袜子类京东直播带货违法监测预警数据
收藏资源简介：
对采集的京东平台带货品类为袜子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：防臭，抗菌，5A防臭袜，3A防臭袜，脚气，脚汗，脚臭，吸湿排汗，吸汗，透气等）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京东直播行为，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 live streaming video content of sock product promotion influencers on JD.com platform. We process and analyze the verbal content delivered by influencers during live broadcasts, and issue warnings or take corresponding measures in accordance with triggering condition rules, based on the count and frequency of violations of pre-configured sensitive violation keywords (e.g., anti-odor, antibacterial, 5A anti-odor socks, 3A anti-odor socks, beriberi, foot sweat, foot odor, moisture wicking, sweat absorption, breathable, etc.) during the live broadcast. This work provides data support for standardizing JD.com live streaming behaviors of enterprises within the jurisdiction of the Market Supervision Administration of Zhuji City. The collected live streaming videos undergo the following five preprocessing steps: 1. Slice the original video files into segments with a maximum duration of 10 minutes per segment. 2. Convert the sliced video content into audio files. 3. Convert the audio content from the sliced videos into text. 4. Extract text from images captured in the original videos using Optical Character Recognition (OCR) technology. 5. Match the obtained text content against the pre-built illegal warning keyword database. Subsequently, a multi-criteria decision analysis (MCDA) model is employed to analyze and calculate the violating statements made by the streamer during the live broadcast, so as to derive the final illegal warning value and the judgment result of whether to trigger a warning. The formula for calculating the illegal warning value is specified as: Illegal warning value = (Number of hits for single illegal warning keyword * 0.25) + (Number of hits for combined illegal 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) The warning triggering rule is as follows: No warning prompt is triggered when the final calculated illegal warning value is ≤ 1, and an illegal warning prompt is triggered when the value is > 1.
