保健食品类小红书直播带货违法监测预警数据

Name: 保健食品类小红书直播带货违法监测预警数据
Creator: 诸暨市市场监督管理局,浙江富润数链科技有限公司
Published: 2024-06-26T10:10:50+08:00

浙江省数据知识产权登记平台2024-07-18 更新2024-07-19 收录

直播监控

违法预警

数据链接：

https://www.zjip.org.cn/home/announce/trends/39280 数据链接链接失效反馈

官方服务：

资源简介：

对采集的小红书平台带货品类为保健食品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：无效退款，保险公司保险，安全，安全无毒副作用，毒副作用小，天然，热销，抢购，试用，家庭必备等）违反的次数和频率，依据触发条件规则提出警告或处理。为诸暨市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 is constructed for the transcription and analysis of influencer live streaming video content focused on health food products from the Xiaohongshu (Little Red Book) platform. The work involves processing and analyzing the verbal content narrated by influencers during live broadcasts, and issuing warnings or taking regulatory actions in accordance with preset trigger rules, based on the count and frequency of violations of pre-defined sensitive violation-related keywords, including but not limited to "invalid refund", "insured by insurance company", "safe", "safe and non-toxic side effects", "low toxic and side effects", "natural", "hot-selling", "rush purchase", "trial use", "household essential". This dataset provides data support for the Zhuji Municipal Market Supervision Administration to standardize the live streaming behaviors of enterprises within its administrative jurisdiction. The processing pipeline for collected live streaming videos is as follows: 1. Slice the original video files into segments with a maximum duration of 10 minutes; 2. Convert the sliced video content into audio; 3. Perform speech-to-text conversion on the extracted audio; 4. Extract text from images captured in the original videos using OCR technology; 5. Match all obtained text content against the pre-built violation warning keyword database. Finally, a multi-criteria decision analysis model is utilized to analyze and compute the violation statements made by the live stream host, to generate the final violation warning score and determine whether a warning should be triggered. The violation warning score is calculated via the following formula: Violation Warning Score = (Number of hits for single violation warning keywords × 0.25) + (Number of hits for combined violation warning keywords × 0.3) + (Number of combined violation warning keywords identified via image recognition × 0.35) + (Number of historical violation records of the live stream host in the past month × 0.1) According to the calculation result: no warning prompt is triggered when the violation warning score ≤ 1, and a violation warning prompt is triggered when the violation warning score > 1.

提供机构：

诸暨市市场监督管理局,浙江富润数链科技有限公司

创建时间：

2024-06-26

原始信息汇总