121,2, z11,22
2,1 | y12, y22) ⇒ [(N11
+ N21)PO
* (N12 + N22)] (x11, x21 | y12, y22) ⇒
NE = (N11
+ N21)PO
* (N12 + N22)
![Page 77: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/77.jpg)
77
Feedforward Rule Based Networks Algorithm 3.1 (Network Design) 1. Define NE, N21, N12 and N22. 2. Find N12
+ N22 by vertical merging of N12 + N22.
3. Derive (N11 + N21)
PO, if possible. 4. Find N11
+ N21 by inverse output permutation of (N11 + N21)
PO. 5. Derive N11 from N11
+ N21, if possible. Algorithm 3.2 (Network Design) 1. Define NE, N11, N12 and N22. 2. Find N12
+ N22 by vertical merging of N12 + N22.
3. Derive (N11 + N21)
PO, if possible. 4. Find N11
+ N21 by inverse output permutation of (N11 + N21)
PO. 5. Derive N21 from N11
+ N21, if possible.
![Page 78: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/78.jpg)
78
Feedforward Rule Based Networks Algorithm 3.3 (Network Design) 1. Define NE, N11, N21 and N22. 2. Find N11
+ N21 by vertical merging of N11 + N21.
3. Find (N11 + N21)
PO by output permutation of N11 + N21.
4. Derive N12 + N22, if possible.
5. Derive N12 from N12 + N22, if possible.
Algorithm 3.4 (Network Design) 1. Define NE, N11, N21 and N12. 2. Find N11
+ N21 by vertical merging of N11 + N21.
3. Find (N11 + N21)
PO by output permutation of N11 + N21.
4. Derive N12 + N22, if possible.
5. Derive N22 from N12 + N22, if possible.
![Page 79: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/79.jpg)
79
Feedback Rule Based Networks Network with Single Local Feedback
• one node embraced by feedback N(F) N N N
N(F) – node embraced by feedback N – nodes with no feedback
N N(F) N N
N N N(F) N
N N N N(F)
![Page 80: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/80.jpg)
80
Feedback Rule Based Networks Example 4 (Network Analysis) [N11 ] (x11,
w11 | z11,12
2,1, v11) * [N 12] (z11,121,1 | y12),
[F11] (v11
| w11) ⇒ [N11 ] (x11,
w11 | z11,12
2,1, v11) * {[N 12] (z11,12
1,1 | y12) + [F11] (v11 | w11)} ⇒
[N11
* (N12 + F11)] (x11,
w11 | y12,
w11) ⇒ [(N11
* (N12 + F11))
EF] (x11, xEF | y12, y
EF) ⇒
![Page 81: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/81.jpg)
81
Feedback Rule Based Networks NE = (N11
* (N12 + F11))
EF F11 – feedback node Algorithm 4.1 (Network Design) 1. Define NE, N11 and N12. 2. Confirm that NE satisfies the feedback constraints, if possible. 3. Make NE
equal to N11 * (N12
+ F11). 4. Derive N12
+ F11 from NE, if possible. 5. Derive F11 from N12
+ F11, if possible.
