Research On Algorithmic Discrimination In Government Governance

The Internet provides opportunities for the combination of algorithms and big data.The extensive application of algorithms makes data fully used in the process of government fine governance,and promotes the reform process of social production and life style and social relations.As the core of the algorithm,the personalized algorithm recommendation based on different algorithm models saves energy cost for algorithm users,the "information cocoon" brought by personalized recommendation increases trade barriers,and the updated and iterative algorithm technology also brings risks and challenges for the government to participate in social governance.The discriminatory results brought by algorithm alienation have brought new challenges to government governance.Whether the algorithm is fair not only affects the fairness of the output of algorithm results,but also affects the interests of the public.Therefore,attention should be paid to the algorithm discrimination in the algorithm field,and the development of the algorithm should be standardized to seek the increase of public welfare.The discriminatory consequences of algorithm alienation will not only damage the interests of non-specific groups,specific personal interests and public interests,but also increase the difficulty of identifying algorithm discrimination.To some extent,it will also damage government credibility,market order and public value.Therefore,government governance algorithm discrimination is imperative.Algorithm as a rapidly developing emerging technology,its unique technical advantages and characteristics provide a new way to solve difficult problems,we should combine the basic national conditions of Our country with the current situation of government governance algorithm discrimination,establish a dynamic governance system dominated by government governance algorithm discrimination,flexible and rigid coexist.According to the characteristics of the algorithm’s own data application,the algorithm data is cleaned from the source,and the "sensitive" elements in the algorithm data are removed to ensure the fairness of the input and output of the algorithm data.Reconstructing the core value of the algorithm,correcting the unfair design value,ensuring the fairness of the algorithm design,and enhancing the interpretability of the algorithm.At the same time,improve the legal system of algorithm governance to enhance the applicable scene of law,implement the subject responsibility to ensure the accountability of algorithm discrimination,establish dynamic evaluation and review mechanism of algorithm discrimination to identify the risk of algorithm discrimination.We should attach importance to diversified regulation of algorithmic discrimination and set up unified regulatory bodies and rules.Improve the public’s algorithm literacy,strengthen the industry ethical self-discipline,the sense of responsibility throughout the algorithm.