Algorithm 'preferences' cannot be arbitrary anymore. Recently, the Secretariat of the Cyberspace Administration of China, the General Office of the Ministry of Industry and Information Technology, the General Office of the Ministry of Public Security, and the General Office of the State Administration for Market Regulation jointly issued a notice on carrying out the "Clear Network Platform Algorithm Typical Problem Governance" special action, targeting typical network platform algorithm problems such as big data "killing", "information cocoons", and "manipulating rankings". This has important and far-reaching significance for maintaining the clarity of cyberspace and protecting the legitimate rights and interests of netizens. In the era of information explosion, algorithm recommendation should have been a powerful tool to improve information acquisition efficiency and meet users' personalized needs. However, some negative phenomena have gradually surfaced: the emergence of "information cocoons" allows users to only access homogeneous content, narrowing their horizons; The manipulation of rankings by the navy has disrupted the public's access to information and affected the normal ecology of online public opinion; And big data 'killing' has seriously damaged consumers' right to fair trade. The existence of these problems not only affects the user experience, but also disrupts the fair competition order in the market, weakens the healthy vitality of the online ecosystem, and even triggers a crisis of social trust to a certain extent. This special action is another round of rectification and standardization aimed at addressing the noise and noise in the process of algorithm development, based on previous governance efforts. It presents many highlights, including a focus on precision. Clearly address the most prominent and strongly reflected typical problems in current network platform algorithms, so that governance can be targeted and avoid vague discussions. Secondly, the measures are effective. For example, it is strictly prohibited to use algorithms to implement big data "killing", and it is proposed to comprehensively use various means, including regulating the price behavior of operators within the platform, supervising the platform to improve the order allocation mechanism, etc., to carry out rectification through multiple measures. Thirdly, focus on collaboration. The joint action of the four departments, integrating resources and strengths from all parties, is conducive to breaking down regulatory barriers, forming a regulatory synergy, and enhancing the authority and effectiveness of governance. From the expected results, the special action is expected to quickly curb the spread of algorithmic chaos in the short term, effectively restrain various violations, and purify the cyberspace. In the long run, it will reshape the order of online information dissemination, promote the standardized, scientific, and humanized development of network platform algorithms, and further enhance people's sense of gain, happiness, and security in the online world. Algorithms are tools, and their value depends on the goals and methods of use. As users and implementers of algorithms, platforms should take practical and effective corrective measures as soon as possible. On the one hand, they should strictly follow the notification requirements for self-examination and self correction, carefully investigate problems and vulnerabilities, and abandon algorithm models that harm user rights and interests; On the other hand, we will strengthen the construction of algorithm transparency, publicize the basic principles, purpose and intention, and operation mechanism of algorithm recommendation services in an appropriate way, so that users can understand consumption and use them with confidence. The relevant government departments should also actively take action, establish strict algorithm review standards and regulatory regulations, supervise the entire process of algorithm development and application, and promptly discover and correct deviations. In addition, users themselves should also enhance their awareness of prevention and rights protection. When encountering problems such as search results being deliberately guided by algorithms, serious deviations in recommended content, and big data "killing", they should be brave enough to complain and report to relevant parties. At the same time, excessive reliance on a single platform should be avoided. By comparing multiple platforms, the risk of being bound by algorithms and misleading consumer decisions can be reduced, forcing platform algorithms to break away from path dependence and build new development models. The ultimate subject of algorithms is humans, and meeting the public's demand for a fair, just, and secure network ecosystem is the starting point of algorithm governance. We look forward to the demonstration effect of the special action and its further transformation into a long-term mechanism, promoting platform algorithms to be more standardized, fair, and transparent, truly becoming a booster for promoting the progress of network civilization and the harmonious development of society. (New Society)