Table 2 Tuning a search engine to attain two different scenarios of retrieval.
Scenario 1. Query with specificity of 99.99% is insufficient for a database of 16 million records. | ||||
|---|---|---|---|---|
| Â | Â | The truth | Â | |
| Â | Â | relevant records | irrelevant records | Â |
search engine | records returned to user | 495 | 1,600 | 2,095 |
| Â | records eliminated | 5 | 15,997,900 | Â |
| Â | Â | 500 | 15,999,500 | 16,000,000 |
odds ratio | 1,000,000.00 | Â | Â | Â |
Specificity | 99.99% | Â | Â | Â |
sensitivity (recall) | 99.01% | Â | Â | Â |
Precision | 23.63% | Â | Â | Â |
Scenario 2. The price for a very high specificity: Missing a large number of relevant records. | ||||
| Â | Â | The truth | Â | |
| Â | Â | relevant records | irrelevant records | Â |
search engine | records returned to user | 250 | 16 | 266 |
| Â | records eliminated | 250 | 15,999,484 | Â |
| Â | Â | 500 | 15,999,500 | 16,000,000 |
odds ratio | 1,000,000.00 | Â | Â | Â |
Specificity | 99.9999% | Â | Â | Â |
sensitivity (recall) | 50.00% | Â | Â | Â |
Precision | 93.99% | Â | Â | Â |