Mirror/MarginaliaSearch
1
0
Fork 0
mirror of https://github.com/MarginaliaSearch/MarginaliaSearch.git synced 2025-02-23 21:18:58 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

(index) Experimental ranking signals

Browse Source
This commit is contained in:
Viktor Lofgren 2024-08-03 10:33:41 +02:00
parent eba2844361
commit c2cedfa83c
2 changed files with 33 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

34
code/index/java/nu/marginalia/index/results/IndexResultScoreCalculator.java
View File

@ -211,13 +211,28 @@ public class IndexResultScoreCalculator {
temporalBias = 0;
}
int numCoherenceAll = coherences.countOptional(positions);
int bestCoherenceAll = coherences.testOptional(positions);
int bestCoherenceTitle = coherences.testOptional(positions, spans.title);
int bestCoherenceHeading = coherences.testOptional(positions, spans.heading);
float coherenceScore = 0.f;
boolean allInTitle = coherences.allOptionalInSpan(positions, spans.title);
boolean allInHeading = coherences.allOptionalInSpan(positions, spans.heading);
// Calculate a bonus for keyword coherences when large ones exist
int largestOptional = coherences.largestOptional();
if (largestOptional >= 2) {
int bestInTitle = coherences.testOptional(positions, spans.title);
int bestInHeading = coherences.testOptional(positions, spans.heading);
int best = coherences.testOptional(positions);
if (largestOptional == bestInTitle) {
coherenceScore = 2.0f * largestOptional;
}
else if (largestOptional == bestInHeading) {
coherenceScore = 1.5f * largestOptional;
}
else if (largestOptional == best) {
coherenceScore = 0.75f * largestOptional;
}
coherenceScore += (float) Math.pow(coherences.countOptional(positions) / (double) coherences.numOptional(), 2);
}
float[] weightedCounts = new float[compiledQuery.size()];
int firstPosition = Integer.MAX_VALUE;
@ -255,12 +270,7 @@ public class IndexResultScoreCalculator {
+ topologyBonus
+ temporalBias
+ flagsPenalty
+ bestCoherenceAll
+ bestCoherenceTitle
+ bestCoherenceHeading
+ numCoherenceAll / 4.
+ (allInTitle ? 5.0 : 0)
+ (allInHeading ? 2.5 : 0);
+ coherenceScore;
double tcfAvgDist = rankingParams.tcfAvgDist * (1.0 / calculateAvgMinDistance(positionsQuery, ctx));
double tcfFirstPosition = rankingParams.tcfFirstPosition * (1.0 / Math.max(1, firstPosition));

11
code/index/java/nu/marginalia/index/results/model/TermCoherenceGroupList.java
View File

@ -82,6 +82,17 @@ public class TermCoherenceGroupList {
return true;
}
public int numOptional() {
return optionalGroups.size();
}
public int largestOptional() {
int best = 0;
for (var coherenceSet : optionalGroups) {
best = Math.max(coherenceSet.size, best);
}
return best;
}
public static final class TermCoherenceGroup {
private final int[] offsets;