
Wirtschaftsinformatik / Very Large Business Applications
Diploma/Master thesis
Title
Beehive Resource Discovery on Top of an Instant Messaging Network
Description
- Pattern of behavior followed by mobile software agents similar to the foraging pattern of honeybees [3]
- When a bee finds a good source of nectar, it recruits other bees
- Good for discovery in dynamic environments
- Theoretical background about agents working as a beehive for resource sharing [1] [2] and [5]
- IM network is a virtualized P2P network
- Semantic annotation opens new horizons
- Attributes like: "Suitable for children" or "MobileOK" have a unified meaning and can be used automatically by machines to filter out, recommend or highlight web content. POWDER [4] provides high flexibility in resource grouping and in the semantic level of annotation.
- Privacy is preserved, your user model and data are on your PC, you have total control over them.
- No need to worry about privacy policies or you personal data being sold to advertisers by social network owners.
- Sending agents is better than sending full DR documents: bandwidth efficiency.
- People’s interests intersect in different small sections
- You share an interest with a friend, but you don't share all of his interests, so you don't need to transfer all of his annotations through the network and then filter them.
- A network of trust is already established
- No need to sign up in any new service and start collecting friends, you already have your network of friends
- You trust your contacts and they trust theirs, spammers are out of the equation.
- File transfer is supported on multiple IM protocols, which is almost the only feature needed for transferring agents.
- As well as management of contact grouping and presence.
- Tasks like presence management and contact grouping can be left to the IM client, which reduces the complexity of the suggested system
- Google Wave is an example of a successful application on top of an IM protocol (Wave Federation Protocol over XMPP)
Ansprechpartner
Dipl.-Inf. Ammar MemariReferences:
[1] Memari, A., Amer, M., & Marx Gómez, J. (2009). A Beehive-like Multi-agent Solution to Enhance Findability of Semantic Web Services and Facilitate Personalization within a P2P Network. In ICT Innovations 2009 (pp. 227-236). Presented at the ICT Innovations 2009, Ohrid, Macedonia: Springer Berlin Heidelberg. doi:http://dx.doi.org/10.1007/978-3-642-10781-8_24
[2] Memari, A., Aalmink, J., & López de Castro, R. P. (2010). A Beehive-like Approach for Dynamic Generation of Integration Services to Enable Adaptive Supply Chains using Enterprise Tomography. In 8th International Conference on Practical Applications of Agents and Multi-Agent Systems. Presented at the 8th International Conference on Practical Applications of Agents and Multi-Agent Systems, Salamanca, Spain: Springer Berlin Heidelberg.
[3] Navrat, P., Kovacik, M., Ezzeddine, A. B., & Rozinajova, V. (2008). Web search engine working as a bee hive. Web Intelli. and Agent Sys., 6(4), 441-452.
[4] Archer, P., Smith, K., & Perego, A. (n.d.). Protocol for Web Description Resources (POWDER): Description Resources. Retrieved September 8, 2009, from http://www.w3.org/TR/2009/REC-powder-dr-20090901/#intro
[5] Heyen, C. (2010, February 17). Diversity in Information Retrieval exemplary in a Multi-agent solution for Findability of Semantic Web Services in a P2P Network as a step towards Personalization (Diploma). Carl von Ossietzky Universität Oldenburg.




