Database Reference
In-Depth Information
Fig. 9.16 Part of the hypernode database in s tance
Finally, transformed data are loaded into the LHD schema. An excerpt of the HD
is shown in Fig. 9.16 .
9.4.2 Social Network Extraction
Using the resulting hypernode database from the previous step, the social network is
extracted. This phase passes through two steps: (1) Entities (people) identification
and (2) detection of relations among people.
The social network is defined by SN
¼
( ESN, RSN ), where:
l ESN is a finite set of entity such ESN
, where h is
the hypernode, which represents e , e n is e 's name, and N e is the set of e 's node.
l RSN is a finite set of the relations between entities such RSN
:¼f
e
n
e
2
HI
;
e
:¼h
h
;
e n ;
N e ig
¼
f
r sn n
r sn
h
n
;
e 1 ;
e 2 ig
, where n is the relation name, and e 1 and e 2 2
ESN .
In what follows, we describe the two steps of the social network extraction.
9.4.2.1 Entities Identification
Entities identification is the process of identifying hypernodes that contain entities
which compose the social network. In this step, we describe the process for
identifying people. The hypernode database schema is used to extract candidate
hypernodes (hypernodes that may contain persons). Then, the hypernodes' instances
Search WWH ::




Custom Search