CN107229348B

CN107229348B - Input error correction method and device for input error correction

Info

Publication number: CN107229348B
Application number: CN201610169993.4A
Authority: CN
Inventors: 朱洋达; 张扬
Original assignee: Beijing Sogou Technology Development Co Ltd
Current assignee: Beijing Sogou Technology Development Co Ltd
Priority date: 2016-03-23
Filing date: 2016-03-23
Publication date: 2021-11-02
Anticipated expiration: 2036-03-23
Also published as: CN107229348A

Abstract

The embodiment of the invention provides an input error correction method, an input error correction device and a device for input error correction, wherein the method comprises the following steps: acquiring click operation of keys corresponding to characters in an input string and click positions of the click operation in a key area; the key area comprises a plurality of sub-areas, and the click position corresponds to the sub-areas; determining a target adjacent key corresponding to the clicking position; and carrying out error correction processing on the input string according to the target adjacent key. The embodiment of the invention can reduce the number of the target adjacent keys used in error correction, thereby reducing the calculation amount of calculation for the target adjacent keys, reducing the calculation amount required by error correction of the target adjacent keys and further improving the error correction efficiency.

Description

Input error correction method and device for input error correction

Technical Field

The present invention relates to the field of input methods, and in particular, to an input error correction method, an input error correction device, and an input error correction device.

Background

For users in chinese, japanese, korean, etc., it is generally necessary to interact with a computer through an input method system. For example, a user can type an input string through a keyboard, and then the input string is converted into a candidate item of a corresponding language and displayed by the input method system according to a preset standard mapping rule, so that the candidate item selected by the user is displayed on a screen.

In practical applications, the user may press a wrong key or click a wrong location, which may cause a wrong input string to be typed. The input error correction is an input method system solution aiming at the errors, and can actively identify the input errors of the user and correct the input strings with the errors into correct input strings, so that the user can input correct characters on the premise of not modifying the input strings.

Currently, a key error correction scheme is introduced for possible deviation when a user clicks a key. For example, when the user inputs a high speed, or blindly clicks, or when the input environment jolts and shakes, the user may easily mistakenly click on the adjacent key of the target key, and perform an adjacent key attempt on each character of the input string by using a key error correction scheme, for example, if the adjacent key of the "s" key specifically includes "a", "d", "w", "x", etc., the adjacent key attempt may replace "s" in the input string with "a", "d", "w", "x", etc., and determine whether the score of the replaced input string reaches a certain threshold, and if so, it is determined that the erroneous click may occur, so the replaced input string may be adopted.

However, in practical applications, since there are multiple adjacent keys for each key, the existing key error correction scheme performs replacement attempts on the adjacent keys of each key and calculates scores, and numerous calculation processes result in low error correction efficiency.

Disclosure of Invention

In view of the above problems, embodiments of the present invention are provided to provide an input error correction method, apparatus, and apparatus for input error correction that overcome or at least partially solve the above problems, and that can reduce the amount of calculation for calculating a target neighboring key, thereby improving error correction efficiency.

In order to solve the above problems, the present invention discloses an input error correction method, comprising:

acquiring click operation of keys corresponding to characters in an input string and click positions of the click operation in a key area; the key area comprises a plurality of sub-areas, and the click position corresponds to the sub-areas;

determining a target adjacent key corresponding to the clicking position;

and carrying out error correction processing on the input string according to the target adjacent key.

Optionally, the step of determining a target neighboring key corresponding to the click position includes:

determining the offset position of the sub-area corresponding to the click position relative to the center of the key area;

and determining at least one key closest to the offset position as a target adjacent key corresponding to the click position.

Optionally, if the clicked position includes coordinate information, the step of determining the target neighboring key corresponding to the clicked position includes:

determining coordinate information of the click position;

and searching for the target adjacent key corresponding to the coordinate information in a mapping relation between the coordinate information and the target adjacent key established in advance.

determining a target adjacent key corresponding to the click position according to an error correction model; the error correction model is established according to collected historical click information of the user, and the historical click information comprises: and inputting the click position of the key corresponding to the character in the string.

Optionally, the step of performing error correction processing on the input string according to the target adjacent key includes:

determining a mapping probability between the click position and the target adjacent key;

and carrying out error correction processing on the input string according to the mapping probability.

Optionally, the step of determining a mapping probability between the click position and the target neighboring key includes:

searching and obtaining the mapping probability between the click position and the target adjacent key in the mapping relation among the pre-established click position, key and mapping probability; and/or

And determining the mapping probability between the clicking position and the target adjacent key according to an error correction model.

Optionally, the step of performing error correction processing on the input string according to the mapping probability includes:

determining an error correction string corresponding to the input string according to the target adjacent key;

obtaining an error correction string score corresponding to the error correction string according to the mapping probability between the click position and the target adjacent key;

and when the score of the error correction string accords with a preset score condition, carrying out error correction processing on the input string according to the error correction string.

Optionally, the error correction model comprises a first error correction model and/or a second error correction model;

the first error correction model is obtained by training according to collected historical click information of different users;

the second error correction model is obtained by training according to collected historical click information of the same user.

Optionally, the step of collecting historical click information of different and/or same users includes:

counting the number of clicks of a key corresponding to an input character and the click position in the key area corresponding to the input character aiming at the screen-on operation of the candidate item corresponding to the input string; and/or

Counting the number of clicks of keys corresponding to correction characters and the click positions in the key areas corresponding to the correction characters aiming at the corrected screen-on operation of the input string; and/or

And counting the click times of the keys corresponding to the error correction characters and the click positions in the key regions corresponding to the error correction characters aiming at the screen-on operation of the error correction candidate items corresponding to the error correction strings.

In another aspect, the present invention discloses an input error correction apparatus, comprising:

the device comprises an acquisition module, a display module and a display module, wherein the acquisition module is used for acquiring the click operation of a key corresponding to a character in an input string and the click position of the click operation in a key area; the key area comprises a plurality of sub-areas, and the click position corresponds to the sub-areas;

the determining module is used for determining a target adjacent key corresponding to the clicking position; and

and the error correction module is used for carrying out error correction processing on the input string according to the target adjacent key.

