Social sketches

• US 10,147,107 B2
• Filed: 06/26/2015
• Issued: 12/04/2018
• Est. Priority Date: 06/26/2015
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Alonso, Omar, Tremblay, Serge-Eric, Kandylas, Vasileios
• Primary Examiner(s): Miller, Alan S
• Application Number:
  US14/752,781
• Publication Number:
  US 20160379232A1
• Time in Patent Office: 1,257 Days
• Field of Search: 705 711- 742
• CPC Class Codes: G06F 16/447 Temporal browsing, e.g. timeline
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 10/00 Administration; Management
• CPC Class Codes: G06Q 30/0204 Market segmentation
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

Systems and methods for generating a social sketch from social communications are provided. Unlike a typical summary of a subject (or of multiple subjects), a social sketch represents a “snapshot” summary of the social communications of people during a given period of time. The social sketch is generated by obtaining a corpus of social communications and filtering the social communications according to time. The filtered results are clustered according to the subject matter/topics of the social communications. Selected clusters are identified and the topic, representative high-quality social communications from non-experts and experts are extracted from each of the selected clusters and saved as a social sketch corresponding to the time period.

Claims

1. A method for generating a social sketch corresponding to a time period in response to a request, the method implemented by at least one processor having executable instructions that, when executed, perform operations comprising:

obtaining social communications from a social networking service, the social communications including social communications generated during a first time period;

filtering obtained social communications according to the first time period such that the filtered social communications correspond to the social communications generated during the first time period;

clustering the filtered social communications according to the subject matter of the social communications to generate a plurality of clusters of filtered social communications, each cluster corresponding to a topic;

identifying a set of clusters of the plurality of clusters of social communications comprising identifying those clusters of the plurality of clusters of social communications that have a sufficient volume of social communications within the cluster, each cluster of the set of the clusters being an identified cluster, and wherein the plurality of clusters of social communications that have a sufficient volume of social communications within the cluster comprises a predetermined number of clusters that have the greatest number of social communications of the plurality of clusters;

for each identified cluster;

extracting a topic from the identified cluster according to the subject matter of the social communications of the identified cluster;

identifying a non-expert set of high-quality communications from the identified cluster, the non-expert set of high-quality communications corresponding to social communications of non-experts on the topic of the identified cluster;

identifying an expert set of high-quality communications from the identified cluster, the expert set of high-quality communications corresponding to social communications of experts on the topic of the identified cluster;

wherein the topic, the non-expert set of high-quality communications, and the expert set of high-quality communications comprise a cluster set of the identified cluster; and

storing the cluster sets of each of the identified clusters as the social sketch corresponding to the identified time period.

2. The computer implemented method of claim 1 further comprising, for each of the identified clusters:

re-clustering the identified cluster of social communications;

identifying a set of sub-clusters of the identified cluster; and

extracting a sub-topic from each of the identified sub-clusters of the set of sub-clusters of the identified cluster;

wherein the topic, the non-expert set of high-quality communications, the expert set of high-quality communications, and the extracted sub-topics comprise a topic set of the identified cluster.

3. The computer implemented method of claim 2 further comprising, for each of the identified clusters:

identifying a representative image of the identified cluster from the social communications of the identified cluster;

wherein the topic, the non-expert set of high-quality communications, the expert set of high-quality communications, the extracted sub topics, and the representative image comprise a topic set of the identified cluster.

4. The computer implemented method of claim 3, wherein the plurality of clusters of social communications that have a sufficient volume of social communications within the cluster comprises those clusters whose volume of social communications within the cluster exceeds a predetermined number.

5. The computer implemented method of claim 3, wherein the plurality of clusters of social communications that have a sufficient volume of social communications within the cluster comprises those clusters whose volume of social communications within the cluster exceeds a predetermined percentage of all of the social communications for the time period.

6. The computer implemented method of claim 3, wherein identifying a set of sub clusters of the identified cluster comprises identifying the set of sub clusters according to any one or more of:

a threshold volume of social communications within the sub cluster,

a threshold percentage of the volume of social communications of the cluster within the sub cluster, or

a threshold number of sub-clusters that have the greatest volume of social communications within the sub cluster.

7. The computer implemented method of claim 2 further comprising:

receiving a search query regarding the time period;

identifying a set of search results relevant to the time period, the set of search results including results of topics not found with the social sketch corresponding to the time period;

obtaining the social sketch corresponding to the time period;

generating a search results page, the search results page including;

at least some of the identified set of results relevant to the time period; and

a social sketch view, the social sketch view including a plurality of user actionable controls, each user actionable control identifying and corresponding to a social sketch topic of the social sketch corresponding to the time period; and

returning the generated search results page to a requesting party in response to the search query.

8. The computer implemented method of claim 2 further comprising:

receiving a search query regarding a first topic;

identifying a set of search results relevant to the first topic;

determining whether the first topic corresponds to a topic of the social sketch and, upon determining that the first topic corresponds to a topic of the social sketch;

generating a search results page, the search results page including;

some of the identified set of results relevant to the time period; and

a timeline information block, the timeline information block including at least one user actionable control, the at least one user actionable control comprising a snippet of information representative of the first topic as occurred during the time period of the social sketch; and

returning the generated search results page to a requesting party in response to the search query.

