Difference between revisions of "Entity Summary"
Line 7: | Line 7: | ||
= Problem = | = Problem = | ||
'''Problem Statement''' - An entity is usually described using a conceptually different set of facts to improve coverage. We want to select a ‘representative’ subset of this set in a good summary to uniquely identify the entity. | '''Problem Statement''' - An entity is usually described using a conceptually different set of facts to improve coverage. We want to select a ‘representative’ subset of this set in a good summary to uniquely identify the entity. | ||
+ | |||
+ | Definitions 1-4 defines basic notions related to entity summaries. They are as stated in [1]. | ||
'''Definition 1''' : A data graph is a digraph G = V, A, Lbl<sub>V</sub> , Lbl<sub>A</sub> , where (i) V is a finite set of nodes, (ii) A is a finite set of directed edges where each a ∈ A has a source node Src(a) ∈ V, a target node Tgt(a) ∈ V, (iii) Llb<sub>V</sub> : V → E ∪ L and (iv) Lbl<sub>A</sub> : A → P are labeling functions that map nodes to entities or literals and edges to properties. | '''Definition 1''' : A data graph is a digraph G = V, A, Lbl<sub>V</sub> , Lbl<sub>A</sub> , where (i) V is a finite set of nodes, (ii) A is a finite set of directed edges where each a ∈ A has a source node Src(a) ∈ V, a target node Tgt(a) ∈ V, (iii) Llb<sub>V</sub> : V → E ∪ L and (iv) Lbl<sub>A</sub> : A → P are labeling functions that map nodes to entities or literals and edges to properties. | ||
Line 145: | Line 147: | ||
=== Dataset === | === Dataset === | ||
Evauation data is available for download [http://knoesis.wright.edu/researchers/kalpa/faces_evaluation.zip download] | Evauation data is available for download [http://knoesis.wright.edu/researchers/kalpa/faces_evaluation.zip download] | ||
+ | |||
+ | |||
+ | = References = | ||
+ | [1] Cheng, Gong, Thanh Tran, and Yuzhong Qu. "RELIN: relatedness and informativeness-based centrality for entity summarization." In The Semantic Web–ISWC 2011, pp. 114-129. Springer Berlin Heidelberg, 2011. |
Revision as of 16:53, 15 May 2014
Check back on Friday May 16th 2014 for more details...
Contents
Creating Faceted (divesified) Entity Summaries
Creating entity summaries has been of contemporary interest in the Semantic Web community in the recet past. In our approach called FACES: FACed Entity Summaries, we are interested in generating diversified and user friendly summaries.
Problem
Problem Statement - An entity is usually described using a conceptually different set of facts to improve coverage. We want to select a ‘representative’ subset of this set in a good summary to uniquely identify the entity.
Definitions 1-4 defines basic notions related to entity summaries. They are as stated in [1].
Definition 1 : A data graph is a digraph G = V, A, LblV , LblA , where (i) V is a finite set of nodes, (ii) A is a finite set of directed edges where each a ∈ A has a source node Src(a) ∈ V, a target node Tgt(a) ∈ V, (iii) LlbV : V → E ∪ L and (iv) LblA : A → P are labeling functions that map nodes to entities or literals and edges to properties.
Defintion 2 : A feature f is a property-value pair where Prop(f ) ∈ P and Val(f ) ∈ E ∪ L denote the property and the value, respectively. An entity e has a feature f in a data graph G = V, A, LblV , LblA if there exists a ∈ A such that LblA (a) = Prop(f ), LblV (Src(a)) = e and LblV (Tgt(a)) = Val(f ).
Definition 3 : Given a data graph G, the feature set of an entity e, denoted by FS(e), is the set of all features of e that can be found in G.
Definition 4 : Given FS(e) and a positive integer k < |FS(e)|, summary of entity e is Summ(e) ⊂ FS(e) such that |Summ(e)| = k.
Evaluation
System | k = 5 | FACES % ↑ | k = 10 | FACES % ↑ | time/entity in seconds |
---|---|---|---|---|---|
FACES | 1.4314 | NA | 4.3350 | NA | 0.76 sec. |
RELIN | 0.4981 | 187 % | 2.5188 | 72 % | 10.96 sec. |
RELINM | 0.6008 | 138 % | 3.0906 | 40 % | 11.08 sec. |
SUMMARUM | 1.2249 | 17 % | 3.4207 | 27 % | NA |
Ideal summ agreement | 1.9168 | 4.6415 |
System | k = 5 | FACES %↑ | k = 10 | FACES %↑ |
---|---|---|---|---|
FACES | 1.8649 | NA | 5.6931 | NA |
RELIN | 0.7339 | 154 % | 3.3993 | 69 % |
RELINM | 0.8695 | 114 % | 4.1551 | 37 % |
SUMMARUM | 1.6484 | 13 % | 4.4919 | 27 % |
Ideal summ agreement | 2.3194 | 5.6228 |
Experiment | FACES % | RELINM % | SUMMARUM % |
---|---|---|---|
Experiment 1 | 84 % | 16 % | NA |
Experiment 2 | 54 % | 16 % | 30 % |
k = 5 | k = 10 | ||
---|---|---|---|
Google search API | Sindice seach API | Google search API | Sindice search API |
3.5 | 3.4 | 0.5333 | 0.5428 |
Dataset
Evauation data is available for download download
References
[1] Cheng, Gong, Thanh Tran, and Yuzhong Qu. "RELIN: relatedness and informativeness-based centrality for entity summarization." In The Semantic Web–ISWC 2011, pp. 114-129. Springer Berlin Heidelberg, 2011.