In yet another aspect, an apparatus for input error correction is disclosed that includes a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured for execution by the one or more processors to include instructions for:

determining a target adjacent key corresponding to the clicking position;

The embodiment of the invention has the following advantages:

according to the embodiment of the invention, the target adjacent key is determined based on the click position of the key by the user, and the input string is subjected to error correction processing according to the target adjacent key. The key area can be divided into a plurality of sub-areas, and the click position corresponds to the sub-area; because the embodiment of the invention can determine the sub-region corresponding to the click position in the key region, and further determine the target adjacent key corresponding to the sub-region, compared with the existing scheme, the embodiment of the invention can position the click position to a smaller sub-region granularity, so as to exclude some adjacent keys and reduce the range of the target adjacent key;

furthermore, the embodiment of the invention can reduce the number of the target adjacent keys used in error correction, thereby reducing the calculation amount of calculation of the target adjacent keys, reducing the calculation amount of error correction of the target adjacent keys and further improving the error correction efficiency.

Drawings

FIG. 1 is a flowchart illustrating steps of a first embodiment of an input error correction method according to the present invention;

FIG. 2 is a flowchart illustrating steps of a second embodiment of an input error correction method according to the present invention;

FIG. 3 is a flowchart illustrating the steps of a third embodiment of an input error correction method according to the present invention;

FIG. 4 is a flowchart illustrating the fourth step of an input error correction method according to a fourth embodiment of the present invention;

FIG. 5 is a flowchart illustrating steps of a sixth embodiment of an input error correction method according to the present invention;

FIG. 6 is a block diagram of an embodiment of an input error correction apparatus according to the present invention;

FIG. 7 is a block diagram of an apparatus 800 for input error correction in accordance with the present invention;

fig. 8 is a schematic diagram of a server according to the present invention.

Detailed Description

In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.

The existing input error correction method carries out adjacent key pressing attempt on each character of an input string, and needs to spend a large amount of calculation, so that the error correction efficiency is low.

The embodiment of the invention discovers that corresponding input intentions are different when the user clicks different positions in the key area in the input process. For example, if the user clicks the position at the upper left corner of the key "j" and the position is very close to the key "u", it means that the probability that the user originally wants to input the character "u" is high, and when the key "u" is clicked, the user may mistakenly click the position at the upper left corner of the key "j".

Therefore, the embodiment of the invention creatively provides that the target adjacent key is determined based on the clicking position of the key by the user, and the input string is subjected to error correction processing according to the target adjacent key. The key area can be divided into a plurality of sub-areas, and the click position corresponds to the sub-area; because the embodiment of the present invention can determine the sub-region corresponding to the click position in the key region, and further can determine the target adjacent key corresponding to the sub-region, compared with the existing scheme, the embodiment of the present invention locates the click position to a smaller sub-region granularity, so as to exclude some adjacent keys, reduce the range of determining the target adjacent key, and further reduce the number of target adjacent keys used in error correction, thereby reducing the calculation amount for calculating the target adjacent key, reducing the calculation amount required in error correction of the target adjacent key, and finally improving the error correction efficiency.

Method embodiment one

Referring to fig. 1, a flowchart illustrating steps of a first embodiment of an input error correction method according to the present invention is shown, which may specifically include the following steps:

step 101, obtaining a click operation of a key corresponding to a character in an input string and a click position of the click operation in a key area; the button area may specifically include a plurality of sub areas, and the click position corresponds to the sub area;

the embodiment of the invention is suitable for inputting various languages such as Chinese, English, Japanese and the like, and for convenience, the Chinese is taken as an example for description. When inputting Chinese, it is usually necessary to type characters such as letters, numbers, symbols, etc., and the input method program converts the acquired user input characters into Chinese. The embodiment of the invention corrects errors which may occur in the input process of a user, and the related error correction scenes can include, but are not limited to, the following types: when a user inputs by using the keyboard, the user mistakenly clicks on a key adjacent to a target key due to high input speed or touch typing; or when the touch virtual keyboard of an intelligent terminal such as a mobile phone is used for inputting, the situation that the key size of the virtual keyboard is relatively small, the key is mistakenly touched by the adjacent key occurs; or, the finger or the touch pen of the intelligent terminal or the mobile terminal is operated to shake due to walking, vehicle bump, collision or the like, so that the situation of mistaken clicking of the adjacent key is caused. Therefore, the embodiment of the invention can be applied to different input devices such as a physical keyboard, a virtual keyboard and the like.

Taking a QWERT keyboard (full keyboard) as an example, the keys adjacent to key "s" may include: "a", "q", "w", "e", "d", "z", "x" and "c", therefore, when the user performs the clicking operation and the target key is the key "s", there is a possibility that the user may mistakenly click on the adjacent keys "q", "w", "e", "d", "z", "x" and "c", and if the error correction attempt is performed on all the adjacent keys like the conventional scheme, a large amount of computing resources are required, resulting in low efficiency.

In order to solve the above problem, in the embodiment of the present invention, the key region is divided into a plurality of sub-regions, for example, the key "s" may be divided into 10 × 10 sub-regions, that is, the key "s" includes 100 sub-regions in total of 10 rows and 10 columns, when the user clicks the key "s", a click position may fall into one of the 100 sub-regions, and probabilities of different sub-regions corresponding to the adjacent keys are different. For example, if the click position is the sub-region at the top left corner of the above 100 sub-regions, the adjacent keys closest to the sub-region only have keys "a", "q", and "w", that is, the probability that the adjacent keys corresponding to the click position are the keys "a", "q", and "w" is higher, so that the error correction attempt is performed on the three adjacent keys, and compared with performing the error correction attempt on all the adjacent keys in the prior art, the calculation amount can be reduced, and the error correction efficiency is improved.

Step 102, determining a target adjacent key corresponding to the click position;

in the embodiment of the invention, the target adjacent key can be used for representing the key which corresponds to the clicking position and can most embody the clicking intention of the user. For example, the key closest to the click position can best reflect the click intention of the user, and therefore, the target adjacent key can be the key closest to the click position.