![Page 82: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/82.jpg)
82
Feedback Rule Based Networks Network with Multiple Local Feedback
• at least two nodes embraced by separate feedback N(F1) N(F2) N N
N(F1), N(F2) – nodes embraced by separate feedback N – nodes with no feedback
N N N(F1) N(F2)
N(F1) N N N(F2)
N(F1) N N(F2) N
N N(F1) N N(F2)
N N(F1) N(F2) N
![Page 83: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/83.jpg)
83
Feedback Rule Based Networks Example 5 (Network Analysis) [N11 ] (x11,
w11 | z11,12
1,1, v11) * [N 12] (z11,121,1, w12
| y12, v12), [F11] (v11
| w11), [F12] (v12 | w12) ⇒
{[N 11] (x11,
w11 | z11,12
1,1, v11) + [I31] (w12 | w12)} *
{[N 12] (z11,12
1,1, w12 | y12, v12) + [F11] (v11
| w11)} * {[I 13] (y12
| y12) + [F12] (v12 | w12) + [I33] (w11| w11)} ⇒
![Page 84: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/84.jpg)
84
Feedback Rule Based Networks [N11 + I31] (x11,
w11, w12
| z11,121,1, v11,
w12) * [N12 + F11] (z11,12
1,1, w12, v11
| y12, v12, w11) *
[I 13 + F12 + I33] (y12, v12,
w11 | y12,
w12, w11) ⇒
[(N11
+ I31) PO] (x11,
w11, w12
| z11,121,1, w12,
v11) * [N12 + F11] (z11,12
1,1, w12, v11
| y12, v12, w11) *
[I 13 + F12 + I33] (y12, v12,
w11 | y12,
w12, w11) ⇒
![Page 85: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/85.jpg)
85
Feedback Rule Based Networks [(N11
+ I31) PO * (N12 + F11) * (I 13 + F12 + I33)]
(x11,
w11, w12
| y12, w12,
w11) ⇒ [((N11
+ I31)PO * (N12 + F11) * (I 13 + F12 + I33))
FE] (x11,
x1FE, x2
FE | y12, y2
FE, y1FE) ⇒
NE = ((N11
+ I31)PO * (N12 + F11) * (I 13 + F12 + I33))
FE F11, F12 – feedback nodes I31, I13, I33 – identity nodes
![Page 86: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/86.jpg)
86
Feedback Rule Based Networks Algorithm 5.1 (Network Design) 1. Define NE, N11, N12, I31, I13, I33 and F12. 2. Confirm that NE satisfies the feedback constraints, if possible. 3. Make NE
equal to (N11 + I31)
PO * (N12 + F11) * (I 13 + F12 + I33). 4. Find N11
+ I31 by vertical merging of N11 and I31.
5. Find (N11 + I31)
PO by output permutation of N11 + I31.
6. Find I13 + F12 + I33 by vertical merging of I13, F12 and I33. 7. Derive N12 + F11 from NE, if possible. 8. Derive F11 from N12 + F11, if possible.
![Page 87: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/87.jpg)
87
Feedback Rule Based Networks Algorithm 5.2 (Network Design) 1. Define NE, N11, N12, I31, I13, I33 and F11. 2. Confirm that NE satisfies the feedback constraints, if possible. 3. Make NE
equal to (N11 + I31)
PO * (N12 + F11) * (I 13 + F12 + I33). 4. Find N11
+ I31 by vertical merging of N11 and I31.
5. Find (N11 + I31)
PO by output permutation of N11 + I31.
6. Find N12 + F11 by vertical merging of N12 and F11. 7. Find (N11 + I31)
PO * (N12 + F11) by horizontal merging of (N11
+ I31)PO and (N12 + F11).
8. Derive I13 + F12 + I33 from NE, if possible. 9. Derive F12 from I13 + F12 + I33, if possible.
![Page 88: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/88.jpg)
88
Feedback Rule Based Networks
Network with Single Global Feedback • one set of at least two adjacent nodes embraced by feedback
N1(F) N2(F) N N N1(F) N N2(F) N {N1(F), N2(F)} – set of nodes embraced by feedback N – nodes with no feedback
N N N1(F) N2(F)
N N1(F) N N2(F)
![Page 89: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/89.jpg)
89
Feedback Rule Based Networks Example 6 (Network Analysis) [N11 ] (x11,
w12,11 | z11,12
1,1) * [N 12] (z11,121,1 | y12, v12,11),
[F12,11] (v12,11
| w12,11) ⇒ [N11] (x11,
w12,11 | z11,12
1,1) * [N 12] (z11,121,1 | y12, v12,11) *
{[I 13] (y12
| y12) + [F12,11] (y12, v12,11 | y12, w12,11)} ⇒
[N11
* N12 * (I13 + F12,11)] (x11, w12,11
| y12, w12,11)
[(N11
* N12 * (I13 + F12,11))EF] (x11,
xEF | y12, yEF) ⇒
![Page 90: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/90.jpg)
90
Feedback Rule Based Networks NE = (N11
* N12 * (I13 + F12,11))EF
F12,11 – feedback node I13 – identity node Algorithm 6.1 (Network Design) 1. Define NE, N11, N12, I13 and F12,11. 2. Confirm that NE satisfies the feedback constraints, if possible. 3. Make NE
equal to N11 * N12 * (I13 + F12,11).
