root/geo
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat(monitoring): match citations by canonical link and scope summary by brand/keyword/platform/date

Browse Source 
Replace the domain+title fuzzy alias scoring with a canonical
candidate-key index built from published-link aliases — only exact
matches now resolve to a SaaS source, eliminating false positives
across articles sharing a host (e.g. m.163.com vs www.163.com).

Surface the summary scoping that already existed on loadCitationRanking
/ loadCitedArticles by accepting brand_id, keyword_id, ai_platform_id
and business_date on CitationSummary, plumbing them through the handler
and admin tracking view.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
liangxu
2026-05-01 11:46:17 +08:00
parent 2436f50c1f
commit b345ee26e4
10 changed files with 592 additions and 307 deletions
Download Patch File Download Diff File Expand all files Collapse all files
server/internal/tenant/app/monitoring_callback_service_test.go
+22 -12
View File
@@ -425,23 +425,33 @@ func TestBuildMonitoringCitationFactCanDisableContentResolution(t *testing.T) {
}
}
func TestScoreMonitoringAliasCandidateUsesDomainWithTitleEvidence(t *testing.T) {
lastPath := "article-42"
func TestScoreMonitoringAliasCandidateUsesCanonicalArticleKey(t *testing.T) {
input := monitoringAliasLookupInput{
NormalizedURL: "https://m.example.com/article-42",
RegistrableDomain: "example.com",
LastPathSegment: &lastPath,
TitleKey: normalizeMonitoringAliasTitle("北京口腔医院全面测评"),
NormalizedURL: "https://m.163.com/dy/article/KRPRE8DJ0556MCR0.html",
MatchKeys: monitoringCitationCandidateKeys("https://m.163.com/dy/article/KRPRE8DJ0556MCR0.html", ""),
}
candidate := monitoringAliasCandidate{
NormalizedURL: "https://www.example.com/posts/article-42",
RegistrableDomain: "example.com",
LastPathSegment: "article-42",
TitleKey: normalizeMonitoringAliasTitle("北京口腔医院全面测评"),
NormalizedURL: "https://www.163.com/dy/article/KRPRE8DJ0556MCR0.html",
MatchKeys: monitoringCitationCandidateKeys("https://www.163.com/dy/article/KRPRE8DJ0556MCR0.html", ""),
}
if score := scoreMonitoringAliasCandidate(input, candidate); score < 90 {
t.Fatalf("scoreMonitoringAliasCandidate() = %d, want strong domain/title match", score)
if score := scoreMonitoringAliasCandidate(input, candidate); score != 100 {
t.Fatalf("scoreMonitoringAliasCandidate() = %d, want canonical URL match", score)
}
}
func TestScoreMonitoringAliasCandidateRejectsSameDomainDifferentArticle(t *testing.T) {
input := monitoringAliasLookupInput{
NormalizedURL: "https://www.163.com/dy/article/KIN5DDMK0556GQZ8.html",
MatchKeys: monitoringCitationCandidateKeys("https://www.163.com/dy/article/KIN5DDMK0556GQZ8.html", ""),
}
candidate := monitoringAliasCandidate{
NormalizedURL: "https://www.163.com/dy/article/KRPRE8DJ0556MCR0.html",
MatchKeys: monitoringCitationCandidateKeys("https://www.163.com/dy/article/KRPRE8DJ0556MCR0.html", ""),
}
if score := scoreMonitoringAliasCandidate(input, candidate); score != -1 {
t.Fatalf("scoreMonitoringAliasCandidate() = %d, want no same-domain match", score)
}
}