The embodiment of the invention can provide the following technical scheme for determining the target adjacent key corresponding to the click position:

technical solution 1

In technical solution 1, the step of determining the target neighboring key corresponding to the click position may specifically include the following sub-steps:

substep S11, determining an offset position of the sub-region corresponding to the click position relative to the center of the key region;

and a substep S12, determining at least one key closest to the offset position as a target neighboring key corresponding to the click position.

In a specific application, when a user performs a click operation, if the sub-region corresponding to the click position is located in the center of the key region, it may be considered that the probability of occurrence of a false click is minimum, and if the sub-region corresponding to the click position has an offset with respect to the center of the key region, and the farther the offset position is, the greater the probability of occurrence of a false click is, so that the target adjacent key is determined according to the offset position of the sub-region corresponding to the click position with respect to the center of the key region, and the accuracy of determining the target adjacent key may be improved. For example, if the sub-region corresponding to the click position is located at the upper left corner of the key "s", that is, the click position has a larger offset position relative to the center of the key region, and the offset position may have directionality, for example, the offset position is located in the upper left direction of the center of the key region, and the keys closest to the offset position may specifically include keys "a", "q", "w", and the like, it may be determined that the target adjacent keys corresponding to the click position are keys "a", "q", "w". Similarly, if the sub-region corresponding to the click position is located at the upper right corner of the key "s", it may be determined that the target adjacent keys are the keys "w", "e", "d".

Technical solution 2

In technical solution 2, the click position may specifically include coordinate information; the step of determining the target neighboring key corresponding to the click position may specifically include the following sub-steps:

substep S21, determining coordinate information of the click position;

and a substep S22, searching for a target adjacent key corresponding to the coordinate information in the mapping relationship between the coordinate information and the target adjacent key established in advance.

In an application example of the present invention, the coordinate information of the center of the key region may be set to [0,0], and assuming that the key region includes n × n (n is an odd number greater than or equal to 3) sub-regions, the range of the coordinate information corresponding to the sub-regions may be within an interval corresponding to [ - (n-1)/2, + (n-1)/2], that is, the maximum value of the coordinate information of the X, Y axis in two dimensions is + (n-1)/2, and the minimum value is- (n-1)/2. For example, when n is 3, the coordinate information corresponding to the 9 sub-regions can be referred to in table 2.

TABLE 2

(-1,1)	(0,1)	(1,1)
			(-1,0)	(0,0)	(1,0)
(-1,-1)	(0，-1)	(1，-1)

In a specific application, the mapping relationship between the coordinate information and the target adjacent key may be stored in advance. For example, if the target neighboring key corresponding to the coordinate information [ -4, +3] of the pre-stored key "s" includes the keys "a", "q", and "w", the corresponding target neighboring key can be directly found according to the coordinate information, so that the time for finding the target neighboring key can be saved. Of course, the above-mentioned manner of setting the coordinate information of the click position is only an application example of the present invention, and in practical applications, a person skilled in the art may flexibly set the coordinate information of the click position as needed.

It is understood that the embodiment of the present invention also does not limit the specific manner of the established mapping relationship between the coordinate information and the target adjacent key. For example, according to the distance between the coordinate information of the click position and the neighboring keys around the click position, the neighboring key with the closest distance may be used as the target neighboring key, and the mapping relationship between the coordinate information and the target neighboring key may be stored. For example, if the coordinate information [ -4, +3] of the key "s" is located at the upper left corner of the key, and the key closest to the coordinate information includes keys "a", "q", and "w", the target adjacent keys corresponding to the coordinate information may be stored as keys "a", "q", and "w" in advance.

In an optional embodiment of the present invention, after determining the target adjacent key corresponding to the click position, the target adjacent key may be further filtered according to a spelling rule, so as to further narrow the range of the target adjacent key and reduce the amount of computation. In an application example of the present invention, assuming that a target input string to be input by a user is "wo" (pinyin corresponding to a chinese character "i"), and "o" is mistakenly input as "i" in an input process, an actual input string is "wi", wherein coordinate information corresponding to a click position of a user click key "i" is [4, -2], which can be obtained according to a mapping relationship between pre-established coordinate information and target adjacent keys, target adjacent keys corresponding to the coordinate information are keys "o" and "k", and since a character string "wk" does not conform to a pinyin rule, a target adjacent key "k" can be filtered, and subsequent error correction processing is performed only for the target adjacent "o", thereby reducing a calculation amount and improving error correction efficiency.

And 103, carrying out error correction processing on the input string according to the target adjacent key.

After the target adjacent key corresponding to the click position is determined, the input string can be subjected to error correction processing according to the target adjacent key. It is understood that the embodiment of the present invention does not limit the specific method for performing error correction processing on the input string according to the target adjacent key. For example, the error correction string may be obtained by first replacing the characters in the input string with the characters corresponding to the target adjacent keys; then, respectively calculating scores of the input string and the error correction string, and when the score of the error correction string is larger than the score of the input string, considering that error click occurs; and finally, acquiring error correction candidate items corresponding to the error correction strings, and outputting the error correction candidate items for a user to select.

In summary, the embodiment of the present invention determines the target neighboring key based on the click position of the key by the user, and performs error correction processing on the input string according to the target neighboring key. The key area can be divided into a plurality of sub-areas, and the click position corresponds to the sub-area; because the embodiment of the present invention can determine the sub-region corresponding to the click position in the key region, and further can determine the target adjacent key corresponding to the sub-region, compared with the existing scheme, the embodiment of the present invention locates the click position to a smaller sub-region granularity, so as to exclude some adjacent keys, reduce the range of determining the target adjacent key, and further reduce the number of target adjacent keys used in error correction, thereby reducing the calculation amount for calculating the target adjacent key, and reducing the calculation amount required in error correction of the target adjacent key, and further improving the error correction efficiency.

