PROACTIVE AND RETROSPECTIVE JOINT WEIGHT ATTRIBUTION IN A STREAMING ENVIRONMENT

• US 20180232434A1
• Filed: 12/22/2017
• Published: 08/16/2018
• Est. Priority Date: 02/16/2017
• Status: Abandoned Application

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Geyik, Sahin Cem, Dialani, Vijay, Gupta, Abhishek
• Application Number:
  US15/852,491
• Publication Number:
  US 20180232434A1
• Time in Patent Office: Days
• CPC Class Codes: G06F 16/3326 using relevance feedback from the user, e.g. relevance feedback on documents, documents sets, document terms or passages
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 10/1053 Employment or hiring

Abstract

Techniques for joint weight attribution for weights of candidate features of a candidate search are described in an example embodiment, disclosed is a system that obtains one or more suggested candidate documents based on a search query specifying one or more parameters. Additionally, the system extracts query intents from the one or more suggested candidate documents, the one or more query intents corresponding to the one or more parameters. Moreover, the system ranks the one or more suggested candidate documents based on the extracted query intents. Furthermore, the system displays top ranked documents on a display device. The system then receives feedback regarding the displayed top ranked documents. Then, weights of a hidden intent are attributed to a set of possible intents based on the received feedback. The feedback can be received retrospectively and proactively. For example, some embodiments perform joint weight attribution based on retrospective and proactive feedback ingestion.

Claims

1. A computer system, comprising:

one or more processors; and

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

obtaining one or more suggested candidate documents based on a search query specifying one or more parameters;

extracting one or more query intents from the one or more suggested candidate documents, the one or more query intents corresponding to the one or more parameters;

ranking the one or more suggested candidate documents based on the extracted one or more query intents;

causing one or more top ranked documents to be displayed on a display device;

receiving feedback regarding the displayed one or more top ranked documents, and

attributing weights of a hidden intent to a set of possible intents based on the received feedback.

2. The system of claim 1, wherein the suggested candidate documents are member profiles in a social networking service.

3. The system of claim 1, wherein the instruction set executable on the processor further cause the computer system to perform operations comprising:

updating the display of the one or more top ranked documents on the display device, the updating being based on the attributing of the weights;

receiving additional feedback regarding the updated display of the one or more top ranked documents; and

repeating the attributing of the weights based on the additional feedback.

4. The system of claim 1 wherein the feedback includes retrospective feedback received from a user interacting with the display of the one or more top ranked documents.

5. The system of claim 4, wherein the attributing of the weights is retrospective weight attribution comprising:

examining the received feedback after a probability of selection of a candidate; and

taking into account aft earlier selection by one arm of a multi-armed bandit (MAB) solution as compared to another arm of the MAB solution.

6. The system of claim 1 wherein the feedback includes proactive feedback.

7. The system of claim 6 wherein the attributing of the weights is proactive weight attribution wherein multiple arms of an MAB solution are improved when a candidate is suggested for the first time the proactive weight attribution comprising:

performing a proactive match of the candidate to a query intent; and

proactively assigning candidates to intents of the one or more query intents.

8. A computer-implemented method, comprising:

obtaining one or more suggested candidate documents based on a search query specifying one or more parameters;

extracting one or more query intents from the one or more suggested candidate documents the one or more query intents corresponding to the one or more parameters;

ranking the one or more suggested candidate documents based on the extracted one or more query intents;

causing one or more top ranked documents to displayed on a display device,

receiving feedback regarding the displayed one or more top ranked documents; and

attributing weights of a hidden intent to a set of possible intents based on the received feedback.

9. The method of claim 8, wherein the suggested candidate documents are member profiles in a social networking service.

10. The method of claim 8, further comprising.

updating the display of the one or more top ranked documents on the display device, the updating being based on the attributing of the weights;

receiving additional feedback regarding the updated display of the one or more top ranked documents, and

repeating the attributing of the weights based on the additional feedback.

11. The method of claim 8, wherein the feedback includes retrospective feedback received from a user interacting with the display of the one or more top ranked documents.

12. The method of claim 11, wherein the attributing of the weights is retrospective weight attribution comprising:

examining the received feedback after a probability of selection of a candidate; and

taking into account an earlier selection by one arm of a multi-armed bandit (MAB) solution as compared to another arm of the MAB solution.

13. The method of claim 8, wherein the feedback includes proactive feedback.

14. The method of claim 13, wherein the attributing of the weights is proactive weight attribution wherein multiple arms of an MAB solution are improved when a candidate is suggested for the first time, the proactive weight attribution comprising:

performing a proactive match of the candidate to a query intent; and

proactively assigning candidates to intents of the one or more query intents.

15. A non-transitory machine-readable storage medium comprising instructions, which when executed by one or more machines, cause the one or more machines to perform operations comprising:

obtaining one or more suggested candidate documents based on a search query specifying one or more parameters;

extracting one or more query intents from the one or more suggested candidate documents, the one or more query intents corresponding to the one or more parameters;

ranking the one or more suggested candidate documents based on the extracted one or more query intents;

causing one or more top ranked documents to be displayed on a display device;

receiving feedback regarding the displayed one or more top ranked documents; and

attributing weights of a hidden intent to a set of possible intents based on the received feedback.

16. The non-transitory machine-readable storage medium of claim 15, wherein the suggested candidate documents are member profiles in a social networking service.

17. The non-transitory machine-readable storage medium of claim 15, wherein the operations further comprise:

updating the display of the one or more top ranked documents on the display device, the updating being based on the attributing of the weights;

receiving additional feedback regarding the updated display of the one or more top ranked documents, and

repeating the attributing of the weights based on the additional feedback.

18. The non-transitory machine-readable storage medium of claim 15, wherein:

the feedback includes retrospective feedback received from a user interacting with the display of the one or more top ranked documents, and

the attributing of the weights is retrospective weight attribution comprising;

examining the received feedback after a probability of selection of a candidate; and

taking into account an earlier selection by one arm of a multi-armed bandit (MAB) solution as compared to another arm of the MAB solution.

19. The non-transitory machine-readable storage medium of claim 15, wherein the feedback includes proactive feedback.

20. The non-transitory machine-readable storage medium of claim 19, wherein the attributing of the weights is proactive weight attribution wherein multiple arms of an MAB solution are improved when a candidate is suggested for the first time, the proactive weight attribution comprising:

performing a proactive match of the candidate to a query intent; and

proactively assigning candidates to interns of the one or more query intents.