The document describes an ontological web-trading model for information retrieval using semantic technologies. It discusses the context of complex information retrieval across multiple heterogeneous sources and proposes using a knowledge representation system and trading service mediated by ontologies to address this. The ontological web-trading model is described as having properties like supporting heterogeneous data, service federation, extensibility and using heuristics to enable scalable information retrieval across distributed sources. An example system called SOLERES is also mentioned.

1. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
Information Retrieval Using an
Ontological Web-Trading Model
José-Andrés Asensio, Nicolás Padilla, Luis Iribarne
{jacortes, npadilla, luis.iribarne}@ual.es
Applied Computing Group (TIC-211)
University of Almería, SPAIN
FedCSIS, ASIR’13, Kraków, POLAND
8-11 September, 2013
Project TIN2010-15588 / Project TIC-6114

2. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
2
1. Context
2. OWT Model Properties
3. OWT Model Operation
4. Case Study: SOLERES System
5. WTA Implementation
6. Conclusions and Future Work
Contents

3. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
3
1. Context (1/4)
• Hypothesis:
Systems
– Specific systems.

4. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
4
1. Context (1/4)
• Hypothesis:
Systems
– Specific systems.
– Huge amount of information
dispersed in multiple sources.

5. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
5
1. Context (1/4)
• Hypothesis:
Systems
– Specific systems.
– Huge amount of information
dispersed in multiple sources.
– Heterogeneous info.
(but structured data).
Source n
Source 5
Source 1
Source 2
Source 3
Source 4

6. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
6
1. Context (1/4)
• Hypothesis:
Systems

Complexity of the information
searching/retrieval processes
– Specific systems.
– Huge amount of information
dispersed in multiple sources.
– Heterogeneous info.
(but structured data).
Source n
Source 5
Source 1
Source 2
Source 3
Source 4

7. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
7
1. Context (2/4)
• Solution:
Source n
Source 5
Source 1
Source 2
Source 3
Source 4

8. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
8
1. Context (2/4)
• Solution:
– Trading Service (Trader).
Source n
Source 5
Source 1
Source 2
Source 3
Source 4
Trader

9. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
9
1. Context (2/4)
• Solution:
– Trading Service (Trader).
Source n
Source 5
Source 1
Source 2
Source 3
Source 4
Trader
KRS

Knowledge Representation
System (KRS)

10. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
10
1. Context (2/4)
• Solution:
– Knowledge Representation
System (KRS).
Source n
Source 5
Source 1
Source 2
Source 3
Source 4
Trader
KRS
Model-Driven
Engineering (MDE):
> System models <
Ontology-Driven
Engineering (ODE):
> Data ontologies <
> Service ontologies <

11. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
11
1. Context (3/4)
• Retrieval mechanism 
 Query–Searching / Recovering–Response Model:
• Query  process of creating and formulating the request.
• Searching  process of locating the data sources.
• Recovering  process of selecting the data from the sources.
• Response  process of creation of the response to the user.

12. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
12
1. Context (3/4)
• Retrieval mechanism 
 Query–Searching / Recovering–Response Model:
• Query  process of creating and formulating the request.
• Searching  process of locating the data sources.
• Recovering  process of selecting the data from the sources.
• Response  process of creation of the response to the user.

13. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
13
1. Context (3/4)
• Retrieval mechanism 
 Query–Searching / Recovering–Response Model:
• Query  process of creating and formulating the request.
• Searching  process of locating the data sources.
• Recovering  process of selecting the data from the sources.
• Response  process of creation of the response to the user.

14. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
14
1. Context (3/4)
• Retrieval mechanism 
 Query–Searching / Recovering–Response Model:
• Query  process of creating and formulating the request.
• Searching  process of locating the data sources.
• Recovering  process of selecting the data from the sources.
• Response  process of creation of the response to the user.

15. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
15
1. Context (3/4)
• Retrieval mechanism 
 Query–Searching / Recovering–Response Model:
• Query  process of creating and formulating the request.
• Searching  process of locating the data sources.
• Recovering  process of selecting the data from the sources.
• Response  process of creation of the response to the user.

16. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
16
1. Context (4/4)
• From the ODP Trading Function a Trader is a software
object that mediates between objects that offer certain
capacities or services (Exporters) and other objects that
demand their use dynamically (Importers).
• Interfaces:
– Lookup.
– Register.
– Admin.
– Link.
– Proxy.
Roles

17. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
17
1. Context (4/4)
• From the ODP Trading Function a Trader is a software
object that mediates between objects that offer certain
capacities or services (Exporters) and other objects that
demand their use dynamically (Importers).
• Interfaces:
– Lookup.
– Register.
– Admin.
– Link.
– Proxy.
Roles

18. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
18
1. Context (4/4)
• From the ODP Trading Function a Trader is a software
object that mediates between objects that offer certain
capacities or services (Exporters) and other objects that
demand their use dynamically (Importers).
• Interfaces:
– Lookup.
– Register.
– Admin.
– Link.
– Proxy.
Roles

19. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
19
1. Context (4/4)
• From the ODP Trading Function a Trader is a software
object that mediates between objects that offer certain
capacities or services (Exporters) and other objects that
demand their use dynamically (Importers).
• Interfaces:
– Lookup.
– Register.
– Admin.
– Link.
– Proxy.
Roles

20. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
20
1. Context (4/4)
• From the ODP Trading Function a Trader is a software
object that mediates between objects that offer certain
capacities or services (Exporters) and other objects that
demand their use dynamically (Importers).
• Interfaces:
– Lookup.
– Register.
– Admin.
– Link.
– Proxy.
Roles

21. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
21
1. Context (4/4)
• From the ODP Trading Function a Trader is a software
object that mediates between objects that offer certain
capacities or services (Exporters) and other objects that
demand their use dynamically (Importers).
• Interfaces:
– Lookup.
– Register.
– Admin.
– Link.
– Proxy.
• Communication  Ontologies:
– Data Ontologies.
– Service Ontologies:
 Lookup Ontology.
 Register Ontology.
 Admin Ontology.
 Link Ontology.
 Proxy Ontology.

22. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
22
2. OWT Model Properties (1/1)
• Ontological Web-Trading (OWT)  Properties:
1. Heterogeneus data model.
2. Federation.
3. Composition and adaptation of services.
4. Weak pairing.
5. Usage of heuristics and metrics.
6. Extensible and scalable.
7. “Storage and forwarding” policy.
8. Delegation.
9. Push and pull storage model.

23. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
23
2. OWT Model Properties (1/1)
• Ontological Web-Trading (OWT)  Properties:
1. Heterogeneus data model.
2. Federation.
3. Composition and adaptation of services.
4. Weak pairing.
5. Usage of heuristics and metrics.
6. Extensible and scalable.
7. “Storage and forwarding” policy.
8. Delegation.
9. Push and pull storage model.

24. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
24
2. OWT Model Properties (1/1)
• Ontological Web-Trading (OWT)  Properties:
1. Heterogeneus data model.
2. Federation.
3. Composition and adaptation of services.
4. Weak pairing.
5. Usage of heuristics and metrics.
6. Extensible and scalable.
7. “Storage and forwarding” policy.
8. Delegation.
9. Push and pull storage model.

25. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
25
2. OWT Model Properties (1/1)
• Ontological Web-Trading (OWT)  Properties:
1. Heterogeneus data model.
2. Federation.
3. Composition and adaptation of services.
4. Weak pairing.
5. Usage of heuristics and metrics.
6. Extensible and scalable.
7. “Storage and forwarding” policy.
8. Delegation.
9. Push and pull storage model.

26. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
26
2. OWT Model Properties (1/1)
• Ontological Web-Trading (OWT)  Properties:
1. Heterogeneus data model.
2. Federation.
3. Composition and adaptation of services.
4. Weak pairing.
5. Usage of heuristics and metrics.
6. Extensible and scalable.
7. “Storage and forwarding” policy.
8. Delegation.
9. Push and pull storage model.

27. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
27
2. OWT Model Properties (1/1)
• Ontological Web-Trading (OWT)  Properties:
1. Heterogeneus data model.
2. Federation.
3. Composition and adaptation of services.
4. Weak pairing.
5. Usage of heuristics and metrics.
6. Extensible and scalable.
7. “Storage and forwarding” policy.
8. Delegation.
9. Push and pull storage model.

28. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
28
2. OWT Model Properties (1/1)
• Ontological Web-Trading (OWT)  Properties:
1. Heterogeneus data model.
2. Federation.
3. Composition and adaptation of services.
4. Weak pairing.
5. Usage of heuristics and metrics.
6. Extensible and scalable.
7. “Storage and forwarding” policy.
8. Delegation.
9. Push and pull storage model.

29. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
29
2. OWT Model Properties (1/1)
• Ontological Web-Trading (OWT)  Properties:
1. Heterogeneus data model.
2. Federation.
3. Composition and adaptation of services.
4. Weak pairing.
5. Usage of heuristics and metrics.
6. Extensible and scalable.
7. “Storage and forwarding” policy.
8. Delegation.
9. Push and pull storage model.

30. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
30
2. OWT Model Properties (1/1)
• Ontological Web-Trading (OWT)  Properties:
1. Heterogeneus data model.
2. Federation.
3. Composition and adaptation of services.
4. Weak pairing.
5. Usage of heuristics and metrics.
6. Extensible and scalable.
7. “Storage and forwarding” policy.
8. Delegation.
9. Push and pull storage model.

31. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
31
2. OWT Model Properties (1/1)
• Ontological Web-Trading (OWT)  Properties:
1. Heterogeneus data model.
2. Federation.
3. Composition and adaptation of services.
4. Weak pairing.
5. Usage of heuristics and metrics.
6. Extensible and scalable.
7. “Storage and forwarding” policy.
8. Delegation.
9. Push and pull storage model.

32. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
32
3. OWT Model Operation (1/1)
• Elements:
<Interface Object (I), Trading Service (T), System Data (D)>

33. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
33
3. OWT Model Operation (1/1)
• Elements:
<Interface Object (I), Trading Service (T), System Data (D)>

34. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
34
3. OWT Model Operation (1/1)
• Elements:
<Interface Object (I), Trading Service (T), System Data (D)>

35. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
35
3. OWT Model Operation (1/1)
• Elements:
<Interface Object (I), Trading Service (T), System Data (D)>

36. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
36
4. Case Study: SOLERES System (1/1)
• SOLERES System
architecture:

37. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
37
4. Case Study: SOLERES System (1/1)
• SOLERES System
architecture:
SOLERES-HCI

38. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
38
4. Case Study: SOLERES System (1/1)
• SOLERES System
architecture:
SOLERES-KRS
– EID (Environmental
Information metaData)
– EIM (Environmental
Information Map)

39. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
39
5. WTA Implementation (1/3)
• Web-Trading
Agent (WTA)
view:

40. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
40
5. WTA Implementation (1/3)
• Web-Trading
Agent (WTA)
view:

41. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
41
5. WTA Implementation (1/3)
• Web-Trading
Agent (WTA)
view:

42. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
42
5. WTA Implementation (1/3)
• Web-Trading
Agent (WTA)
view:

43. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
43
5. WTA Implementation (1/3)
• Web-Trading
Agent (WTA)
view:

44. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
44
5. WTA Implementation (2/3)
• Data ontologies  EID metadata:

45. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
45
5. WTA Implementation (2/3)
• Data ontologies  EID metadata:

46. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
46
5. WTA Implementation (2/3)
• Data ontologies  EID metadata:

47. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
47
5. WTA Implementation (2/3)
• Data ontologies  EID metadata:

48. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
48
5. WTA Implementation (2/3)
• Data ontologies  EID metadata:

49. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
49
5. WTA Implementation (2/3)
• Data ontologies  EID metadata:

50. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
50
5. WTA Implementation (2/3)
• Data ontologies  EID metadata:

51. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
51
5. WTA Implementation (3/3)
• Service ontologies  Lookup Ontology:

52. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
52
5. WTA Implementation (3/3)
• Service ontologies  Lookup Ontology:
Concepts

53. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
53
5. WTA Implementation (3/3)
• Service ontologies  Lookup Ontology:
Action

54. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
54
5. WTA Implementation (3/3)
• Service ontologies  Lookup Ontology:
Predicates

55. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
55
6. Conclusions and Future Work (1/1)
• Conclusions:
– Web-based Information Systems (WIS) facilitate information
search and retrieval, favoring user cooperation and decision
making.

56. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
56
6. Conclusions and Future Work (1/1)
• Conclusions:
– Web-based Information Systems (WIS) facilitate information
search and retrieval, favoring user cooperation and decision
making.

– Ontologies provide them a shared vocabulary.
– Trading services improve the component interoperability and
information retrieval.

57. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
57
6. Conclusions and Future Work (1/2)
• Conclusions:
– Web-based Information Systems (WIS) facilitate information
search and retrieval, favoring user cooperation and decision
making.

– Ontologies provide them a shared vocabulary.
– Trading systems improve the component interoperability and
information retrieval.

– We have introduced Ontological Web-Trading (OWT) model as
an extension of the traditional ODP trading service.

58. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
58
6. Conclusions and Future Work (2/2)
• Future work:
– The implementation of SOLERES-HCI by using multi-agent
architectures.
– To study how to decompose the user tasks into actions to be
performed by the SOLERES-KRS.
– To incorporate evaluation and validation techniques.

59. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
Information Retrieval Using an
Ontological Web-Trading Model
Thank you for your attention!!
Contact: jacortes@ual.es
Applied Computing Group (TIC-211)
University of Almería, SPAIN
FedCSIS, ASIR’13, Kraków, POLAND
8-11 September, 2013
Project TIN2010-15588 / Project TIC-6114

60. 3rd International Workshop on Advances in Semantic Information Retrieval
Kraków, POLAND, 8-11 September, 2013
60
Contraportada