From 67daea2f52f1536b4ca5e3a838e607c451639d36 Mon Sep 17 00:00:00 2001 From: MythEclipse Date: Wed, 3 Jun 2026 20:15:24 +0700 Subject: [PATCH] fix(analytics): filter out common file extensions from topic words Excludes image, video, audio, and document file extensions from the topic word extraction process to improve the quality of generated analytics topics. --- .../src/modules/analytics/analytics.repository.ts | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/services/backend/src/modules/analytics/analytics.repository.ts b/services/backend/src/modules/analytics/analytics.repository.ts index bb23a8f..93b02bb 100644 --- a/services/backend/src/modules/analytics/analytics.repository.ts +++ b/services/backend/src/modules/analytics/analytics.repository.ts @@ -361,6 +361,15 @@ export class AnalyticsRepository { 'mana','cuma','kah','udh','gitu','gini','gtu','gni', 'mending','wkakak','wakak' ) + AND word NOT LIKE '%.jpg' + AND word NOT LIKE '%.jpeg' + AND word NOT LIKE '%.png' + AND word NOT LIKE '%.gif' + AND word NOT LIKE '%.webp' + AND word NOT LIKE '%.mp4' + AND word NOT LIKE '%.mp3' + AND word NOT LIKE '%.pdf' + AND word NOT LIKE '%.zip' GROUP BY word ORDER BY COUNT(*) DESC LIMIT 10