Table 3.
Evaluation of the LitInspector homonym resolution compared to three other data mining tools by means of the homonym MIZ-1
| Text mining software | MIZ-1 Myc-interacting zinc finger-1 (Gene ID: 7709) |
MIZ-1 Msx-interacting-zinc finger-1 (Gene ID: 9063) |
Unresolved homonyms | Correctly resolved homonyms (percent) | ||||
|---|---|---|---|---|---|---|---|---|
| Number matches | Correctly resolved | False positives | Number matches | Correctly resolved | False positives | |||
| LitInspector | 53 | 44 | 2 | 36 | 27 | 2 | 7 | 87% |
| EBIMed | 11 | 10 | 1 | 17 | 9 | 8 | 0 | 67% |
| PubGene | 45 | (9) | (1) | 34 | (2) | (8) | (0) | (55%a) |
| iHOP | 36 | 14 | 0 | 31 | 12 | 6 | 35 | 39% |
The programs were used with default parameters. All numbers refer to abstracts, i.e. sentences with the same identified gene from one paper were counted only once for this gene. The evaluation was performed in July 2008.
aPubGene provides only 10 example papers for each search, therefore, an evaluation of the complete results is not possible. This evaluation was performed using the 20 (2 × 10) abstracts available.