# US 2017/0255879 A1 · Searching method and device based on artificial intelligence

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2017/0255879 A1
- Document ID: US20170255879A1
- Status: Published application
- Kind: A1
- Application number: 15/392,017
- Filed: 2016-12-28
- Published: 2017-09-07
- Earliest claimed: 2016-03-01
- Assignees: Beijing Baidu Netcom Science Technology Co., Ltd.
- Applicants: Beijing Baidu Netcom Science And Technology Co., Ltd.
- Inventors: Li CHEN; Qian XU; Hao TIAN; Jingzhou HE; Lei SHI; Fan WANG; Shiwei HUANG; Derong ZHENG
- Web page: https://patentgrep.com/patent/US20170255879A1

## CPC

- G06F 16/90335: {Query processing}
- G06F 16/904: Browsing; Visualisation therefor (for navigating the web G06F16/954; browsing optimisation for the web G06F16/957)
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- G06N 20/00: Machine learning
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A searching method and device based on artificial intelligence is provided in the present disclosure. The searching method includes: obtaining a query; obtaining a first search result corresponding to the query according to a Markov Decision Process MDP model; displaying the first search result; and obtaining a reward for the first search result from a user so as to obtain a second search result according to the MDP model, and displaying the second search result. According to the searching method, the interaction with the user may be more effective, the user's demand is better satisfied, and the user experience is improved.

## Claims

### Claim 1

A searching method based on artificial intelligence, comprising:
obtaining a query;
obtaining a first search result corresponding to the query according to a Markov Decision Process (MDP) model;
displaying the first search result; and
obtaining a reward for the first search result from a user so as to obtain a second search result according to the MDP model, and displaying the second search result.

### Claim 2 (depends on 1)

The searching method according to claim 1, wherein, parameters of the MDP model comprise:
a state, represented by the query and a context;
an action, represented by the first search result; and
a reward, represented by the reward for the first search result.

### Claim 3 (depends on 1)

The searching method according to claim 1, wherein, the query comprises:
a query inputted by the user initially, a query recommended to the user, or a switched query inputted by the user.

### Claim 4 (depends on 1)

The searching method according to claim 1, wherein, the first search result comprises:
a webpage result, and a query recommended to the user.

### Claim 5 (depends on 4)

The searching method according to claim 4, wherein, the reward comprises one or more of:
clicking the webpage result by the user;
clicking the query recommended to the user by the user;
the switched query inputted by the user;
a clicking and buying action of the user;
a clicking duration; and
a searching duration.

### Claim 6

A searching device based on artificial intelligence, comprising:
one or more computing devices configured to execute one or more software modules, the one or more software module comprising:
an obtaining module, configured to obtain a query;
a calculating module, configured to obtain a first search result corresponding to the query according to a Markov Decision Process (MDP) model;
a displaying module, configured to display the first search result; and
a reward module, configured to obtain a reward for the first search result from a user, such that a second search result is obtained according to the MDP model and the second search result is displayed.

### Claim 7

The searching device according to claim, 6, wherein, parameters of the MDP model used in the calculating module comprise:
a state, represented by the query and a context;
an action, represented by the first search result; and
a reward, represented by the reward for the first search result.

### Claim 8 (depends on 6)

The searching device according to claim 6, wherein the query comprises:
a query inputted by the user initially, a query recommended to the user, or a switched query inputted by the user.

### Claim 9 (depends on 6)

The searching device according to claim 6, wherein, the first search result comprises:
a webpage result, and a query recommended to the user.

### Claim 10 (depends on 9)

The searching device according to claim 9, wherein, the reward comprises one or more of:
clicking the webpage result by the user;
clicking the query recommended to the user by the user;
the switched query inputted by the user;
a clicking and buying action of the user;
a clicking duration; and
a searching duration.

### Claim 11

A non-transitory computer readable storage medium having stored therein instructions that, when executed by a processor of a terminal, causes the terminal to perform a searching method based on artificial intelligence, the searching method comprising:
obtaining a query;
obtaining a first search result corresponding to the query according to a Markov Decision Process (MDP) model;
displaying the first search result; and
obtaining a reward for the first search result from a user so as to obtain a second search result according to the MDP model, and displaying the second search result.