4. Find N11 * N12 by horizontal merging of N11
and N12. 5. Derive I13 + F12,11 from NE, if possible. 6. Derive F12,11 from I13 + F12,11, if possible.
![Page 91: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/91.jpg)
91
Feedback Rule Based Networks
Network with Multiple Global Feedback • at least two sets of nodes embraced by separate feedback with
at least two adjacent nodes in each set
N1(F1) N2(F1) N1(F2) N2(F2) {N1(F1), N2(F1)}, {N1(F2), N2(F2)} – sets of nodes embraced by separate feedback
N1(F1) N1(F2) N2(F1) N2(F2)
![Page 92: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/92.jpg)
92
Feedback Rule Based Networks Example 7 (Network Analysis) [N11 ] (x11,
w12,11 | z11,12
1,1) * [N 12] (z11,121,1, w13,12
| z12,131,1, v12,11) *
[N13] (z12,13
1,1 | y13, v13,12), [F12,11] (v12,11
| w12,11), [F13,12] (v13,12 | w13,12) ⇒
{[N 11 ] (x11,
w12,11 | z11,12
1,1) + [I31] (w13,12 | w13,12)} *
[N12] (z11,12
1,1, w13,12 | z12,13
1,1, v12,11) * {[N 13] (z12,13
1,1 | y13, v13,12) + [I33] (v12,11 | v12,11)} *
![Page 93: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/93.jpg)
93
Feedback Rule Based Networks {[I 14] (y13
| y13) + [F13,12] (v13,12 | w13,12) +
[F12,11] (v12,11
| w12,11)} ⇒ [(N11
+ I31) * N12 * (N13 + I33) * (I 14 + F13,12 + F12,11)] (x11,
w11, w12
| y12, w12,
w11) ⇒ [((N11
+ I31) * N12 * (N13 + I33) * (I 14 + F13,12 + F12,11)) FE]
(x11,
x1FE, x2
FE | y12, y2
FE, y1FE) ⇒
![Page 94: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/94.jpg)
94
Feedback Rule Based Networks NE = ((N11
+ I31) * N12 * (N13 + I33) * (I 14 + F13,12 + F12,11)) FE
F12,11, F13,12 – feedback nodes I31, I33, I14 – identity nodes Algorithm 7.1 (Network Design) 1. Define NE, N11, N12, N13, I14, I31, I33 and F13,12. 2. Confirm that NE satisfies the feedback constraints, if possible. 3. Make NE
equal to (N11
+ I31) * N12 * (N13 + I33) * (I 14 + F13,12 + F12,11) .
4. Find N11 + I31 by vertical merging of N11
and I31. 5. Find (N11
+ I31) * N12 by horizontal merging of (N11
+ I31) and N12.
![Page 95: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/95.jpg)
95
Feedback Rule Based Networks 6. Find N13
+ I33 by vertical merging of N13 and I33.
7. Find (N11 + I31) * N12 * (N13 + I33) by horizontal merging of
(N11 + I31) * N12 and (N13 + I33) .
8. Derive I14 + F13,12 + F12,11 from NE, if possible. 9. Find I14 + F13,12 by vertical merging of I14 and F13,12. 10. Derive F12,11 from I14 + F13,12 + F12,11, if possible. Algorithm 7.2 (Network Design) 1. Define NE, N11, N12, N13, I14, I31, I33 and F11,12. 2. Confirm that NE satisfies the feedback constraints, if possible. 3. Make NE
equal to (N11
+ I31) * N12 * (N13 + I33) * (I 14 + F13,12 + F12,11). 4. Find N11
+ I31 by vertical merging of N11 and I31.
