[PPT] - Extended Context Patterns A Visual Language for Context-Aware PowerPoint Presentation

SLIDE 1

Extended Context Patterns – A Visual

Language for Context-Aware Applications

Andrei Olaru and Adina Magda Florea

cs@andreiolaru.ro

AI-MAS Group, University Politehnica of Bucharest 10.10.2016

0 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 2

Context Focus Graphs & Patterns Extended Patterns Conclusion Future Work

Extended Context Patterns – A Visual Language for Context-Aware Applications

verview

0 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 3

· Extended Context Patterns · ·

|Context

Focus Graphs & Patterns Extended Patterns Conclusion Future Work

| Context

1 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 4

· Extended Context Patterns · ·

|Context

Focus Graphs & Patterns Extended Patterns Conclusion Future Work

| Context

◮ Ambient Intelligence /

Internet of Things with applications in AAL.

◮ AmI systems must make

complex processes usable by everyday people.

1 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 5

· Extended Context Patterns · ·

|Context

Focus Graphs & Patterns Extended Patterns Conclusion Future Work

| Context

◮ Ambient Intelligence /

Internet of Things with applications in AAL.

◮ AmI systems must make

complex processes usable by everyday people.

· part of the AmIciTy initiative and the CAMI project.

aimas.cs.pub.ro/amicity camiproject.eu 1 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 6

· Extended Context Patterns · · Context

|Focus

Graphs & Patterns Extended Patterns Conclusion Future Work

| Focus

? How to represent

◮ situation ◮ patterns ◮ rules

2 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 7

· Extended Context Patterns · · Context

|Focus

Graphs & Patterns Extended Patterns Conclusion Future Work

| Focus

? How to represent

◮ situation ◮ patterns (detected or searched) ◮ rules

2 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 8

· Extended Context Patterns · · Context

|Focus

Graphs & Patterns Extended Patterns Conclusion Future Work

| Focus

? How to represent

◮ situation ◮ patterns (detected or searched) ◮ rules

. . . in a manner that is

◮ visual ◮ clear ◮ incremental to

learn

2 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 9

· Extended Context Patterns · · Context

|Focus

Graphs & Patterns Extended Patterns Conclusion Future Work

| Focus

? How to represent

◮ situation ◮ patterns (detected or searched) ◮ rules

. . . in a manner that is

◮ visual ◮ clear ◮ incremental to

learn · take inspiration from Conceptual Graphs, Semantic Maps and RDF.

[Sowa, 2006, Sowa, 2008, Lassila and Swick, 1998]

2 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 10

· Extended Context Patterns · · Context

|Focus

Graphs & Patterns Extended Patterns Conclusion Future Work

| Focus

We represent:

◮ situation – information about the current

state of the assisted user.

◮ patterns – information about situations that

are likely to occur.

◮ rules – information about what to do in

particular situations. · we use a representation that is based on graphs.

[Olaru et al., 2011]

3 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 11

· Extended Context Patterns · · Context

|Focus

Graphs & Patterns Extended Patterns Conclusion Future Work

| Focus

We represent:

◮ situation – information about the current

state of the assisted user.

◮ patterns – information about situations that

are likely to occur.

◮ rules – information about what to do in

particular situations. · we use a representation that is based on graphs.

[Olaru et al., 2011]

3 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 12

· Extended Context Patterns · · Context

|Focus

Graphs & Patterns Extended Patterns Conclusion Future Work

| Focus

We represent:

◮ situation – information about the current

state of the assisted user.

◮ patterns – information about situations that

are likely to occur.

◮ rules – information about what to do in

particular situations. · we use a representation that is based on graphs.

[Olaru et al., 2011]

3 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 13

· Extended Context Patterns · · Context

|Focus

Graphs & Patterns Extended Patterns Conclusion Future Work

| Focus

We represent:

◮ situation – information about the current

state of the assisted user.

◮ patterns – information about situations that

are likely to occur.

◮ rules – information about what to do in

particular situations. · we use a representation that is based on graphs.

[Olaru et al., 2011]

3 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 14

· Extended Context Patterns · · Context

|Focus

Graphs & Patterns Extended Patterns Conclusion Future Work

| Focus

We represent:

◮ situation – information about the current

state of the assisted user.

◮ patterns – information about situations that

are likely to occur.

◮ rules – information about what to do in

particular situations. · we use a representation that is based on graphs.

[Olaru et al., 2011]

3 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 15

· Extended Context Patterns · · Context Focus

|Graphs & Patterns

Extended Patterns Conclusion Future Work

| Context Graphs and Patterns

◮ the basic element of the context graph / pattern

is the relation between concepts. · the labels of concepts are chosen from the user’s familiar space – his/her contacts, familiar places, etc.

