Question distribution method and a question distribution system for a Q and A platform

Patent metadata

Field	Value
Publication number	US-10339188-B2
Application number	US-201414579329-A
Country	US
Kind code	B2
Filing date	Dec 22, 2014
Priority date	Dec 20, 2013
Publication date	Jul 2, 2019
Grant date	Jul 2, 2019

Abstract

The present invention provides a question distribution method and a question distribution system for a Q & A platform, wherein the method comprises: obtaining a question entering into a Q & A platform; determining a user suitable for answering the question based on a question distribution model, and distributing the question to a third party platform where the user suitable for answering the question is located, wherein the question distributing model is established with user data collected from the third party platform interfaced with the Q & A platform; obtaining an answer to the question from the third party platform, and providing the answer to the question to a user posing the question. The present invention extends the source of answering a question to a third party platform interfaced with the Q & A platform, which breaks the limitation of the original Q & A platform, so that a user more suitable for answering the question in the third party platform can provide an answer, thereby enhancing the probability of the question being timely answered.

First claim

The invention claimed is:

1. A question distribution method for a question and answer (Q & A) platform, the method comprising:
obtaining a question entering into a Q & A platform through a first interface associated with the Q & A platform;
displaying the question on the first interface;
determining a user suitable for answering the question based on a question distribution model;
distributing the question to a third party platform where the user suitable for answering the question is located, wherein the third party platform interfaces with the Q & A platform through an application programming interface (API) of the Q & A platform, and the question distributing model is established by using user data collected from the third party platform interfaced with the Q & A platform, wherein the third party platform is different from the Q & A platform, and the third party platform is modified according to API of the Q & A platform to interface with the Q & A platform;
displaying the question on a second interface associated with the third party platform;
obtaining an answer to the question from the third party platform through the second interface; and
displaying the answer to the question on the first interface,
wherein the question distribution model comprises a platform model, and the determining a user suitable for answering the question based on a question distribution model comprises:
extracting features of respective types from the question based on features on respective types in the platform model, the platform model being built based on multi-type features;
mapping the extracted features to the features on the respective types in the platform model;
calculating, with weight values of respective features, score values of answers of respective platforms to the question; and
selecting a platform having a score value satisfying a preset condition from the platforms.

2. The method according to claim 1 , wherein the obtaining a question entering into a Q & A platform comprises:
obtaining the question generated by a user of the Q & A platform; or,
receiving the question pushed from the third party platform interfaced with the Q & A platform; or,
actively pulling the question from the third party platform interfaced with the Q & A platform.

3. The method according to claim 1 , wherein the method further comprises:
collecting user data of the third party platform interfaced with the Q & A platform; and
establishing the question distribution model by using the collected user data.

4. The method according to claim 3 , wherein, before establishing the question distribution model with the collected user data, the method further comprises: performing code conversion to the collected user data, the code conversion specifically comprises:
performing code conversion to the collected user data, thereby obtaining user data in a uniform code format;
performing normalization processing to code-converted user data; and
performing data structure standardization to the normalized user data.

5. The method according to claim 1 , wherein the question distribution model comprises a user interest model, wherein;
the user interest model is built based on at least one of following multi-type features and respective weight values thereof: a user's activity feature, a user's interest keyword feature, a user's interest tag feature, a user's interest classification feature, a user's geographical location feature, and a feature of platform originating the user; and
the platform model is built based on at least one of following multi-type features and respective weight values thereof: a keyword feature at which the platform is good, a tag feature at which the platform is good, a classification feature at which the platform is good, a geographical location feature at which the platform is good, an authority feature of the platform, and a platform activity feature.

6. The method according to claim 5 , wherein the determining a user suitable for answering the question based on a question distribution model comprises:
determining, utilizing the platform model, the third party platform suitable for answering the question, and determining a user in the determined third party platform as a user suitable for answering the question; or,
determining, utilizing the platform model, a third party platform suitable for answering the question, and then determining, utilizing the user interest model, the user suitable for answering the question in the third party platform suitable for answering the question; or
determining a third party designated by the user as the third party platform suitable for answering the question, and then determining, utilizing the user interest model, the user suitable for answering the question in the third party platform suitable for answering the question; or
determining, utilizing an extended user interest model derived from merging the platform model and the user interest model, the user suitable for answering the question.

7. The method according to claim 1 , wherein, after distributing the question to a third party platform where the user suitable for answering the question is located, the method further comprises:
pushing, by the third party platform, the question to the user suitable for answering the question in a message sending manner specific to the third party platform; or,
providing a questioning page for the question on the Q & A platform on the third party platform through a Q & A platform application embedded in the third party platform.

8. The method according to claim 2 , wherein the providing the answer to the question to a user posing the question specifically comprises:
if the question is generated by the user of the Q & A platform, pushing the answer to the question to the user posing the question using at least one of an SMS, an email, and an internal site message; or,
if the question is originated from the third party platform interfaced with the Q & A platform, providing the answer to the question to the third party platform originating the question, and pushing, by the third party platform, the answer to the question to the user posing the question in a manner specific to the third party platform.

9. A question distribution system for a question and answer (Q & A) platform, the system comprising:
at least one processor; and
a memory storing instructions, wherein the instructions when executed by the at least one processor, cause the at least one processor to perform operations, the operations comprising:
providing an application programming interface (API) and perform an interaction between the Q & A platform and the third party platform;
obtaining a question entering into a Q & A platform through a first interface associated with the Q & A platform, the first interface displaying the entered question;
determining a user suitable for answering the question based on a question distribution model, and distributing the question to a third party platform where the user suitable for answering the question is located, wherein the question distributing model is established with user data collected from the third party platform interfaced with the Q & A platform; and
obtaining an answer to the question from the third party platform through a second interface associated with the third party platform, the second interface displaying the question, and providing the answer to the question to first interface,
wherein the question distribution model comprises a platform model, and the determining a user suitable for answering the question based on a question distribution model comprises:
extracting features of respective types from the question based on features on respective types in the platform

Assignees

• Baidu online network technology beijing co ltd

Inventors

• Liu Shi
• Yang Fan
• Li Guohong

Classifications

• G06F16/9535 Primary

  Search customisation based on user profiles and personalisation · CPC title

• H04L67/26

  Electricity · mapped topic

• G09B7/02 Primary

  of the type wherein the student is expected to construct an answer to the question which is presented or wherein the machine gives an answer to the question presented by a student · CPC title

• G06Q10/063112

  Skill-based matching of a person or a group to a task · CPC title

• H04L67/55

  Push-based network services · CPC title