Suggested keywords

• US 10,042,944 B2
• Filed: 06/18/2014
• Issued: 08/07/2018
• Est. Priority Date: 06/18/2014
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Jain, Aastha, Lytkin, Nikita Igorevych, Gordon, Vitaly, Lau, Gloria
• Primary Examiner(s): Gortayo, Dangelino N
• Application Number:
  US14/308,496
• Publication Number:
  US 20150370805A1
• Time in Patent Office: 1,511 Days
• Field of Search: 707730, 707748, 707728, 707749, 707723, 707732, 707737, 707758, 707784
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/252 between a Database Management System and a front-end application
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/335 Filtering based on additional data, e.g. user or group profiles filtering in web context G06F16/9535, G06F16/9536
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9536 Search customisation based on social or collaborative filtering
• CPC Class Codes: G06F 16/958 Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

A suggested keywords system is configured for identifying phrases, which are most relevant to experience and expertise of a professional network member, and which the member may be interested in weaving into their profile summary. The suggested keywords system generates a model, for each phrase, that calculates probability of that phrase being present in a profile that is characterized by the absence of certain attributes and by the presence of certain attributes. Based on the model, the suggested keywords system calculates a ranking value for the phrase for a particular target profile. The phrases with the higher rank are considered to be more relevant in describing professional background of the target member. A certain number of phrases that have the highest ranking are presented to the member as suggested keywords to be included in their professional summary.

Claims

1. A method comprising:

accessing a target profile from profiles in an on-line social network system, the on-line social networking system maintaining a set of different attributes, an attribute from the set of different attributes corresponds to a characteristic of a member represented by a member profile from the profiles in the on-line social network system;

generating, for the target profile, a profile feature vector, the profile feature vector has a number of dimensions equal to the number of items in the set of different attributes, each dimension corresponding to a respective attribute from the set of different attributes, each feature in the profile feature vector being either 1 or 0 to indicate a presence or an absence of an associated attribute in the target profile;

extracting a plurality of phrases from the profiles in the on-line social network system;

calculating, using at least one processor, for phrases from the plurality of phrases, respective ranking values based on the profile feature vector, a ranking value from the respective ranking values expresses probability of the phrase being included in the target profile; and

including a phrase selected from the plurality of phrases based on its ranking value into a profile summary user interface generated for a member represented by the target profile.

2. The method of claim 1, comprising:

calculating respective weights for each attribute from the plurality of attributes with respect to a phrase from the plurality of phrases, a weight from the respective weights for a particular attribute being indicative of the likelihood of the phrase being present in a profile that includes the particular attribute;

from the profile feature vector, determining attributes from the plurality of attributes that are present in the target profile; and

combining respective weights of the attributes that are present in the target profile to generate a ranking value for the n-gram with respect to the target profile.

3. The method of claim 2, wherein the combining of the respective weights of the attributes that are present in the target profile comprises applying a scoring function to the respective weights.

4. The method of claim 1, comprising identifying suggested keywords for a summary section of the target profile, based on respective ranking values of the phrases from the plurality of phrases.

5. The method of claim 4, wherein the identifying of the suggested keywords for the summary section of the target profile is in response to detecting an access request to a webpage comprising a summary section of the target profile.

6. The method of claim 1, comprising storing one or more phrases from the plurality of phrases and their respective ranking values for future access.

7. The method of claim 1, comprising:

selecting a predetermined number of phrases from the plurality of phrases, based on their respective ranking values; and

presenting the selected phrases to a member represented by the target profile.

8. The method of claim 7, comprising:

determining that the summary section of the target profile contains one or more phrases that are within a certain threshold number of phrases having highest respective ranking values; and

presenting the selected phrases to a member represented by the target profile, excluding the one or more phrases that are present in the summary section of the target profile.

9. The method of claim 7, comprising presenting a notification to the member represented by the target profile, in response to the determining that the summary section of the target profile contains the one or more phrases.

10. The method of claim 1, comprising recalculating, periodically, a ranking value for each phrase in the plurality of phrases.

11. A computer-implemented system comprising:

one or more processors; and

a non-transitory computer readable storage medium comprising instructions that when executed by the one or processors cause the one or more processors to perform operations comprising;

accessing a target profile from profiles in an on-line social network system, the on-line social networking system maintaining a set of different attributes, an attribute from the set of different attributes corresponds to a characteristic of a member represented by a member profile from the profiles in the on-line social network system;

generating for the target profile, a profile feature vector, the profile feature vector has a number of dimensions equal to the number of items in the set of different attributes, each dimension corresponding to a respective attribute from the set of different attributes, each feature in the profile feature vector being either 1 or 0 to indicate of a presence or an absence of an associated attribute in the target profile;

extracting a plurality of phrases from the profiles in the on-line social network system;

calculating for phrases from the plurality of phrases, respective ranking values based on the profile feature vector, a ranking value from the respective ranking values expresses probability of the phrase being included in the target profile; and

including a phrase selected from the plurality of phrases based on its ranking value into a profile summary user interface generated for a member represented by the target profile.

12. The system of claim 11, comprising:

calculating respective weights for each attribute from the plurality of attributes with respect to a phrase from the plurality of phrases, a weight from the respective weights for a particular attribute being indicative of the likelihood of the phrase being present in a profile that includes the particular attribute;

determining, from the profile feature vector, attributes from the plurality of attributes that are present in the target profile, the ranking module is to combine respective weights of the attributes that are present in the target profile to generate a ranking value for the phrase with respect to the target profile.

13. The system of claim 12, comprising applying a scoring function to the respective weights.

14. The system of claim 11, identifying suggested keywords for a summary section of the target profile, based on respective ranking values of the phrases from the plurality of phrases.

15. The system of claim 14, comprising identifying the suggested keywords for the summary section of the target profile in response to detecting an access request to a webpage comprising a summary section of the target profile.

16. The system of claim 11, comprising storing one or more phrases from the plurality of phrases and their respective ranking values for future access.

17. The system of claim 11, comprising:

selecting a predetermined number of phrases from the plurality of phrases, based on their respective ranking values; and

presenting the selected phrases to a member represented by the target profile.

18. The system of claim 17, comprising:

determining that the summary section of the target profile contains one or more phrases that are within a certain threshold number of phrases having highest respective ranking values; and

presenting the selected phrases to a member represented by the target profile, excluding the one or more phrases that are present in the summary section of the target profile.

19. The system of claim 17, comprising presenting a notification to a member represented by the target profile, in response to the determining that the summary section of the target profile contains the one or more phrases.

20. A machine-readable non-transitory storage medium having instruction data to cause a machine to perform operations comprising:

accessing a target profile from profiles in an on-line social network system, the on-line social networking system maintaining a set of different attributes, the an attribute from the set of different attributes corresponds to a characteristic of a member represented by a member profile from the profiles in the on-line social network system;

generating, for the target profile, a profile feature vector, the profile feature vector has a number of dimensions equal to the number of items in the set of different attributes, each dimension corresponding to a respective attribute from the set of different attributes, each feature in the profile feature vector being either 1 or 0 to indicate of a presence or an absence of an associated attribute in the target profile;

extracting a plurality of phrases from the profiles in the on-line social network system;

calculating, for phrases from the plurality of phrases, respective ranking values based on the profile feature vector, a ranking value from the respective ranking values expresses probability of the phrase being included in the target profile; and

including a phrase selected from the plurality of phrases based on its ranking value into a profile summary user interface generated for a member represented by the target profile.