9. The computer implemented method of claim 2 further comprising:

receiving a search query regarding a first topic and the time period, wherein the first topic corresponds to a topic of the social sketch;

identifying a set of search results relevant to the first topic;

generating a search results page, the search results page including;

at least some of the identified set of results relevant to the first topic; and

a social sketch view, the social sketch view comprising the topic of the social sketch, at least some of the non-expert set of high quality communications from the identified cluster, and at least some of the expert set of high quality communications for the identified cluster; and

returning the generated search results page to a requesting party in response to the search query.

10. The computer implemented method of claim 9, wherein the social sketch view further comprises a user actionable control linking the current social sketch view to a social sketch view corresponding to a social sketch of another time period that includes the first topic, wherein the user actionable control linking the current social sketch view to a social sketch view corresponding to a social sketch of another time period is configured to cause to replace the social sketch view corresponding to the first topic of a social sketch of the time period with a social sketch view corresponding to the first topic of a social sketch of the another time period.

11. A computing device configured to generate a social sketch corresponding to a time period, the computing device comprising a processor and a memory, and further comprises additional components for generating the social sketch, the additional components comprising:

a social sketch generator configured to generate a social sketch from social communications obtained from one or more social networking services;

a clustering module configured to cluster a corpus of social communications according to similarity; and

a social sketch data store;

wherein, in operation, the social sketch generator;

obtains a corpus of social communications from one or more social networking services over a network via a network communication component, the corpus of social communications including social communications generated during a first time period;

filters the obtained corpus of social communications according to the first time period such that the filtered social communications correspond to the social communications of the corpus of social communications generated during the first time period;

clusters, by way of the clustering module, the filtered social communications according to the subject matter of the social communications to generate a plurality of clusters of filtered social communications;

identifies a set of clusters of the plurality of clusters of social communications comprising identifying those clusters of the plurality of clusters of social communications that have a sufficient volume of social communications within the cluster, wherein the plurality of clusters of social communications that have a sufficient volume of social communications within the cluster comprises a predetermined number of clusters that have the greatest number of social communications of the plurality of clusters, each cluster of the set of the clusters being an identified cluster;

for each identified cluster;

extracts a topic from the identified cluster according to the subject matter of the social communications of the identified cluster;

identifies a representative image of the identified cluster from the social communications of the identified cluster;

identifies a non-expert set of high-quality communications from the identified cluster, the non-expert set of high-quality communications corresponding to social communications of non-experts on the topic of the identified cluster;

identifies an expert set of high-quality communications from the identified cluster, the expert set of high-quality communications corresponding to social communications of experts on the topic of the identified cluster;

re-clusters, via the clustering module, the identified cluster of social communications;

identifies a set of sub-clusters of the identified cluster; and

extracts a sub-topic from each of the identified sub-clusters of the set of sub-clusters of the identified cluster;

wherein the topic, the non-expert set of high-quality communications, the expert set of high-quality communications, the extracted sub-topics, and the representative image comprise a cluster set of the identified cluster; and

stores the cluster sets of each of the identified clusters as the social sketch corresponding to the identified time period in the social sketch data store.

12. The computing device of claim 11, wherein the plurality of clusters of social communications that have a sufficient volume of social communications within the cluster comprises those clusters whose volume of social communications within the cluster exceeds a predetermined number.

13. The computing device of claim 11, wherein the plurality of clusters of social communications that have a sufficient volume of social communications within the cluster comprises those clusters whose volume of social communications within the cluster exceeds a predetermined percentage of all of the social communications for the time period.

14. The computing device of claim 11, wherein identifying a set of sub clusters of the identified cluster comprises identifying the set of sub clusters according to any one or more of:

a threshold volume of social communications within the sub cluster,

a threshold percentage of the volume of social communications of the cluster within the sub cluster, or

a threshold number of sub-clusters that have the greatest volume of social communications within the sub cluster.

15. A method for providing a targeted social sketch in response to a request, the method implemented by at least one processor having executable instructions that, when executed, perform operations comprising:

receiving a request for a targeted social sketch corresponding to a first topic and a time period;

accessing social communications from one or more social networking services;

filtering the social communications according to the time period, the results being social communications generated during the time period;

clustering the social communications according to topics as found in the subject matter of the social communications;

identifying a cluster of social communications corresponding to the first topic from a plurality of clusters of social communications having a sufficient volume of social communication within the cluster, and wherein the plurality of clusters of social communications that have a sufficient volume of social communications within the cluster comprises a predetermined number of clusters that have the greatest number of social communications of the plurality of clusters;

identifying a non-expert set of high-quality communications from the identified cluster, the non-expert set of high-quality communications corresponding to social communications of non-experts on the first topic of the identified cluster;

identifying an expert set of high-quality communications from the identified cluster, the expert set of high-quality communications corresponding to social communications of experts on the first topic of the identified cluster;

storing the targeted social sketch, wherein the targeted social sketch comprises the first topic, the non-expert set of high-quality communications, and the expert set of high-quality communications;

generating a targeted social sketch presentation according to the targeted social sketch; and

returning the targeted social sketch presentation to a requesting party in response to the request.

16. The computer implemented method of claim 15 further comprising:

re-clustering the identified cluster of social communications;

identifying a set of sub-clusters of the identified cluster; and

extracting a sub-topic from each of the identified sub-clusters of the set of sub-clusters of the identified cluster;

wherein the targeted social sketch comprises the first topic, the non-expert set of high-quality communications, the expert set of high-quality communications, and the extracted sub-topics.