Method embodiment two

This embodiment is an optional embodiment of the first method embodiment, and details a process of performing error correction processing on the input string according to the target neighboring key are described on the basis of the first method embodiment. Referring to fig. 2, a flowchart illustrating steps of a second embodiment of an input error correction method according to the present invention is shown, which may specifically include the following steps:

step 201, obtaining a click operation of a key corresponding to a character in an input string and a click position of the click operation in a key area; the button area may specifically include a plurality of sub areas, and the click position corresponds to the sub area;

step 202, determining a target adjacent key corresponding to the click position;

step 203, determining the mapping probability between the click position and the target adjacent key;

and 204, carrying out error correction processing on the input string according to the mapping probability.

In the embodiment of the invention, the mapping relation among the click position, the key and the mapping probability can be pre-established; wherein the mapping probability can be used to reflect the click intention of the user.

In an optional embodiment of the present invention, the step of determining the mapping probability between the click position and the target neighboring key may specifically include:

and searching for the mapping probability between the click position and the target adjacent key in the mapping relation among the pre-established click position, key and mapping probability.

Application example 1

Generally, the probability of the error click between the keys with the shorter distance is higher, and therefore, according to the distance from the coordinate information corresponding to the click position to the center of the target adjacent key region, the application example establishes the mapping relation among the click position, the keys and the mapping probability so as to improve the accuracy of the mapping probability.

Assuming that coordinate information corresponding to a click position of a click button "s" is [ -4, +3], a target adjacent button corresponding to the click position includes buttons "a", "q", "w", and a mapping probability can be determined according to distances from the coordinate information [ -4, +3] to centers of respective button regions of the target adjacent buttons "a", "q", "w", the mapping probability is greater as the distance is smaller, and it is assumed that the mapping probability between the click position [ -4, +3] and the button "s" is: 0.4, the mapping probability with the target neighboring key "a" is: 0.3; the mapping probability with the target neighboring key "q" is: 0.1; the mapping probability with the target neighboring key "w" is: 0.2, and the sum of the mapping probabilities is 1. A mapping relationship between the click position [ -4, +3], the key "s", "a", "q", "w", and the mapping probability 0.4, 0.3, 0.2, 0.1 can be established. Referring to table 1, a specific illustration of the mapping relationship between click positions, keystrokes, and mapping probabilities of the present invention is shown.

TABLE 1

Click position of key "s	Push-button	Mapping probabilities
			[-4,+3]	s	0.4
[-4,+3]	a	0.3
			[-4,+3]	q	0.2
[-4,+3]	w	0.1

In an optional embodiment of the present invention, when storing the mapping probability, in addition to considering a distance from the coordinate information corresponding to the click position to the center of the target adjacent key region, whether the input string conforms to the spelling rule may be considered, and specifically, the mapping probability between the key corresponding to the character conforming to the spelling rule and the click position may be increased, and the mapping probability between the key corresponding to the character not conforming to the spelling rule and the click position may be decreased. For example, for the key "i" and the key "o", in the case of "w", the character string "wo" is more in accordance with the spelling rule than the character string "wi", and therefore, in the case of "w", even if the key clicked by the user is the key "i", the mapping probability between the click position and the key "o" may be larger than the mapping probability between the click position and the key "i".

Application example two

The application example takes whether the characters corresponding to the keys and the spelling rule are met or not as an example, and explains a specific process for establishing a mapping relation among the click positions, the keys and the mapping probabilities, so that the accuracy of the mapping probabilities is further improved on the basis of the application example I.

Assuming that the input string is "wi", and acquiring the click position of the click key "w" and the click position of the click key "i", querying the target adjacent key corresponding to the click position of the obtained key "w" as follows: keys "q", "a", "s", since "w" is the first character of the input string, in a specific application, a mapping relationship between the click position of the click key "w" and the mapping probability between the corresponding target adjacent keys may be stored in advance when "w" is the first character, e.g., when w "is the first character, the mapping probability between the click position of the key" w "and the key" w "is 0.6, the mapping probability with the target adjacent key" q "is 0.1, the mapping probability with the target adjacent key" a "is 0.2, and the mapping probability with the target adjacent key" s "is 0.1; the target adjacent key corresponding to the click position of the key "i" is: keys "o" and "k", in a case where only the distance between the click position and the target adjacent key is considered, without considering the above "w", the mapping probability between the click position of the key "i" and the key "i" is 0.5, the mapping probability with the target adjacent key "o" is 0.2, and the mapping probability with the target adjacent key "k" is 0.3; and in the case of considering the above "w", the mapping probability between the click position of the key "i" which does not conform to the spelling rule and the key "i" is 0.2, the mapping probability with the target adjacent key "o" which conforms to the spelling rule is 0.7, and the mapping probability with the target adjacent key "k" which does not conform to the spelling rule is 0.1. Therefore, the obtained mapping probability is more consistent with the spelling rule on the basis of being consistent with the key position distribution, and the accuracy of error correction is further improved.

In the embodiment of the present invention, the input string may be subjected to error correction processing according to the mapping probability. Specifically, at least one target adjacent key with the highest mapping probability may be obtained, and characters in an input string are replaced with the at least one target adjacent key with the highest mapping probability to obtain an error correction string corresponding to the input string, and the error correction string is presented to a user for selection by the user.

To sum up, in the embodiment of the present invention, the key region is divided into a plurality of sub-regions, and the click position corresponds to the sub-region, because the directions and distances of different sub-regions and the same target neighboring key are different, different sub-regions and the target neighboring key have different mapping probabilities. According to the embodiment of the invention, the error correction processing is carried out on the input string according to the mapping probability between the target adjacent key and the corresponding click position, and compared with the prior art that the error correction is tried on all adjacent keys around the key, the error correction accuracy can be further improved on the basis of improving the error correction efficiency.

Method embodiment three

This embodiment is an optional embodiment of the second method embodiment, and details a process of performing error correction processing on the input string according to the mapping probability are described on the basis of the second method embodiment. Referring to fig. 3, a flowchart illustrating a third step of an input error correction method according to an embodiment of the present invention is shown, which may specifically include the following steps:

301, acquiring a click operation of a key corresponding to a character in an input string and a click position of the click operation in a key area; the button area may specifically include a plurality of sub areas, and the click position corresponds to the sub area;

step 302, determining a target adjacent key corresponding to the click position;

step 303, determining the mapping probability between the click position and the target adjacent key;

step 304, determining an error correction string corresponding to the input string according to the target adjacent key;

in a specific application, characters in the input string may be replaced with characters corresponding to the target adjacent key to obtain a corresponding error correction string, for example, in the second application example, characters in the input string "wi" may be replaced with characters corresponding to the target adjacent key to obtain at least one error correction string, which may specifically include: "qi", "ai", "si", "wo", "wk".

