On the Information Cocoon Effect in the TikTok Recommendation Algorithm

Although the commercial improvement of the recommendation algorithm has led to the rapid rise of short video platforms represented by TikTok, it has also contributed to the social problem of information cocoons. The recommendation algorithm for business services has the characteristics of customization, circle and entertainment, and in the process of operation, the algorithm gradually shows an extreme tendency-personal customization closed loop, group polarization and entertainment to death. In this regard, this paper proposes to establish a people-oriented human-computer interaction mechanism, optimize the display mechanism of the comment area, and rationally allocate the proportion of recommended content types to cope with the alienation caused by the current TikTok recommendation algorithm due to lack of value.