![](bg8.png)
If the number of hops to a remote node exceeds the
depth of the taxonomy tree, no information is kept
about this remote node. The taxonomy thus serves
here as a summarised directory of documents in the
local neighbourhood and is not designed for routing
queries to far distant nodes.
In (Koloniari and Pitoura, 2004), knowledge is
kept in XML files by nodes that are organised in a
hierarchy. Each node possesses a local Bloom filter
that is triggered whenever an incoming query poten-
tially matches with locally stored knowledge. Nodes
also contain a merged Bloom filter that encompasses
the Bloom filters of the child nodes. A query match-
ing to a merged filter is redirected to child nodes. If
no positive local or child matching is obtained, the
query is redirected to the parent node. The approach
is rather suited for static P2P networks as nodes have
to stay in a hierarchy. The resource capacities of the
root node have to be sufficiently dimensioned since it
has to handle the most search and update queries. In
addition, the usage of Bloom filters may lead to a high
number of false positive matchings and thus routings,
if they are not configured and adapted well.
In (Jacobson et al., 2009) and (Gritter and Cheri-
ton, 2001), hierarchical names, similar to the taxon-
omy paths in our work, are used for content-based
routing. Also, routing entries are merged to reduce
the size of routing tables. However, taxonomies are
not applied. Instead, a hierarchical name could start
with the URL of the owner of a content. Consistent
names and annotations of content are not given. Ad-
ditionally, typical semantic queries cannot be used as
no solver is included in the routing procedure.
6 CONCLUSION
In this paper, we have presented a novel routing mech-
anism for loosely coupled networks. The main con-
tribution are the creation of semantic routing tables
and an efficient update mechanism. Their entries are
based on semantic aggregations, which rely on tax-
onomies and a distance measure to obtain the shortest
path to queried individuals. The usage of ASP enable
the application of the same formalism for knowledge
representation, reasoning, and query resolution. The
applied taxonomies are automatically extracted from
a commonsense knowledge source and are provided
as an ontology in ASP. Additionally, they can be dy-
namically adapted to the current situation. The frame-
work for the extraction has been presented in (Jakob
et al., 2021).
In our future work, we plan to integrate the
presented semantic routing mechanism into the dis-
tributed and multi-agent-based knowledge base intro-
duced in (Jakob et al., 2020). To prevent that all rout-
ing entries are merged to the same base class, we plan
to provide methods to adjust the depth of the aggre-
gation. Finally, we want to comprehensively evaluate
the proposed routing algorithm in a large scale search
& rescue scenario simulation
3
and aim to compare the
presented approach with other semantic or content-
based approaches.
REFERENCES
Brewka, G., Eiter, T., and Truszczy
´
nski, M. (2011). Answer
Set Programming at a Glance. Communications of the
ACM, 54(12):92–103.
Gebser, M., Kaminski, R., Kaufmann, B., and Schaub,
T. (2014). Clingo=ASP+Control: Extended Report.
Technical report, Knowledge Processing and Informa-
tion Systems.
G
´
omez Santill
´
an, C., Cruz Reyes, L., Meza Conde, E.,
Schaeffer, E., and Castilla Valdez, G. (2010). A
Self-Adaptive Ant Colony System for Semantic Query
Routing Problem in P2P Networks. Computaci
´
on y
Sistemas, 13(4):433–448.
Gritter, M. and Cheriton, D. R. (2001). An architecture for
content routing support in the internet. In USITS, vol-
ume 1, pages 4–4.
Jacobson, V., Smetters, D. K., Thornton, J. D., Plass, M. F.,
Briggs, N. H., and Braynard, R. L. (2009). Network-
ing named content. In Proceedings of the 5th inter-
national conference on Emerging networking experi-
ments and technologies, pages 1–12.
Jakob, S., Jahl, A., Baraki, H., and Geihs, K. (2020). A Self-
Organizing Multi-Agent Knowledge Base. Accepted
for publication at IEEE ICWS 2020.
Jakob, S., Jahl, A., Baraki, H., and Geihs, K. (2021). Gen-
erating Commonsense Ontologies with Answer Set
Programming. Accepted for publication at ICAART
2021.
Koloniari, G. and Pitoura, E. (2004). Content-based Rout-
ing of Path Queries in Peer-to-Peer Systems. In Inter-
national Conference on Extending Database Technol-
ogy, pages 29–47. Springer.
Michlmayr, E., Pany, A., and Kappel, G. (2007). Using Tax-
onomies for Content-based Routing with Ants. Com-
puter Networks, 51(16):4514–4528.
Pireddu, L. and Nascimento, M. A. (2004). Taxonomy-
Based Routing Indices for Peer-to-Peer Networks. In
Workshop on Peer-to-Peer Information Retrieval.
Speer, R., Chin, J., and Havasi, C. (2017). ConceptNet 5.5:
An Open Multilingual Graph of General Knowledge.
In Proceedings of the Thirty-First AAAI Conference
on Artificial Intelligence, pages 4444–4451.
Visala, K., Lagutin, D., and Tarkoma, S. (2009). LANES:
An Inter-Domain Data-Oriented Routing Architec-
ture. In Proceedings of the 2009 workshop on Re-
architecting the internet, pages 55–60.
3
https://www.emergencity.de/ (December 16, 2020)
ICAART 2021 - 13th International Conference on Agents and Artificial Intelligence
1004