305, obtaining an error correction string score corresponding to the error correction string according to the mapping probability between the click position and the target adjacent key;

in an alternative embodiment of the present invention, the error correction string score may be obtained by:

step S31, determining the mapping probability between the keys corresponding to the characters in the error correction string and the click positions;

step S32, the product of the mapping probabilities corresponding to the plurality of characters is used as the score of the error correction string corresponding to the error correction string.

Here, a process of calculating an error correction string score of an error correction string will be described by taking the above second application example as an example. Assuming that the input string is "wi", wherein a mapping probability between a click position of a click key "w" and the key "w" is 0.6, a mapping probability with a target adjacent key "q" is 0.1, a mapping probability with a target adjacent key "a" is 0.2, and a mapping probability with a target adjacent key "s" is 0.1; and the mapping probability between the click position of the click key "i" and the key "i" is 0.2, the mapping probability with the target adjacent key "o" is 0.7, and the mapping probability with the target adjacent key "k" is 0.1. Then the error correction string score corresponding to the error correction string "qi" can be calculated by the following steps: firstly, acquiring the mapping probability between the clicking position of a clicking key 'w' and a target adjacent key 'q' as 0.1; then, acquiring the mapping probability between the click position of the click key i and the key i to be 0.2; finally, the product of the two mapping probabilities is used as an error correction string score corresponding to the error correction string "qi", specifically, 0.1 × 0.2 — 0.02. Similarly, the error correction string score for which the pair of error correction string "wo" is obtained may be calculated to be 0.6 × 0.7 — 0.42, and the error correction string scores for all error correction strings may be calculated.

It is to be understood that the above-mentioned process of calculating the score of the error correction string according to the mapping probability is only an application example of the present invention, and in practical applications, the embodiment of the present invention does not impose a limitation on a specific method for calculating the score of the error correction string. For example, the mapping probability between the corresponding key of each character in the error correction string and the click position may be obtained by obtaining a logarithmic value and then summing the logarithmic values to obtain the score of the error correction string.

And step 306, when the score of the error correction string meets a preset score condition, performing error correction processing on the input string according to the error correction string.

In an optional embodiment of the present invention, the preset scoring condition may specifically be that the error correction string score meets a preset threshold range, for example, greater than or equal to 0.4. In practical applications, a person skilled in the art can flexibly determine the threshold range, and the embodiment of the present invention does not limit the specific value of the threshold range and the determination manner thereof. Further, the embodiment of the present invention may also dynamically adjust the threshold range, and if it is found that the error correction result generated for the current threshold range of the user is not ideal, the threshold range may be adjusted accordingly. For example, for a parkinson patient, or an elderly person, who may have difficulty performing accurate procedures, the threshold range may be increased; for another example, for a user who is skilled in accurate operation, the threshold range can be narrowed to improve the accuracy of error correction; for another example, the corresponding relationship between the user input content and the user category may be stored in the server, when the input content of the user is received, the user category (for example, an accurate user or a non-accurate user) matching the input content may be obtained by querying the server, and the threshold range may be adjusted according to the user category.

In another optional embodiment of the present invention, the preset score condition may specifically be that the error correction string score is greater than the input string score, and the input string score may specifically be a product of mapping probabilities between keys corresponding to characters in the input string and click positions. In the second application example, the specific step of obtaining the score of the input string corresponding to the input string "wi" may include: firstly, acquiring a mapping probability p1 between a click position of a click key "w" and the key "w"; then, acquiring a mapping probability p2 between the click position of the click key "i" and the key "i"; finally, the product of p1 and p2 is defined as an input string score corresponding to the input string "wi", and specifically, 0.6 × 0.2 is 0.12. Since the score 0.42 of the error correction string of the pair of error correction strings "wo" is greater than the score 0.12 of the input string corresponding to the input string "wi", which indicates that the error correction string is better than the input string, it can be determined that the error correction string "wo" satisfies the preset score condition, and further the error correction string can be displayed to the user to prompt the user to correct the error. Further, the candidate item corresponding to the error correction string can be obtained and output for the user to select.

According to the embodiment of the invention, the error correction string corresponding to the input string is determined according to the mapping probability between the target adjacent key and the clicking position, and when the score of the error correction string accords with the preset score condition, the error correction processing is carried out on the input string according to the error correction string. The character corresponding keys and the click positions in the error correction string meeting the preset score condition have high mapping probability generally, and the input intention of a user can be reflected most, so that the accuracy of error correction can be improved by using the error correction string to correct errors.

Method example four

In a specific application, in order to enable the mapping probability between the click position and the target adjacent key to better conform to the input habit of the user, the embodiment can obtain the error correction model by collecting the historical click information of the user and training, so that the mapping probability of the target adjacent key corresponding to the click position can be calculated by the error correction model.

Referring to fig. 4, a flowchart illustrating a fourth step of an input error correction method according to an embodiment of the present invention is shown, which may specifically include the following steps:

step 401, obtaining a click operation of a key corresponding to a character in an input string and a click position of the click operation in a key area; the button area may specifically include a plurality of sub areas, and the click position corresponds to the sub area;

step 402, determining a target adjacent key corresponding to the click position according to an error correction model; the error correction model is established according to collected historical click information of the user, and the historical click information may specifically include: inputting the click position of a key corresponding to the character in the string;

step 403, determining the mapping probability between the click position and the target adjacent key according to the error correction model;

and step 404, carrying out error correction processing on the input string according to the target adjacent key and the mapping probability.

