| Institution: |
Slovak University of Technology |
| Technologies used: |
C#.NET, MongoDB |
| Inputs: |
maintenance rules, information tags, changesets |
| Outputs: |
consistent information tags |
Addressed problem
Semantics over large information spaces (e.g., a content management system of a software company) is crucial for systems, e.g. for effective information exploration, navigation or search. But current coverage of information systems by semantics is insufficient, so systems create various metadata. These metadata contain simple structured information retrieved from the source content (e.g., keywords, concepts) and users’ activity (e.g., clicks on links, a time spend on a webpage). By metadata systems build a lightweight semantics over information spaces, in which they can share metadata about the content in information spaces and interconnections between information artifacts, which include not only content based metadata, but also metadata based on user activities and collaboration. Crucial problem of metadata lies in dynamicity of information spaces those content is continually modified. This with influence of time can lead to invalidation of metadata which are closely related to source content. Due to invalidation of metadata and their amount, these metadata have to be automatically maintained and represented in a model which allows effective maintenance of metadata.
Description
Problem of metadata maintenance has not any sufficient solution. But this problem can be divided to two partly indifferent sub-problems. The first is maintenance of metadata anchoring or identification of source code after modifications, which has been addressed by metadata. We solved this problem by proposition of a robust location descriptor for source code [DEXA’13]. The robust location descriptor use specific source code features and context of the location for identification of source code addressed by metadata.
The second problem is maintenance of bodies of metadata. This problem is not solved in current approaches of metadata maintenance at all. We proposed information tags maintenance which engages maintenance rules, that consists of stream queries and actions. Stream queries are evaluated on streamed activities and source code changesets. When a query is matched an assigned update action is executed and affected information tags are maintained.
References
Rástočný, Karol – Bieliková, Mária: Maintenance of Knowledge Tags within Heterogeneous Web Content. In Proceedings of Current Trends in Web Engi-neering: ICWE 2012 International Workshops MDWE, Composable Web, WeRE, QWE, and Doctoral Consortium, LNCS, Vol. 7703, Springer-Verlag, Ber-lin Heidelberg, 2012, pp. 216–220.
Bieliková, Mária – Rástočný, Karol: Lightweight Semantics over Web Information Systems Content Employing Knowledge Tags. In: S. Castano et al. (Eds.): ER Workshops 2012, LNCS, Vol. 7518, Springer-Verlag, Berlin Heidelberg, 2012, pp. 327–336.
Rástočný, Karol – Bieliková, Mária: Metadata Anchoring for Source Code: Ro-bust Location Descriptor Definition, Building and Interpreting. In: H. Decker et al. (Eds.): DEXA 2013, Part II, LNCS, Vol. 8056, Springer, Heidelberg, 2013, pp. 372-379.
Rástočný, Karol - Bieliková, Mária: Enriching Source Code by Empirical Metadata. In ESEM 2014: 8th ACM/IEEE International Symposium on Empiri-cal Software Engineering and Measurement, New York: ACM, 2014, ISBN 978-1-4503-2774-9, p. 1.
Rástočný, Karol: Knowledge Tags Maintenance in Heterogeneous Web Content: The Repository. In: Information Sciences and Technologies. Bulletin of the ACM Slovakia. - ISSN 1338-1237. - Vol. 4, No. 2, 2012, pp. 59-61.
Barla, Michal – Rástočný, Karol – Tvarožek, Jozef – Bieliková, Mária: Značky ako integrujúci prvok heterogénnych metód pre adaptívnu podporu vývoja softvéru. In Proceedings of the Annual Database Conference (DATAKON 2012), October 14-16, 2012. Košice: Technická univerzita, 2012, ISBN 978-80-553-1049-7, pp. 107-116. (in Slovak)