• Aucun résultat trouvé

Reality is not a Game! Extracting Semantics From Unconstrained Navigation on Wikipedia

N/A
N/A
Protected

Academic year: 2022

Partager "Reality is not a Game! Extracting Semantics From Unconstrained Navigation on Wikipedia"

Copied!
1
0
0

Texte intégral

(1)

Reality is not a game!

Extracting Semantics from Unconstrained Navigation on Wikipedia

Thomas Niebler, Daniel Schl¨or, Martin Becker, and Andreas Hotho

University of Wuerzburg Tel.: +49-931-31-89094

{niebler,schloer,becker,hotho}@informatik.uni-wuerzburg.de

Abstract. Semantic relatedness between words has been successfully extracted from navigation on Wikipedia pages. However, the navigational data used in the corresponding works are sparse and expected to be biased since they have been collected in the context of games. In this paper, we raise this limitation and ex- plore if semantic relatedness can also be extracted from unconstrained naviga- tion. To this end, we first highlight structural differences between unconstrained navigation and game data. Then, we adapt a state of the art approach to extract semantic relatedness on Wikipedia paths. We apply this approach to transitions derived from two unconstrained navigation datasets as well as transitions from WikiGame and compare the results based on two common gold standards. We confirm expected structural differences when comparing unconstrained naviga- tion with the paths collected by WikiGame. In line with this result, the mentioned state of the art approach for semantic extraction on navigation data does not yield good results for unconstrained navigation. Yet, we are able to derive a related- ness measure that performs well on both, unconstrained navigation data as well as game data. Overall, we show that unconstrained navigation data on Wikipedia is suited for extracting semantics.

The original paper is currently under review [Niebler et al(2015)].

References

Niebler et al(2015). Niebler T, Schl¨or D, Becker M, Hotho A (2015) Extracting semantics from unconstrained navigation on wikipedia. K¨unstliche Intelligenz - Special Issue: Semantic Web currently under review

Copyrightc 2015 by the paper’s authors. Copying permitted only for private and academic purposes.In: R. Bergmann, S. G¨org, G. M¨uller (Eds.): Proceedings of the LWA 2015 Work- shops: KDML, FGWM, IR, and FGDB. Trier, Germany, 7.-9. October 2015, published at http://ceur-ws.org

19

Références

Documents relatifs

In order to determine patterns in the perception of well-received structures (as per the building rankings), we correlate the influential factors of buildings with concrete prop-

These examples underline the need to augment current navigation languages with capabilities to extract fragments (i.e., subgraphs) of the graphs being navigated besides of sets

The approach extracts structured discourse graphs (SDGs) from texts, a representation introduced in [5] which focuses on a RDF compatible graph representation which maximizes

We imple- mented a processing pipeline consisting of a high-performance semantic role labeler to extract predicate–argument structures and a converter using VerbNet thematic roles

Ongoingly, we constructed a simple but versatile neural net- work for neural network argumentation, so that it can decide which argumentation semantics (admissible, stable,

i.OB cropping: this step is responsible for loading a list of classes, instances and properties of the OB domain ontology, which will be the basis for the search of the

We apply a distant supervised learning algorithm on a sub-set of Wikipedia in French made of disambiguation pages where we manually annotated hypernym relations.. The learned

This work presents a proposal for the development of a natural language processing module for event and temporal analysis of biographies as available in Wikipedia. At the current