◮ patterns contain generic nodes, that match any concept. ◮ to support describing conditions for and effects of pattern detection −

→ edges have properties such as characteristic and actionability.

4 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 16

· Extended Context Patterns · · Context Focus

|Graphs & Patterns

Extended Patterns Conclusion Future Work

| Context Graphs and Patterns

◮ the basic element of the context graph / pattern

is the relation between concepts. · the labels of concepts are chosen from the user’s familiar space – his/her contacts, familiar places, etc.

◮ patterns contain generic nodes, that match any concept. ◮ to support describing conditions for and effects of pattern detection −

→ edges have properties such as characteristic and actionability.

4 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 17

· Extended Context Patterns · · Context Focus

|Graphs & Patterns

Extended Patterns Conclusion Future Work

| Context Graphs and Patterns

◮ the basic element of the context graph / pattern

is the relation between concepts. · the labels of concepts are chosen from the user’s familiar space – his/her contacts, familiar places, etc.

◮ patterns contain generic nodes, that match any concept. ◮ to support describing conditions for and effects of pattern detection −

→ edges have properties such as characteristic and actionability.

4 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 18

· Extended Context Patterns · · Context Focus Graphs & Patterns

|Extended Patterns

Conclusion Future Work Hyperedges Negation Repetition Alternative

| Extended Patterns

◮ in order to increase the power of context patterns, we introduce further

structuring elements, in the shape of several types of hyperedges. ← − this is a hyperedge. it has an inbound arity of 1 and an

utbound arity of one, for a total

pattern-arity of 2.

5 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 19

· Extended Context Patterns · · Context Focus Graphs & Patterns

|Extended Patterns

Conclusion Future Work Hyperedges Negation Repetition Alternative

| Extended Patterns

“The place is not a building.”

6 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 20

· Extended Context Patterns · · Context Focus Graphs & Patterns

|Extended Patterns

Conclusion Future Work Hyperedges Negation Repetition Alternative

| Extended Patterns

“There may be various spaces contained in each other, of which the topmost is a building. In this case, the emergency contact is the administrator of the building.”

7 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 21

· Extended Context Patterns · · Context Focus Graphs & Patterns

|Extended Patterns

Conclusion Future Work Hyperedges Negation Repetition Alternative

| Extended Patterns

“If the place where the user is located is not a building, then contact the Police or Fire Department which covers the place.”

8 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 22

· Extended Context Patterns · · Context Focus Graphs & Patterns Extended Patterns

|Conclusion

Future Work

| Conclusion

◮ We have developed a formalism that relies on graph theory. ◮ We have introduced three types of hyperedges which can extend the

power of expression of graph patterns.

◮ Matching of the three types of hyperedges can be integrated into

the existing algorithm for context pattern matching.

[Olaru and Florea, 2015]

9 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 23

· Extended Context Patterns · · Context Focus Graphs & Patterns Extended Patterns Conclusion

|Future Work

| Future Work

◮ Test the expressiveness of the introduced formalism in more AmI

and AAL scenarios.

◮ Test the performance of the matching algorithm on real-life

scenarios.

◮ Create a visual interface in which the user can easily created

extended context patterns.

10 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 24

Thank You!

Any Questions? cs@andreiolaru.ro

10 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 25

· Extended Context Patterns · ·

Lassila, O. and Swick, R. (1998). Resource description framework (RDF) model and syntax specification. Technical report, The World Wide Web Consortium. Olaru, A. and Florea, A. M. (2015). A platform for matching context in real time. In Onieva, E., Santos, I., Osaba, E., Quinti´ an, H., and Corchado, E., editors, Proceedings of HAIS 2015, The 10th International Conference on Hybrid Artificial Intelligent Systems, June 22-24, Bilbao, Spain, volume 9121 of Lecture Notes in Computer Science, pages 99–110. Springer International Publishing. Olaru, A., Florea, A. M., and El Fallah Seghrouchni, A. (2011). Graphs and patterns for context-awareness. In Novais, P., Preuveneers, D., and Corchado, J., editors, Ambient Intelligence - Software and Applications, 2nd International Symposium on Ambient Intelligence (ISAmI 2011), University of Salamanca (Spain) 6-8th April, 2011, volume 92 of Advances in Intelligent and Soft Computing, pages 165–172. Springer Berlin / Heidelberg. Sowa, J. (2008). Conceptual graphs. Foundations of Artificial Intelligence, 3:213–237. Sowa, J. F. (2006). Semantic networks. Encyclopedia of Cognitive Science.

10 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016

SLIDE 26

Thank You!

Any Questions? cs@andreiolaru.ro

10 / 10

. Andrei Olaru and Adina Magda Florea . cs@andreiolaru.ro . IDC 2016, Paris, France 10.10.2016