In an optional embodiment of the present invention, the error correction model may specifically include a first error correction model and/or a second error correction model; the first error correction model is obtained by training according to collected historical click information of different users and can be used for reflecting input habits of most users. Further, when the first error correction model is trained, the users can be divided into different groups according to the historical input information of the users, for example, the groups with different ages, such as an old group, a middle-aged group, a young group and the like; or, the method can also be divided into groups with different culture levels, such as a primary culture group, a middle culture group, a high culture group and the like, and then the first error correction model is obtained through training according to historical click information of different groups, so that the first error correction model can be suitable for users of different groups, and the error correction accuracy can be improved.

In an application example of the present invention, the range of the mistaken clicks may be larger due to eye-flowers or hand trembling in the elderly population, for example, the click position is located at the upper left corner of the "s" key, and the target adjacent keys may be generally considered to include keys "a", "q", "w"; the old people may also click the keys "z", "e", etc. by mistake. Therefore, when the first error correction model is trained according to historical click information of the old people group, more target adjacent keys can be added, so that the trained first error correction model can better accord with the input characteristics of the old people group.

The second error correction model is obtained by training according to collected historical click information of the same user and can be used for reflecting individual differences and personal input behavior habits of the user.

The embodiment of the invention corrects the error based on the first error correction model and/or the second error correction model, so that the error correction process can accord with the behavior habits of most users and can reflect individual differences. When the user personalized data is not accumulated enough, the error correction can be carried out by taking the first error correction model as the main mode, and when the user personalized data is accumulated enough, the error correction can be carried out by taking the second error correction model as the main mode. The establishment of the first error correction model is derived from historical click information of a large number of different users, and can reflect average click tendency of most users. However, since there are often individual differences between users, the first error correction model is not sufficient to describe the click tendencies of all users. Therefore, according to the embodiment of the invention, the historical click information of the user is collected and the second error correction model corresponding to the user is obtained by training according to the process that the user uses the input method, so as to reflect the individual click tendency and the input habit of the user.

In an optional embodiment of the present invention, the step of collecting historical click information of different and/or the same users may specifically include:

It is to be understood that the above three steps of collecting the historical click information of different and/or the same users are only an application example of the present invention, and in practical applications, the embodiment of the present invention does not impose a limitation on the manner of collecting the historical click information of the users.

Application example three

The specific process of establishing the error correction model by collecting the historical click information of the user is described below by a specific application example.

First, historical click information of the user is collected. Supposing that input information of a user for 100 input strings containing characters "s" is collected, the click positions of keys corresponding to the characters "s" in the 100 input strings are [ -4, +3], the number of times of screen-up of the user for candidate items corresponding to the input strings is 60, the number of times of screen-up of the obtained candidate items corresponding to the corrected strings is 20 after the user actively corrects the characters "s" to the characters "w", the number of times of screen-up of the obtained candidate items corresponding to the corrected strings is 3 after the user actively corrects the characters "s" to the characters "a", and the number of times of screen-up of the obtained candidate items corresponding to the corrected strings is 2 after the user actively corrects the characters "s" to the characters "q"; and after the user replaces the character "s" with the character "w" for the input method program, the number of times of screen-up of the obtained candidate item corresponding to the error correction string is 10, the number of times of screen-up of the obtained candidate item corresponding to the error correction string is 2 after the character "s" is replaced with the character "a", and the number of times of screen-up of the obtained candidate item corresponding to the error correction string is 3 after the character "s" is replaced with the character "q".

Then, the collected historical click information can be used as sample data to train an error correction model, so that when a click position is input, a target adjacent key corresponding to the click position and mapping probabilities between the click position and each target adjacent key can be output through the error correction model. Specifically, the number of times that the user actually responds to the character at the click position can be obtained from the historical click information, for example, the number of times that the user actually responds to the character "s" at the click position [ -4, +3] of the "s" key is 60 times, the number of times that the user actually responds to the character "w" at the click position [ -4, +3] of the "s" key is 20+10 ═ 30 times, and the number of times that the user actually responds to the character "a" at the click position [ -4, +3] of the "s" key is 3+2 ═ 5 times, etc., then the error correction model can be trained by using the character that the user actually responds to the click position and the number of times that the response is a character as sample data.

In an application example of the present invention, when the click position of the click key "s" is [ -4, +3], substituting the click position into the error correction model to obtain the target neighboring key corresponding to the click position includes: key "w", key "a", and key "q"; and the mapping probability between the click position and the key "s" is: 0.6, the mapping probability with the target neighboring key "w" is: 0.3, the mapping probability with the target neighboring key "a" is: 0.05, the mapping probability with the target neighboring key "q" is: 0.05.

application example four

The following describes a specific process of obtaining the target neighboring key corresponding to the clicked position and the mapping probability of the corresponding target neighboring key by using an error correction model through a specific application example.

Assuming that the input string is "wi", and the click positions of the key "w" and the key "i" are respectively substituted into the error correction model, the error correction model can output that the target adjacent key corresponding to the key "w" is: keys "q", "a", "s", and the mapping probability between the click position of the key "w" and the key "w" are 0.6, the mapping probability with the target adjacent key "q" is 0.1, the mapping probability with the target adjacent key "a" is 0.2, and the mapping probability with the target adjacent key "s" is 0.1; the target adjacent key corresponding to key "i" is: keys "o" and "k", and in consideration of the above "w", the mapping probability between the click position of the key "i" and the key "i" is 0.2, the mapping probability with the target adjacent key "o" is 0.7, and the mapping probability with the target adjacent key "k" is 0.1.

In a specific application, after a first error correction model is obtained through training of historical click information of a large number of different users, the first error correction model can be stored in a binary file of a server and issued to a client through the server, and the first error correction model can be updated through a network.

In the embodiment of the present invention, the training of the second error correction model is from the accumulation of each click operation of the user, and if the user uses the input method for the first time and the accumulated user personalized data is empty, the first error correction model may be used to correct the error. And in the process of using the input method by the user later, training by collecting the historical click information of the user to obtain a second error correction model. For the second error correction model, the training process is similar to that of the first error correction model, except that the initial state of the second error correction model may be null, that is, starting from the user installing the input method, the null second error correction model of the user is initialized, and the second error correction model of the user is gradually adjusted by continuously collecting the historical click information of the user. Optionally, the initial state of the second error correction model may also be the established first error correction model, and the first error correction model is gradually adjusted through collecting and analyzing the historical click information of the user to form the second error correction model of the user.

