Cluster-Based Friend Suggestion Aggregator
Techniques are described to generate a list of suggested friends for a client of a social networking service. In an implementation, a cluster of clients is identified from among the clients of the social networking service. The cluster of clients includes the client for which the list of suggested friends is to be generated. The list of suggested friends for the client may then be generated and may include one or more clients of the social networking service selected based on connections with the cluster of clients. The list indicates at least one other client of the social networking service that is suggested as a friend.
1 . A computer implemented method comprising:
selecting a client of a social networking service; and
generating a list of suggested friends for the selected client based on interactions of the selected client with other clients of the social networking service via the social networking service.
2 . A method as recited in claim 1 , wherein the interactions via the social networking service include sending or receiving emails.
3 . A method as recited in claim 1 , wherein the interactions via the social networking service include sending or receiving instant messages.
4 . A method as recited in claim 1 , wherein the interactions via the social networking service include posting comments.
5 . A method as recited in claim 1 , wherein the interactions include interactions described in data received from one or more data providers.
6 . A method as recited in claim 1 , wherein the generating includes ranking the one or more clients in the list
7 . A method as recited in claim 6 , wherein the ranking is based on the connections of the one or more clients with a cluster of clients to generate a ranked suggestion list of the one or more clients, the list of suggested friends being based on the ranked suggestion list.
8 . A method as recited in claim 6 , wherein the ranking is based on one or more weighting factors.
9 . A method as recited in claim 8 , wherein at least one said weighting factor is adjusted based on past acceptance or rejection of friends suggested to the first client by the data provider.
10 . A method as recited in claim 1 , wherein the selected client is part of a cluster of clients that were identified based on whether the clients have connections, one to another.
11 . One or more computer-readable storage media comprising instructions that are executable to implement a social networking service that is configured to generate a list of suggested friends for a selected client of the social networking service based on interactions of the selected client with one or more other clients of the social network service, the interactions described by data received from one or more data providers.
12 . One or more computer readable media as recited in claim 11 , wherein the one or more friend suggestions are weighted with at least one weighting factor.
13 . One or more computer readable media as recited in claim 11 , wherein the one or more other clients are ranked based on the connections of the one or more other clients with a cluster of clients to generate a ranked suggestion list, the list of suggested friends being based on the ranked suggestion list.
14 . One or more computer readable media as recited in claim 13 , wherein the cluster of clients is identified by selecting clients of the plurality of clients to be in the cluster of clients based on whether the clients have connections to the cluster of clients.
15 . A system implemented by one or more computing devices, the system configured to perform operations comprising:
identifying a cluster of clients from a plurality of clients of a social networking service; and
generating a list of suggested friends for a selected client from the cluster based on interactions of the clients to communicate, one with another.
16 . A system as recited in claim 15 , wherein the interactions include sending or receiving emails.
17 . A system as recited in claim 15 , wherein the interactions include sending or receiving instant messages.
18 . A system as recited in claim 15 , wherein the interactions include posting comments.
19 . A system as recited in claim 15 , wherein the interactions are described in data received from one or more data providers.
20 . A system as recited in claim 15 , wherein the generating is performed using a weighting factor.