![Page 96: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/96.jpg)
96
Feedback Rule Based Networks 5. Find (N11
+ I31) * N12 by horizontal merging of (N11
+ I31) and N12. 6. Find N13
+ I33 by vertical merging of N13 and I33.
7. Find (N11 + I31) * N12 * (N13 + I33) by horizontal merging of
(N11 + I31) * N12 and (N13 + I33) .
8. Derive I14 + F13,12 + F12,11 from NE, if possible. 9. Derive F13,12 from I14 + F13,12 + F12,11, if possible.
![Page 97: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/97.jpg)
97
Evaluation of Rule Based Networks Structural Evaluation Metrics
• number of non-identity network nodes • number of non-identity network connections • overall number of cells in grid structure • number of populated cells in grid structure • average path length from first layer nodes to last layer nodes • average path depth from first level nodes to last level nodes
![Page 98: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/98.jpg)
98
Evaluation of Rule Based Networks Linguistic Evaluation Metrics
• composing hierarchical into standard rule based systems • decomposing standard into hierarchical rule based systems
Composition Formula NE,x = *p=1
x–1 (N1,p + +q=p+1x–1 Iq,p)
NE,x – equivalent node for a rule based network with x inputs
![Page 99: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/99.jpg)
99
Evaluation of Rule Based Networks Decomposition Algorithm 1. Find N1,1 from the first two inputs and the output. 2. If x=2, go to step 9. 3. Set k=3. 4. While k≤x, do steps 5-7. 5. Find NE,k from the first k inputs and the output. 6. Derive N1,k–1 from the formula for NE,k, if possible. 7. Set k=k+1. 8. Endwhile. 9. End. NE,k – equivalent node for a rule based network with first k inputs
![Page 100: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/100.jpg)
100
Evaluation of Rule Based Networks Functional Evaluation Metrics
• model performance indicators • applications to case studies
Feasibility Index (FI) FI = sum i=1
n (pi / n)
n – number of non-identity nodes p i – number of inputs to the i-th non-identity node lower FI ⇒ better feasibility
![Page 101: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/101.jpg)
101
Evaluation of Rule Based Networks
Accuracy Index (AI) AI = sum i=1
nl sum j=1qil sum k=1
vji (|yjik – dji
k| / vij)
nl – number of nodes in the last layer qil – number of outputs from the i-th node in the last layer vji – number of discrete values for the j-th output from the i-th node in the last layer yji
k , djik – simulated and measured k-th discrete value for the j-th
output from the i-th node in the last layer lower AI ⇒ better accuracy
![Page 102: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/102.jpg)
102
Evaluation of Rule Based Networks
Efficiency Index (EI) EI = sum i=1
n (qiIOM . ri
IOM)
n – number of non-identity nodes IOM – input output mapping qi
IOM – number of outputs from the i-th non-identity node with an
associated IOM ri
IOM – number of rules for the i-th non-identity node with an
associated IOM lower EI ⇒ better efficiency
![Page 103: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/103.jpg)
103
Evaluation of Rule Based Networks
Transparency Index (TI) TI = (p + q) / (n + m)
p – overall number of inputs q – overall number of outputs n – number of non-identity nodes m – number of non-identity connections lower TI ⇒ better transparency
![Page 104: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/104.jpg)
104
Evaluation of Rule Based Networks
Case Study 1 (Ore Flotation) Inputs: x1 – copper concentration in ore pulp x2 – iron concentration in ore pulp x3 – debit of ore pulp Output: y – new copper concentration in ore pulp
![Page 105: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/105.jpg)
105
Evaluation of Rule Based Networks
Figure 4: Standard rule based system for case study 1
![Page 106: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/106.jpg)
106
Evaluation of Rule Based Networks
Figure 5: Hierarchical rule based system for case study 1
![Page 107: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/107.jpg)
107
Evaluation of Rule Based Networks
Figure 6: Rule based network for case study 1
![Page 108: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/108.jpg)
108
Evaluation of Rule Based Networks Table 1: Models performance for case study 1 Index / Model Standard rule
