By using Internet media such as websites, web pages, and Internet applications, Internet advertising has become the main driving force for the sustainable development and innovation of the overall advertising industry by virtue of its strong interactive, situational, and accurate communication effects.
However, due to the virtual space characteristics of the network, in recent years, there have been false advertisements, misleading advertisements and infringing advertisements in some parts of the Internet advertising market. The marketing of various content such as "planting grass" and "exploring shops" has not been marked according to regulations. The "viral" communication chain has made the traditional advertising supervision method relying on manual retrieval more difficult. The above chaos not only infringes consumers' right to know, but also challenges the supervision of Internet advertising.
It is worth noting that the current Internet advertising has realized real-time update and accurate delivery with the help of artificial intelligence (AI) technologies such as algorithm technology and natural language processing, and has begun to transform to intelligence. Accordingly, the massive processing capacity and uninterrupted operation efficiency of AI technology are highly adaptable to the needs of Internet advertising monitoring, and can play an important role in the identification, identification, punishment and other important regulatory links of illegal advertising. Therefore, in order to achieve efficient, scientific and reasonable Internet advertising supervision, the traditional manual management mode needs to be further upgraded. Strengthening the application of artificial intelligence technology in the field of Internet advertising supervision is one of the paths.
In terms of advertising recognition, big data technology can be used to build an advertising library, and cloud computing technology can be used to calculate and monitor advertisements. At present, there are three main ways of Internet advertising monitoring, namely, dynamic IP pool simulation browsing detection, modeling detection and large-scale parallel computing detection. They have been relatively mature in dealing with the more conventional types of linked advertising, search engine advertising and rich media advertising, but for the more novel types of "soft advertising", such as "grass planting, shop exploration, live marketing", the monitoring methods and bottom algorithms are not yet mature, and still rely on manual review to a large extent, reducing the monitoring efficiency. Given that the "soft advertising" type has developed for several years and accumulated a large amount of data, the feasibility of using big data technology to further build a new illegal advertising library has been enhanced. Therefore, the regulatory authorities can establish a database in accordance with the Advertising Law of the People's Republic of China, Internet Advertising Management Measures and other laws and regulations, model, compare and identify illegal advertisements.

In terms of advertising recognition, distributed technology can be used to collect and evaluate advertisements. At present, Internet advertising has basically realized the whole process online, meeting the preconditions for the use of blockchain technology. Relevant departments can promote the placement of advertising transaction processes in the blockchain, and record advertising data and other information on the chain. Its tamper proof and traceable characteristics will improve the accuracy and efficiency of advertising identification. It should be noted that Internet advertising supervision belongs to administrative law enforcement. At present, artificial intelligence technology is not mature in the identification of illegal advertising, so the illegal advertising detected by technology still needs to be manually identified again.

In terms of penalties for illegal advertising, the use of artificial intelligence technology can improve accuracy. The punishment for traditional advertising is usually determined based on advertising duration, overall advertising costs, etc., and the fine amount is determined according to the fine provisions of the Advertising Law of the People's Republic of China for each illegal subject. However, there are magnitude differences between Internet advertising and traditional advertising in terms of delivery efficiency and influence. Therefore, we can rely on artificial intelligence technology to calculate the click rate, conversion rate and other data of advertising, and conduct accurate measurement based on communication data and influence surface to provide reliable basis for punishment determination.
The development of Internet advertising market has formed a huge system so far. Advertisers, publishers and consumers are closely related. While strengthening supervision, attention should also be paid to the standardized governance of the industry.   (New Society)