Algorithmic recommendation on social media platforms has become an important mechanism for users' daily information reception and emotional regulation. As social media platforms such as Douyin

Rednote

Weibo

and Bilibili have become an important part of young people's daily lives

the way users access information has gradually shifted from active search to platform algorithmic recommendation. Users' ways of accessing information have gradually shifted from active searching to personalized content distribution dominated by platform algorithms. Algorithm recommendation uses data such as users' likes

duration of stay

search records

viewing history

and comment interactions to determine users' interests and continuously generate personalized information streams. Nevertheless

it is worth noting that many users continue to use the platform even if they are aware that algorithmic recommendations may cause anxiety

irritability

fatigue

or a sense of being induced. Therefore

the core issue that this study focuses on how does algorithmic recommendation shape users' continuous usage behavior through emotional satisfaction

companionship experience

and instant feedback while triggering negative emotions. This study uses the semi-structured in-depth interview method

targeting Chinese young social media users aged 20-30 as the research subjects

and analyzes their perception

emotional experience

and coping strategies regarding algorithmic recommendations when using social media platforms. Research has found that users generally have algorithm awareness and can perceive that the platform pushes content based on their personal behavioral data. On the one hand

algorithmic recommendation may generate negative emotions; on the other hand

it meets users' needs through emotional compensation such as entertainment and relaxation. It is found that the relationship between users and algorithms is not simply one of control and being controlled

but rather a contradictory emotional dependence relationship.