This paper explores the perfect fit relationship between mediatization and living landscape, grasps the rules of online social interaction of the elderly, redefines the mediatized social scene, and explains the new behaviors arising from the new scene, i.e., the elderly’s behaviors of interacting with the mediatized society. Using cosine similarity, the angle between query vectors and document vectors is calculated, and a probabilistic retrieval model is constructed based on BM25 for portraying the social attributes of the elderly, and deriving global, behavioral, public, and private factors in the process of mediated social interactions of the elderly. Through the BM25 model, the social media information of the elderly, such as date, APP name, start time of use, end time of use, and GPS location information, is retrieved. The retrieval results show that the elderly use the two APPs, Zhihu and iReader, less frequently, with the length of use of 1103.594 and 796.452 respectively, and the frequency of use is only 203 and 98 times. As for the three age groups of 60-70 years old, 71-75 years old and 76-80 years old, the number of elderly people who use the cell phone medium for 2.1-4 hours per day on average is the highest, which is 65, 90 and 30 respectively, accounting for 45.14%, 47.12% and 47.62% respectively.