酒类美团直播带货违法监测预警数据
收藏资源简介：
对采集的美团平台带货品类为酒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：不宿醉，喝不醉，不上头，醒酒快，温肾，除积冷，消除紧张、消除焦虑，增加体力，改善疲劳感，再喝一杯，千杯不醉等）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美团直播行为，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 focuses on transcription and analysis of influencer live stream video content for alcohol-related product categories on the Meituan platform. The verbal speech content of influencers during live broadcasts is processed and analyzed, and warnings or corresponding treatments are proposed based on trigger rules, according to the frequency and count of violations of pre-set sensitive violation keywords by influencers during live streaming. The pre-set sensitive violation keywords include, but are not limited to: avoiding hangovers, not getting drunk, not having a heavy head, fast hangover relief, warming the kidneys, removing accumulated cold, relieving tension and anxiety, boosting physical stamina, alleviating fatigue, have another drink, never getting drunk no matter how much one drinks, etc. This work provides data support for standardizing the live streaming behaviors of enterprises under the jurisdiction of the Zhuji Municipal Market Supervision Administration via the Meituan platform. The collected live stream videos undergo the following preprocessing steps: 1. Slice the original video files into segments with a maximum duration of 10 minutes; 2. Convert the sliced video content into audio; 3. Convert the audio from the sliced videos into text; 4. Extract text from images captured from the original videos using OCR technology; 5. Match the obtained text content against the violation warning keyword database. Finally, a multi-criteria decision analysis model is used to analyze and calculate the violation statements made by the streamer during the live broadcast, to derive the final violation warning score and the judgment of whether to trigger a warning. The formula for calculating the violation warning score is: Violation Warning Score = (Number of hits for single violation warning keyword * 0.25) + (Number of hits for combined violation warning keywords * 0.3) + (Number of combined violation warning keywords detected via image recognition * 0.35) + (Number of historical violation records of the live stream room in the past month * 0.1) If the final violation warning score ≤ 1, no warning prompt is triggered; if the score > 1, a violation warning prompt is triggered.
