基于算法推荐模式的社会性反思:个体困境、群体极化与媒体公共性

Social Reflection based on Algorithm Recommendation Mode:Individual Dilemma,Group Polarization and Media Publicity

互联网信息过载,催生了算法推荐信息分发模式的产生与发展。然而,算法推荐的实践应用,日渐凸显了一些不容忽视的社会性问题:对于个体而言,算法推荐在个体选择性接触和媒体“流量为王”思维的共同作用下,信息分发的“千人千面”,演变为个体信息获取的“信息茧房”困境;对于群体而言,算法推荐对个体的“信息茧房”,加剧了群体成员间“千人一面”的“回音室”效应,并与网络群体中“判断型话题”、同质性群体结构、网络传播的匿名性等“群体极化”互动影响因素共同作用,产生“群体极化”现象,进而消解算法时代的“媒体公共性”。优化现有算法推荐模式,是突破其社会困境的可行之路:首先,赋予算法推荐正确的价值观,在满足用户兴趣的前提下,优先推荐正能量的优质信息内容;其次,将“编辑算法”与“推荐算法”相结合,实行兼具信息分发多样化与个性化的混合推荐算法,将媒体公共性重建与个性化需求满足有机结合。

Information overload in the era of mobile internet has given birth to the generation and development of algorithm recommendation information distribution mode.However,the practical application of algorithm recommendation increasingly highlights some social problems that cannot be ignored:for individuals,under the joint action of individual selective contact and media“traffic is king”thinking,the“thousands of people and thousands of faces”of information distribution has evolved into the dilemma of“information cocoons”for individual information acquisition;for groups,the“information cocoons”for individuals intensifies the“echo chamber”effect among group members.Interacting with“judgment topic”,homogeneous group structure,and anonymity of network communication,dynamic factors work together to produce the phenomenon of“group polarization”,and then eliminate the“media publicity”in the era of algorithm.Optimizing the existing algorithm recommendation mode is a feasible way to break through its social dilemma:First,give the algorithm the correct values,and give priority to recommending positive energy information content on the premise of meeting the user s interest.Secondly,combining“editing algorithm”with“recommendation algorithm”to implement a hybrid recommendation algorithm with both diversification and personalization of information distribution.The reconstruction of publicity and the satisfaction of individual needs are organically combined.