Abstract: Social relationship chain obtaining method and terminal and a readable medium are provided. The method includes: obtaining a target user set and mutual-friend information between any two target users in the target user set, target users in the target user set being friends of a center user, and the mutual-friend information indicating whether the two target users are in a friend relationship; constructing a reference network according to the center user and the mutual-friend information between the any two target users in the target user set, nodes in the reference network being the target users in the target user set, and there being a connection line between two target users in the reference network that are in a friend relationship; and determining, from the target user set according to the reference network, a target user being in a parent-child relationship chain with the center user.

Abstract: A contact management method and apparatus are described. The method includes: calculating a score of interaction activity between a user and a contact according to an interaction measure, interaction frequency, a time attenuation factor, and a preset weight of each interaction type of the user and the contact, and a simultaneous satisfaction degree of the interaction types; identifying a relation chain type of the user and the contact, and querying a preset score table to obtain a score of the relation chain type; calculating a score of relational closeness between the user and the contact; and managing the contact according to the score of the relational closeness.

Abstract: Embodiments of the present invention provide a method, system and server for delivering an advertisement based on user characteristic information. The method includes performing, by a server, data mining on user raw data corresponding to a client to obtain user characteristic information, generating a characteristic label based on the user characteristic information; determining, by the server, a type of advertisement according to the characteristic label, and delivering an advertisement of the type to the client. In the embodiments of the present invention, a great amount of user raw data is collected and stored in a server, data mining is performed on the user raw data, a characteristic label is generated based on user characteristic information obtained, and a network advertisement is delivered according to the characteristic label. Therefore, the advertisement can be delivered to particular users and the click ratio of the advertisement is increased.