药食同源类抖音直播带货违法监测预警数据

对采集的抖音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抖音直播行为，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 ≤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 performs transcription and analysis on the live streaming video content of influencers promoting medicinal and edible homologous products on the Douyin platform. It processes and analyzes the spoken content of these influencers, and issues warnings or takes corresponding regulatory actions according to the trigger condition rules, based on the frequency and count of violations against pre-set sensitive violation keywords (e.g., thoroughly cure, absolutely safe, no side effects whatsoever, rapid effect, immediate effect, cure all diseases, 100% pure natural, prevent diseases, anti-cancer, anti-aging, government recommendation, etc.) during live broadcasts. This work provides data support for local market supervision bureaus to standardize the live streaming behaviors of enterprises operating on the Douyin platform within their jurisdictions.

The collected live streaming videos undergo the following preprocessing workflow:
Step 1: Segment the original video files into chunks with a maximum duration of 10 minutes.
Step 2: Convert the content of each segmented video into speech.
Step 3: Convert the generated speech content from the segmented videos into textual format.
Step 4: Extract text from images captured from the original videos using OCR technology.
Step 5: Match all collected text content against the pre-built violation warning keyword database.

Finally, a multi-criteria decision analysis (MCDA) model is employed to analyze and calculate the violating statements made by the live streamers during their broadcasts, to derive the violation warning score and make a warning determination. If the violation warning score is ≤ 1, no warning prompt will be triggered; if the score exceeds 1, a violation warning prompt will be activated.