based system Hierarchical rule based system
Rule based network
Feasibility 3 2 2 Accuracy 4.35 4.76 4.60 Efficiency 1331 242 1331 Transparency 4 1.33 1.33 FI – RBN superior to SRBS and equivalent to HRBS AI – RBN inferior to SRBS and superior to HRBS EI – RBN equivalent to SRBS and inferior to HRBS TI – RBN superior to SRBS and equivalent to HRBS
![Page 109: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/109.jpg)
109
Evaluation of Rule Based Networks
Case Study 2 (Retail Pricing) Inputs: x1 – expected selling price for retail product x2 – difference between selling price and cost for retail product x3 – expected percentage to be sold from retail product Output: y – maximum cost for retail product
![Page 110: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/110.jpg)
110
Evaluation of Rule Based Networks
0 20 40 60 80 100 120 1400
10
20
30
40
50
60
70
80
90
100
input permutations
outp
ut
Figure 7: Standard rule based system for case study 2
![Page 111: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/111.jpg)
111
Evaluation of Rule Based Networks
0 20 40 60 80 100 120 1400
20
40
60
80
100
input permutations
outp
ut
Figure 8: Hierarchical rule based system for case study 2
![Page 112: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/112.jpg)
112
Evaluation of Rule Based Networks
0 20 40 60 80 100 120 1400
20
40
60
80
100
input permutations
outp
ut
Figure 9: Rule based network for case study 2
![Page 113: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/113.jpg)
113
Evaluation of Rule Based Networks Table 2: Models performance for case study 2 Index / Model Standard rule
based system Hierarchical based based system
Rule based network
Feasibility 3 2 2 Accuracy 2.86 5.57 3.64 Efficiency 125 80 125 Transparency 4 1.33 1.33 FI – RBN superior to SRBS and equivalent to HRBS AI – RBN inferior to SRBS and superior to HRBS EI – RBN equivalent to SRBS and inferior to HRBS TI – RBN superior to SRBS and equivalent to HRBS
![Page 114: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/114.jpg)
114
Evaluation of Rule Based Networks Composition of Hierarchical into Standard Rule Based System
• full preservation of model feasibility • maximal improvement of model accuracy • fixed loss of model efficiency • full preservation of model transparency
Decomposition of Standard into Hierarchical Rule Based System
• no change of model feasibility • minimal loss of model accuracy • fixed improvement of model efficiency • no change of model transparency
![Page 115: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/115.jpg)
115
Rule Based Network Toolbox
Details • developed by Erasmus and project students • implemented in the Matlab environment • downloadable from the Mathworks web site • visible from the Springer web site
Structure • Matlab files for basic operations on nodes • Matlab files for advanced operations on nodes • Matlab files for auxiliary operations • Word files with illustration examples
![Page 116: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/116.jpg)
116
Conclusion Theoretical Significance of Rule Based Networks
• novel application of discrete mathematics and control theory • detailed validation by test examples and case studies
Methodological Impact of Rule Based Networks
• extension of standard and hierarchical rule based systems • bridge between standard and hierarchical rule based systems • novel framework for different types of rule based systems
Application Areas of Rule Based Networks
• modelling and simulation in the mining industry • modelling and simulation in the retail industry
![Page 117: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/117.jpg)
117
Conclusion Other Applications of Rule Based Networks
• finance (mortgage evaluation) • healthcare (patient management) • transport (commuter modelling) • robotics (path planning) • construction (crane control) • football (score prediction) • hospitality (rice cooking) • navigation (boat sailing) • environment (weather forecasting) • business (risk analysis)
![Page 118: IARIA 2013 TUTORIALTitle Microsoft Word - IARIA_2013_TUTORIAL.doc Author Alexander Gegov Created Date 6/4/2013 10:55:43 PM](https://reader034.vdocuments.site/reader034/viewer/2022051805/5ff48b4d04fe03606f0dfeb9/html5/thumbnails/118.jpg)
118
References A.Gegov, Fuzzy Networks for Complex Systems: A Modular Rule Base Approach, Springer, Berlin, 2010