In the embodiment of the invention, in the input process of the user, the second error correction model can be further adjusted and updated continuously according to the input behavior of the user, so that the second error correction model is continuously optimized and better conforms to the behavior habit of the user. Specifically, the embodiment of the present invention may provide at least four following ways to update the second error correction model:

the method comprises the steps that firstly, if a user selects displayed error correction candidate items to be displayed on a screen, the error correction aiming at a click position is effective, and then the error correction model is subjected to forward accumulation for one time, for example, the mapping probability between the click position and a key corresponding to a character in an error correction string is increased;

if the user normally selects the candidate word corresponding to the input string to be displayed on the screen, the input of the user is correct, and the mapping probability between the click position and the key corresponding to the character in the input string is increased;

if the user performs a backspace operation on the input string and inputs other input strings after correction, performing forward accumulation on the error correction model once according to the correction string, for example, increasing the mapping probability between the click position and the key corresponding to the character in the correction string;

and if the user performs modification operation on the provided error correction candidate item, which indicates that error correction performed on the click position is wrong, performing negative accumulation on the error correction model for one time, for example, reducing the mapping probability between the click position and the key corresponding to the character in the error correction string.

Through the four updating modes, the user can know the keys which are frequently pressed by mistake, the clicking position information of the pressed by mistake and the corresponding clicking position information under the condition of correct clicking. Therefore, the error click condition and the click tendency of the user can be obtained, and the second error correction model is continuously adjusted so as to continuously improve the accuracy of error correction.

In practical application, because the error correction of the user input string cannot be guaranteed to be completely correct, if the user input does not have errors, the input method program makes wrong automatic error correction behavior, and the influence on the user experience is more serious than that of the missed judgment, so that the negative accumulation of the false judgment can be increased when the error correction model is trained to solve the problems. Optionally, when the error correction candidate corresponding to the error correction string is output as a candidate, the error correction candidate may be marked, for example, the error correction candidate is highlighted in a marking or highlighting manner to prompt the user to select or modify, so as to form an interaction with the user, which is more favorable for forming a second error correction model conforming to the input habit of the user.

Method example five

The embodiment is an optional embodiment of the fourth embodiment, and on the basis of the fourth embodiment, the first error correction model and the second error correction model are fused, so that the error correction process can meet the average click tendency of most users and further conform to specific user behaviors with individual differences, and the accuracy of error correction can be improved.

The following describes a fusion mode of two error correction models according to an embodiment of the present invention by taking a formula as an example:

A_i＝Sum_(x,y)*P_i+f_per*S_i (1)

f_per＝0.4×(S_per+1)+1 (2)

wherein A is_iDenotes the number of times key i is mapped at the click position (x, y), Sum_(x,y)Representing the order of magnitude of the number of clicks, P, of key i at click position (x, y)_iRepresenting the position between the click (x, y) and the key i in the first error correction modelThe mapping probability of f_perRepresenting the weight occupied by the second error correction model, S_iIndicating that the click position (x, y) in the accumulated personalization data corresponds to the number of clicks of the key i. S_perRepresents the sum of the personalized click data accumulated at the coordinate, which is positively correlated with the scale of the personalized data accumulation, i.e. S_perThe smaller, f_perThe smaller; s_perThe larger, f_perThe larger. Therefore, the first error correction model is mainly used when the personalized data are less, and the second error correction model is mainly used when the personalized data are more.

Application example five

The application example specifically illustrates a process of fusing a first error correction model and a second error correction model, where a (x, y) represents that coordinate information corresponding to a click position of a click button "a" is (x, y), and the click position a (x, y) is an order of magnitude Sum of actual click times_(x,y)Is 10000.

Step a1, substituting the click position a (x, y) into the first error correction model, and calculating the first error correction model to obtain the following first error correction model data: the mapping probabilities between the click position a (x, y) and the key are respectively: a: 0.980; s: 0.020;

step a2, obtaining the following second error correction model data according to the counted number of clicks of the key corresponding to the click position a (x, y) in the personalized data of the user: the times that the user actually responds to the characters 'a','s', 'w' at the click position a (x, y) are 100 times, 3 times and 1 time respectively;

step A3, calculating the weight f occupied by the second error correction model according to the formula (2)_perComprises the following steps:

{f_per＝0.4*(100+5+1+1)+1＝43.8}；

and step A4, fusing the first error correction model data and the second error correction model data according to the formula (1), and calculating the times of mapping the click position a (x, y) to the key i.

Specifically, the number of times the click position a (x, y) is mapped to the key "a" is:

10000*0.980+fper(a)*100＝9800+4380＝13940

the number of times the click position a (x, y) is mapped to key "s" is:

10000*0.020+fper(s)*5＝419

the number of times click position a (x, y) is mapped to key "w" is:

fper(w)*1≈44

similarly, the number of times the click position a (x, y) is mapped to each key can be calculated by the above-described process.

The following describes an error correction process according to an embodiment of the present invention, taking specific operations of a user as an example:

firstly, a user clicks a key on touch screen equipment to input pinyin, and an input method program carries out click position analysis by collecting click coordinates of the user on a screen. The key clicked by the user and which of the n × n sub-regions of the key region the click position corresponds to can be obtained through analysis.

Then, the input method program transmits the collected input string, the click position of the key and other related information to the input method kernel for analysis, and the input method kernel obtains a corresponding syllable structure by establishing a syllable network, for example, for the input string "fangan", after establishing the syllable network, the input string may be divided into the following syllable structures: [ fan ] [ gan ], [ fang ] [ an ]. The error correction of the key-press of the embodiment of the invention occurs in the process of establishing the syllable network. In practical application, key sound mapping exists in an input method, namely keys correspond to syllables one by one, and the embodiment of the invention adds adjacent keys around one key into a syllable network and endows different mapping probabilities.

Next, in the process of establishing the syllable network, the input method kernel calculates and obtains the score of the error correction string corresponding to each error correction string according to the mapping probability between the key corresponding to each character in the input string and the click position.

