From 9d4881a01ce0d369d3f9abb27cadb3edd241fe35 Mon Sep 17 00:00:00 2001
From: hub-gif <2487812171@qq.com>
Date: Thu, 16 Apr 2026 14:21:32 +0800
Subject: [PATCH] =?UTF-8?q?fix(report):=20=E6=89=A9=E5=85=85=E6=AD=A3?=
 =?UTF-8?q?=E5=90=91=E5=8C=85=E8=A3=85=E7=9F=AD=E8=AF=AD=E5=B9=B6=E6=BE=84?=
 =?UTF-8?q?=E6=B8=85=E5=8F=A3=E8=AF=AD=E6=9D=A1=E5=BD=A2=E5=9B=BE=E5=8F=A3?=
 =?UTF-8?q?=E5=BE=84?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Made-with: Cursor
---
 .../crawler_copy/jd_pc_search/jd_competitor_report.py | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/backend/crawler_copy/jd_pc_search/jd_competitor_report.py b/backend/crawler_copy/jd_pc_search/jd_competitor_report.py
index e0686f0..b74634f 100644
--- a/backend/crawler_copy/jd_pc_search/jd_competitor_report.py
+++ b/backend/crawler_copy/jd_pc_search/jd_competitor_report.py
@@ -631,6 +631,15 @@ _POS_LEXEME_DETAIL = (
     "物流很快",
     "包装很用心",
     "包装完好",
+    # 包装/物流等维度：若未命中下列句式，条形图仍可能偏低；关注词「包装」等反映提及频率
+    "包装不错",
+    "包装很好",
+    "独立包装",
+    "小包装很方便",
+    "包装精美",
+    "密封性好",
+    "包装严实",
+    "快递很快",
     "口感很好",
     "味道不错",
     "很好吃",
@@ -768,6 +777,8 @@ def _comment_sentiment_lexicon(texts: list[str]) -> dict[str, Any]:
             "「正向短语」仅在命中正向词表的评价条内统计（含混合条）；"
             "「负向短语」仅在命中负向词表的评价条内统计（含混合条）；每条每短语最多计 1 次；"
             "统计的是预设口语片段，非分词模型。"
+            "条形图表示的是「预设短语命中条数」，不是全文情感或某维度（如包装、物流）的完整满意度；"
+            "若用户用「盒子不错」「没压坏」等未收录说法，仍可能落在关注词「包装」子串与语义池原文中，条形图偏低不代表该维度无人提及。"
             "预设表无法覆盖全部说法（如「一袋就一点点」），条形图条数低不代表用户未在别处表述同类不满，须结合语义池原文。"
         ),
     }