# US 2017/0255646 A1 · Method and Apparatus of Recommending Candidate Terms Based on Geographical Location

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

- Number: US 2017/0255646 A1
- Document ID: US20170255646A1
- Status: Published application
- Kind: A1
- Application number: 15/601,695
- Filed: 2017-05-22
- Published: 2017-09-07
- Earliest claimed: 2012-05-29
- Assignees: Alibaba Group Holding Limited
- Applicants: Alibaba Group Holding Limited
- Inventors: Maojian Fu
- Web page: https://patentgrep.com/patent/US20170255646A1

## CPC

- G06F 16/29: Geographical information databases
- G06F 16/248: Presentation of query results
- G06F 16/9537: Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
- G06F 40/274: Converting codes to words; Guess-ahead of partial word inputs
- H04M 1/72457: according to geographic location
- H04W 4/029: Location-based management or tracking services
- H04M 2250/10: including a GPS signal receiver
- H04M 2250/70: methods for entering alphabetical characters, e.g. multi-tap or dictionary disambiguation
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- H04M: TELEPHONIC COMMUNICATION (circuits for controlling other apparatus via a telephone cable and not involving telephone switching apparatus G08)
- H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72)

## Abstract

The disclosure provides a method and device for recommending a candidate word according to a geographic position. The method may include receiving a coded character string of a user by a computing device. The computing device may collect geographic position information corresponding to the coded character string, and then determine a geographic area in which the geographic position information is located. The computing device may obtain a geographic candidate word corresponding to the coded character string according to a geographic word stock of the determined geographic area. The geographic word stock of the geographic area may store the coded character strings and a corresponding geographic word according to the geographic area. As compared to current technologies, complexity of input can be reduced and intelligence of an input method can be improved.

## Claims

### Claim at position 1

1-20. (canceled)

### Claim 21

A method, comprising:
receiving a sequence of at least one letter from a user device;
determining location information associated with the user device;
mapping the received sequence of at least one letter and the location information with a database of multiple terms, a term of the multiple terms being related to a stored sequence of at least one letter and a context location based on a historical correlation between the term and the stored sequence of at least one letter in the context location; and
providing a candidate term out of the multiple terms as corresponding to the received sequence of at least one letter on the mapping.

### Claim 22 (depends on 21)

The method of claim 21, wherein the context location includes a range of geographic location points.

### Claim 23 (depends on 21)

The method of claim 21, wherein the providing the candidate term includes causing to display the candidate term in a manner configured to receive a user feedback on the candidate term.

### Claim 24 (depends on 21)

The method of claim 21, further comprising updating the database of multiple terms based on a user feedback on the provided candidate term.

### Claim 25 (depends on 21)

The method of claim 21, wherein the database includes multiple terms as related to the stored sequence of at least one letter and the context location, each of the multiple terms being assigned a utilization score in the context location.

### Claim 26 (depends on 21)

The method of claim 21, wherein the receiving the sequence of at least one letter from the user device includes receiving a first sequence of at least one letter and a second sequence of at least one letter, and wherein the second sequence includes the first sequence.

### Claim 27 (depends on 26)

The method of claim 26, wherein the providing the candidate term based on the mapping includes providing a first candidate term based on mapping the first sequence with the database and providing a second candidate term based on mapping the second sequence with the database.

### Claim 28 (depends on 21)

The method of claim 21, wherein the received sequence of at least one letter includes a word abbreviation.

### Claim 29 (depends on 21)

The method of claim 21, wherein the received sequence of at least one letter is in a first form of a language and the candidate term is in a second form of a language.

### Claim 30

A machine learning method, comprising:
receiving, from a user device, a sequence of at least one letter;
determining location information associated with the user device;
mapping the received sequence of at least one letter and the location information with a database of multiple terms, the database relating a term to a stored sequence of at least one letter and a location range based on a historical correlation between the term and the stored sequence of at least one letter in the location range;
providing a candidate term out of the multiple terms as corresponding to the received sequence of at least one letter based on the mapping;
receiving a user feedback on the candidate term; and
updating the database based on the user feedback.

### Claim 31 (depends on 30)

The machine learning method of claim 30, wherein the mapping includes determining that the location information is within a location range related to the candidate term.

### Claim 32 (depends on 30)

The machine learning method of claim 30, wherein updating includes updating a utilization score of the candidate term as related to the received sequence of at least one letter in a location range that encloses the location information.

### Claim 33 (depends on 30)

The machine learning method of claim 30, wherein the database includes multiple terms as related to the stored sequence of at least one letter and the location range, each of the multiple terms being assigned a utilization score in the location range.

### Claim 34 (depends on 30)

The machine learning method of claim 30, wherein the received sequence of at least one letter includes a word abbreviation.

### Claim 35 (depends on 30)

The machine learning method of claim 30, wherein the received sequence of at least one letter is in a first form of a language and the candidate term is in a second form of a language.

### Claim 36 (depends on 30)

The machine learning method of claim 30, wherein the database includes a geographic word stock related to a location range, and the geographic word stock is constructed by:
determining a utilization frequency of a geographic word in the location range;
determining that the utilization frequency of the geographic word is greater than a threshold value; and
storing the geographic word as corresponding to a corresponding stored sequence of at least one letter in the database.

### Claim 37 (depends on 36)

the machine learning method of claim 36, wherein the utilization frequency of the geographic word in the location range is determined in relation to the corresponding stored sequence of at least one letter.

### Claim 38

A method, comprising:
providing a server including a database of multiple terms, the database relating a term to a stored sequence of at least one letter and a geographic area based on a historical correlation between the term and the stored sequence of at least one letter in the geographic area; and
providing a set of executable instructions configured to be installed in a user device, which, when installed in the user device and executed by a processing unit of the user device, are operable to configure the processing unit to perform operations including:
detecting a user input into the user device;
acquiring a sequence of at least one letter from the detected user input;
transmitting the acquired sequence of at least one letter to the server for the server to map the acquired sequence of at least one letter with the database of the multiple terms based on location information of the user device to obtain a candidate term;
receiving the candidate term from the server; and
displaying the candidate term as corresponding to the user input.

### Claim 39 (depends on 38)

The method of claim 38, wherein the detecting the user input includes detecting a partial input of an on-going user input.

### Claim 40 (depends on 38)

The method of claim 38, wherein the acquired sequence of at least one letter is in a first form of a language and the candidate term is in a second form of a language
