药品类拼多多直播带货违法监测预警数据

对采集的拼多多平台带货品类为药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，免费治疗等）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拼多多直播行为，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 designed for the transcription and analysis of live streaming video content from drug-category product promotion influencers on the Pinduoduo platform. The spoken content of these influencers is processed and analyzed, and warnings or corresponding regulatory actions are issued based on the frequency and count of violations against pre-configured sensitive prohibited keywords (including but not limited to: invalid refund, insurance covered by insurance companies, safety, safe with no toxic and side effects, low toxic and side effects, natural, hot sale, rush purchase, trial use, household necessity, free treatment, etc.), in accordance with the preset trigger rules. This work provides data support for the Market Supervision Administration of Zhuji City to regulate the live streaming behaviors of enterprises within its jurisdiction on the Pinduoduo platform.

The collected live streaming videos undergo the following standardized preprocessing procedures:
1. Slice the original video files into segments with a maximum duration of 10 minutes;
2. Convert the content of each sliced video segment into audio via video-to-speech (VTS) conversion;
3. Transcribe the converted audio content into text via speech-to-text (STT) conversion;
4. Extract text from images extracted from the original video using Optical Character Recognition (OCR) technology;
5. Match all obtained text content against the pre-built prohibited warning keyword database.

Finally, a multi-criteria decision analysis (MCDA) model is employed to analyze and calculate the violative utterances made by the live streamer during the broadcast, to derive the final violation warning score and make a judgment on whether to trigger a warning alert.
The violation warning score is calculated using the following weighted formula:
Violation Warning Score = (Number of hits for individual prohibited keywords × 0.25) + (Number of hits for combined prohibited keywords × 0.3) + (Number of combined warning keywords detected via OCR-based image recognition × 0.35) + (Number of historical violation records of the live stream room in the past 30 days × 0.1)
After calculating the final violation warning score, no warning alert will be activated if the score is less than or equal to 1, and a warning alert will be activated if the score exceeds 1.

搜集汇总

数据集介绍

特点

该数据集用于监测拼多多药品类直播带货中的违法行为，包含1899条记录，每季度更新，通过分析直播内容中的违规敏感词，为市场监督局提供数据支持。算法规则涉及视频处理和多标准决策分析，计算违法预警值以判断是否触发预警。