.. (לתיקייה המכילה) | ||
Q3: For the scoring can we iterate through the entire doc set? | |
Performance (complexity) is definitely a concern and you absolutely should not iterate through the entire doc stream. |
Q3: What types of heuristics can we use in the scoring? | |
A random value is not a justifiable heuristic. However, a default arbitrary number might be reasonable, as long as you can reason about why you chose it. In general we are looking for an approximation of a reasonable scoring value for each scorer. |
Wet part: Which version of Lucene should we use? | |
We suggest using Lucene 2.4. Please do not use the 3.X version. |