Skip to main content

Table 4 Diverse Expansion Scenarios: Results of Word Expansion Using Varied Similarity and Probability Settings

From: Empowering health geography research with location-based social media data: innovative food word expansion and energy density prediction via word embedding and machine learning

 

Food word expansion scenarios

 

A

B

C

D

E

F

G

Similarity Level

0.55

0.55

0.6

0.6

0.65

0.7

0.8

Probability Level

0.7

0.8

0.7

0.8

0.75

0.7

0.8

Number of expanded words

32,637

32,637

19,826

19,826

11,957

5,636

378

Accuracy

71%

75%

83%

92%

94%

94%

94%

Percentage of L-ED food words

6.08%

6.08%

5.98%

5.98%

5.02%

4.67%

5.82%

Percentage of H-ED food words

93.92%

93.92%

94.02%

94.02%

94.98%

95.33%

94.18%

  1. The columns labeled with alphabetic identifiers A to G depict distinct expansion scenarios, each defined by a unique combination of similarity and probability levels