Table 4.
Relevance of newly found videos by the number of links to the original set of colonoscopy videos (total degree).
Total degreea | Count of videos with total degree, N | Number of videos coded as “relevant” (relevancy), n (%) | Cumulative count of nonduplicate videos, N | Cumulative count of nonduplicate relevant videos, n | Cumulative precisionb (%) | Cumulative recallc, (%) | Cumulative F1-scored, (%) |
44 | 1 | 1 (100) | 1 | 1 | 100 | 2.7 | 5.3 |
41 | 1 | 1 (100) | 2 | 2 | 100 | 5.4 | 10.3 |
26 | 1 | 1 (100) | 3 | 3 | 100 | 8.1 | 15.0 |
23 | 1 | 1 (100) | 4 | 4 | 100 | 10.8 | 19.5 |
22 | 1 | 1 (100) | 5 | 5 | 100 | 13.5 | 23.8 |
21 | 1 | 1 (100) | 6 | 6 | 100 | 16.2 | 27.9 |
20 | 2 | 2 (100) | 8 | 8 | 100 | 21.6 | 35.6 |
19 | 1 | 1 (100) | 9 | 9 | 100 | 24.3 | 39.1 |
18 | 1 | 1 (100) | 10 | 10 | 100 | 27.0 | 42.6 |
17 | 2 | 2 (100) | 12 | 12 | 100 | 32.4 | 49.0 |
16 | 1 | 1 (100) | 13 | 13 | 100 | 35.1 | 52.0 |
15 | 2 | 2 (100) | 15 | 15 | 100 | 40.5 | 57.7 |
14 | 1 | 1 (100) | 16 | 16 | 100 | 43.2 | 60.4 |
13 | 1 | 1 (100) | 17 | 17 | 100 | 45.9 | 63.0 |
12 | 2 | 2 (100) | 19 | 19 | 100 | 51.4 | 67.9 |
11 | 2 | 2 (100) | 21 | 21 | 100 | 56.8 | 72.4 |
10 | 1 | 1 (100) | 22 | 22 | 100 | 59.5 | 74.6 |
9 | 1 | 1 (100) | 23 | 23 | 100 | 62.2 | 76.7 |
7 | 2 | 1 (50) | 25 | 24 | 96 | 64.9 | 77.4 |
6 | 1 | 1 (100) | 26 | 25 | 96 | 67.6 | 79.4 |
5 | 2 | 0 (0) | 28 | 25 | 89 | 67.6 | 76.9 |
4 | 2 | 2 (100) | 30 | 27 | 90 | 73.0 | 80.6 |
3 | 2 | 1 (50) | 32 | 28 | 88 | 75.7 | 81.2 |
2 | 5 | 1 (20) | 37 | 29 | 78 | 78.4 | 78.4 |
1 | 5 | 1 (20) | 42 | 30 | 71 | 81.1 | 75.9 |
0 | 71 | 7 (10) | 113 | 37 | 33 | 100 | 49.3 |
aThe sum of connections each new video has with the videos in the original colonoscopy video set.
bThe cumulative count of relevant videos divided by the cumulative count of all videos.
cCumulative count of relevant videos divided by the total number of new and nonduplicate 37 relevant videos.
dThe harmonic mean of cumulative precision and cumulative recall.