药食同源类小红书直播带货违法监测预警数据

对采集的小红书平台带货品类为药食同源类的达人直播视频内容进行转译分析，对达人口播语言内容进行处理、分析，根据达人在直播过程中对预先设置的违规敏感词（比如：彻底根治、绝对安全、无任何副作用、快速见效、立竿见影、包治百病、100%纯天然、预防疾病、抗癌、抗衰老、政府推荐等）违反的次数和频率，依据触发条件规则提出警告或处理。为服务辖区市场监督局管理区域内规范企业小红书直播行为，提供数据支持。将采集完成的直播视频进行进行预处理，第一步：基于原始视频文件，以最大10分钟单位对原始视频进行切片。第二步：对于已完成的切片视频，进行视频内容转语音操作。第三步：对于已完成视频转语音操作的切片，进行语音转文本操作。第四步：使用OCR技术对原始视频中抓取的图片进行文字提取操作。第五步：将所得到的文字内容与违法预警关键词库进行匹配。最终运用多标准决策分析模型，对主播在直播过程中出现的违规语句进行分析计算，得出违法预警值和是否预警判断。
违法预警值=（违法预警单关键词命中次数*0.25）+（违法预警组合关键词命中次数* 0.3）+（图片识别命中预警组合关键词个数*0.35）+（直播间近一个月历史违规记录数*0.1）
通过公式计算出最终违法预警值，违法预警值 ≤1 时，不触发预警提示，违法预警值 ＞1 时触发违法预警提示。

This dataset involves transcription and analysis of livestream video content from influencers on the Xiaohongshu (Little Red Book) platform, focusing on the medicine-food homology product promotion category. We first process and analyze the oral spoken content of the influencers during their livestreams, then issue warnings or take corresponding enforcement actions based on the triggering rules, using the frequency and count of violations of pre-set sensitive violation keywords (such as "completely cure", "absolutely safe", "no side effects whatsoever", "quick effect", "instant effect", "cure all diseases", "100% pure natural", "disease prevention", "anti-cancer", "anti-aging", "government recommended", etc.) made by the influencers during their broadcasts. This dataset provides data support for the local market supervision bureau to standardize the livestream behaviors of enterprises within its jurisdiction.

Preprocessing of the collected livestream videos is conducted as follows:
Step 1: Slice the original video files into segments with a maximum duration of 10 minutes each.
Step 2: Convert the sliced video files into audio content.
Step 3: Transcribe the audio from the sliced videos into text.
Step 4: Extract text from images captured in the original videos using OCR technology.
Step 5: Match the obtained text content against the violation warning keyword database.

Finally, a multi-criteria decision analysis model is applied to analyze and calculate the violation statements made by the streamer during the livestream, so as to obtain the violation warning value and the judgment of whether to trigger a warning.
The violation warning value is calculated via the following formula:
Violation Warning Value = (Number of hits for individual violation warning keywords * 0.25) + (Number of hits for combined violation warning keywords * 0.3) + (Number of combined warning keywords identified via image recognition * 0.35) + (Number of historical violation records of the livestream room in the past month * 0.1)

If the final calculated violation warning value is ≤ 1, no warning prompt will be triggered; if the value exceeds 1, a violation warning prompt will be triggered.

搜集汇总

数据集介绍

特点

该数据集用于监测小红书平台上药食同源类直播带货中的违法行为，通过分析直播内容中的关键词匹配和违法预警值计算，提供预警支持。数据集规模为1030条，每季度更新，服务于市场监督局的监管需求。