饲料类淘宝直播带货违法监测预警数据
收藏资源简介：
对采集的淘宝平台带货品类为饲料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：绝对安全，100%有效，保证增重，无副作用，提高治愈率，有效率高达，无效退款，保险公司保险，比药品更有效，优于其他保健品，催肥，速肥猛长等）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淘宝直播行为，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 performs transcription and analysis on collected live streaming video content of Taobao influencers promoting feed products. The spoken language content of the influencers is processed and analyzed, and warnings or corresponding enforcement actions are issued based on the frequency and count of violations against pre-set sensitive violation keywords (such as: absolutely safe, 100% effective, guaranteed weight gain, no side effects, improved cure rate, effective rate as high as, money-back guarantee if ineffective, covered by insurance companies, more effective than medications, superior to other health supplements, fattening, rapid weight gain, etc.) in accordance with trigger condition rules. This dataset provides data support for standardizing Taobao live streaming operations of enterprises within the administrative jurisdiction of Zhuji Municipal Market Supervision Administration. The collected live streaming videos undergo the following preprocessing steps: 1. Segment the original video files into clips with a maximum duration of 10 minutes each; 2. Convert the content of each segmented video clip to audio; 3. Conduct speech-to-text transcription on the converted audio from the clips; 4. Extract text from images captured from the original videos using Optical Character Recognition (OCR) technology; 5. Match the obtained text content against the violation warning keyword database. Finally, a multi-criteria decision analysis (MCDA) model is utilized to analyze and calculate the violating statements made by live streamers during the live streaming session, to derive the Violation Warning Score and make a warning trigger decision. The formula for calculating the Violation Warning Score is as follows: Violation Warning Score = (Number of hits for individual violation warning keywords * 0.25) + (Number of hits for combined violation warning keywords * 0.3) + (Number of combined warning keywords identified via image recognition * 0.35) + (Number of historical violation records of the live broadcast room in the past month * 0.1) According to the calculation results: no warning prompt is triggered when the Violation Warning Score ≤ 1; a violation warning prompt is triggered when the Violation Warning Score > 1.
