Modeling user behaviors is crucial for many web applications. In addition, it is also essential to distinguish the behaviors of different users because of low precision of sensing individual users. In this talk, I will introduce two of my works about modeling and distinguishing user behaviors in web search and online streaming services. In the first study, we use context-aware query suggestion as an example to demonstrate the benefits of understanding user behaviors for web search applications [1]. More specifically, we model how a user reformulates queries in a search session and then precisely infer the next reformulation as well as the next query. Because many users tend to share accounts in online streaming services, the recommendation systems would be disgraced by the uncertainty about real users. In the second study, we aim to distinguish user behaviors under each shared account, thereby improving the recommendation performance [2]. [1] Jyun-Yu Jiang and Wei Wang. Rin: Reformulation inference network for context-aware query suggestion. In Proceedings of the 27th ACM International Conference on Information and Knowledge Management (CIKM 18), pages 197-206. ACM, 2018. [2] Jyun-Yu Jiang, Cheng-Te Li, Yian Chen, and Wei Wang. Identifying users behind shared accounts in online streaming services. In The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval (SIGIR 18), pages 65-74. ACM, 2018.