Semantic Similarity based on Weighted Ontology - Geographical Information Systems: Trends and Technologies - page 316

impacts signifi cantly on the commonality parameters w c att , and w c part as are
shown in Table 4, and presented graphically in Figs. 6 and 7. In Fig. 6, we
observe that, as w att increases, w c part decreases (note that w c att + w c part =1). We
also observe that the correlation of GSim c ( GC c for short) increases with the
increase of w c att and varies from 0.72 to 0.83.

Table 4. Correlation of GSim and parameters by varying attributes' occurrences.

Occ. of Att. w part w att w part w c att GC v GC c

15 0.423 0.577 0.702 0.298 0.3335 0.7191

17 0.430 0.570 0.675 0.325 0.3902 0.7248

19 0.436 0.564 0.650 0.350 0.5708 0.7751

21 0.443 0.557 0.627 0.373 0.6067 0.7772

23 0.450 0.550 0.605 0.395 0.6503 0.7903

25 0.458 0.542 0.585 0.415 0.7214 0.8198

27 0.465 0.535 0.567 0.433 0.7748 0.8310

29 0.473 0.527 0.549 0.451 0.7997 0.8340

31 0.481 0.519 0.532 0.468 0.8148 0.8342

33 0.490 0.510 0.517 0.483 0.8242 0.8343

35 0.499 0.501 0.502 0.498 0.8332 0.8345

Commonality

1

w att
w part
GC

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

15 19 23 27 31 35

Attributes

Fig. 6. GC c correlation and commonality parameters by varying attributes' occurrences.

Color image of this figure appears in the color plate section at the end of the topic.