基于深度学习的现场促销AI稽核方法的研究

针对通信运营商社会渠道网点在申报现场促销活动酬金补贴时,需要投入大量人力对促销场次真实性、促销照片规范性等进行人工稽核的问题,文中通过对现场促销位置、时间、工号进行多维度的合理性自动审核,基于深度学习技术开发现场促销AI稽核模型。该模型能够高效稽核促销真实性,使稽核效率提升80%,有效降低了套利风险,节约了31.14%的促销包干酬金支出。

Research on AI Audit Method of On-site Promotion Based on Deep Learning

In view of the problem that the social channel outlets of communication operators need to invest a lot of manpower to manually audit the authenticity of promotion sessions, the standardization of promotion photos and other issues when applying for on-site promotion subsidy, this paper develops an AI audit model for on-site promotion based on in-depth learning technology through multi-dimensional automatic verification of the rationality of on-site promotion location, time, and job number, and effectively audit the authenticity of promotion, the audit efficiency was improved by 80%, effectively reducing arbitrage risk, and saving 31.14% of the promotion lump sum remuneration.