奶粉类阿里巴巴直播带货违法监测预警数据

对采集的阿里巴巴平台带货品类为奶粉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：母乳、增加抵抗力、促进脑部发育、益智、预防感冒、治疗、强化体质、增强免疫力等）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阿里巴巴直播行为，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 conducts transcription and analysis on the live streaming video content of influencers promoting milk powder products collected from the Alibaba platform. We process and analyze the verbal content spoken by the influencers, and issue warnings or take corresponding actions based on trigger rules according to the frequency and count of violations of pre-set sensitive violation words such as "breast milk", "increase resistance", "promote brain development", "enhance intelligence", "prevent colds", "treat diseases", "strengthen physique", "boost immunity" and so on. The dataset aims to provide data support for the local Market Supervision Bureau to regulate the live streaming behaviors of enterprises within its jurisdiction on the Alibaba platform.

The preprocessing steps for the collected live streaming videos are as follows:
1. Slice the original video files into segments with a maximum duration of 10 minutes;
2. Convert the sliced videos into audio content;
3. Transcribe the audio from the sliced videos into text;
4. Extract text from images captured in the original videos using OCR technology;
5. Match the obtained text and image-extracted 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 live broadcast, to derive the final violation warning value and the judgment of whether to trigger a warning.

The formula for calculating the violation warning value is:
Violation Warning Value = (Number of hits for single violation warning keyword * 0.25) + (Number of hits for combined violation warning keywords * 0.3) + (Number of combined warning keyword hits identified via image recognition * 0.35) + (Number of historical violation records of the live room in the past month * 0.1)

According to the calculation rules: when the violation warning value ≤ 1, no warning prompt is triggered; when the violation warning value > 1, a violation warning prompt is triggered.

搜集汇总

数据集介绍

特点

该数据集包含1167条奶粉类阿里巴巴直播带货的违法监测预警数据，每季度更新一次。数据集通过分析直播视频内容中的违规敏感词，如母乳、增加抵抗力等，计算违法预警值，为市场监督局提供规范企业直播行为的数据支持。