And finally, acquiring candidate items corresponding to the error correction strings meeting preset scoring conditions for error correction, and displaying the error correction candidate items to a user and carrying out error correction prompt.

Application example six

Assuming that a target character string which a user intends to input is 'shuju' (pinyin of 'data'), the user firstly inputs 'sh' in the input process, but when the right thumb wants to click the key 'u' after clicking the key 'h', the action amplitude is deviated and is at the upper left corner of the key 'j', and then the user inputs the rest character string 'ju', namely the actual input string input by the user is 'shjju'.

Wherein, the click position of the key "j" is on the upper left, and the mapping probability between the click position and the key "j" is 0.5, the mapping probability between the key "u" and the target adjacent key "u" is 0.4, and the mapping probability between the key "i" and the target adjacent key is 0.1 by inquiring the mapping relation of the mapping probabilities between the click position and the key; then, respectively calculating the scores of the error correction strings, and adding the character u into the subsequent pinyin string creating and word searching processes if the error correction string shuju has the highest error correction string score; and finally, performing comprehensive score calculation by using a language model and other strategies according to the word frequency obtained by word searching, wherein the comprehensive score of the error correction string 'shuju' is higher than that of the original input string 'shju', and displaying the error correction candidate item 'data' corresponding to the error correction string 'shuju' to a user and performing error correction prompt. The language model is a measure for judging whether the input string meets the spelling specification of the language and the fluency thereof by using a statistical method. For example, for an error correction string, the probability of passing each character in the string as it appears at its position in the error correction string may be calculated based on the language model. Further, the user can directly modify the original input string through the error correction prompt or directly click the error correction candidate item to perform the screen-on operation. Optionally, after the input method program records the screen-on operation, the error correction model may be updated to improve the accuracy of correcting the error.

In an alternative embodiment of the present invention, in order to improve the error correction efficiency, common error input strings may be stored in advance. For example, by analyzing input information of a user, it is known that the user inputs "shuju" into "shjju" for many times, and corrects the shjju "with a backspace key for many times, it may be considered that" shjju "is a common misinput string of the user, a mapping relationship between the common misinput string" shjju "and a correct input string" shuju "may be stored for the user, and when the user inputs" shjju ", the input string may be directly corrected through a common misinput string query, and a candidate item corresponding to the correct input string" shuju "is displayed, so that a calculation amount may be reduced, and an error correction efficiency may be improved.

The embodiment of the invention combines the first error correction model and the second error correction model together, and takes the first error correction model as the main part when the personalized data is less; when the individualized data has accumulated sufficiently, the second error correction model is dominant. Therefore, better error correction effect can be achieved at different stages, and the use experience of a user is improved.

Method example six

The present embodiment is an optional embodiment of the fourth method embodiment, and on the basis of the fourth method embodiment, a corresponding relationship between the user account and the second error correction model may also be established. Referring to fig. 5, a flowchart illustrating steps of a sixth embodiment of an input error correction method according to the present invention is shown, which may specifically include the following steps:

step 501, downloading a corresponding second error correction model from a server after logging in a user account;

in a specific application, the server may specifically be any type of server, and for example, may be a general server, a cloud server, or the like. In the embodiment of the invention, a mapping relation between the user account and the second error correction model can be established, and the established second error correction model is synchronized to the server. Therefore, when the user uses any intelligent terminal, the second error correction model corresponding to the user account can be downloaded from the server for use only by logging in the user account. In addition, in the input process of the user, the second error correction model is updated according to the input information of the user, and the updated data information is uploaded to the server, so that the latest second error correction model is always stored in the server.

In a specific application, when a user uses an input method program in a certain intelligent terminal for input, if the user logs in an input method account and the account has a corresponding second error correction model in a server, the input method program downloads the second error correction model corresponding to the user from the server and performs error correction processing on an input string of the user according to the second error correction model and the existing first error correction model, so as to improve the accuracy.

Step 502, acquiring a click operation of a key corresponding to a character in an input string and a click position of the click operation in a key area; the button area may specifically include a plurality of sub areas, and the click position corresponds to the sub area;

step 503, determining a target adjacent key corresponding to the click position according to the second error correction model;

step 504, determining the mapping probability between the click position and the target adjacent key according to the second error correction model;

step 505, determining an error correction string corresponding to the input string according to the target adjacent key;

step 506, obtaining an error correction string score corresponding to the error correction string according to the mapping probability between the click position and the target adjacent key;

and 507, when the score of the error correction string meets a preset score condition, performing error correction processing on the input string according to the error correction string.

According to the embodiment of the invention, the click position of the user on the key is introduced, so that the error correction model is closer to the use habit of the user, and the error correction model which is more in line with the click tendency of the user is favorably established; in addition, the second error correction model is uploaded to the server, so that the user can use the second error correction model corresponding to the user on any intelligent terminal without re-establishing the second error correction model of the user on each device, and the error correction efficiency can be improved.

It should be noted that, for simplicity of description, the method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the illustrated order of acts, as some steps may occur in other orders or concurrently in accordance with the embodiments of the present invention. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred and that no particular act is required to implement the invention.

Device embodiment

Referring to fig. 6, a block diagram of an embodiment of an input error correction apparatus according to the present invention is shown, which may specifically include the following modules:

an obtaining module 601, configured to obtain a click operation of a key corresponding to a character in an input string and a click position of the click operation in a key region; the button area may specifically include a plurality of sub areas, and the click position corresponds to the sub area;

a determining module 602, configured to determine a target neighboring key corresponding to the click position; and

and an error correction module 603, configured to perform error correction processing on the input string according to the target neighboring key.

In an optional embodiment of the present invention, the determining module 602 may specifically include:

the offset position determining submodule is used for determining the offset position of the sub-area corresponding to the click position relative to the center of the key area;

and the first target adjacent key determining submodule is used for determining at least one key closest to the offset position as a target adjacent key corresponding to the clicking position.

In another optional embodiment of the present invention, the clicked position may specifically include coordinate information, and the determining module 602 may specifically include:

the coordinate determination submodule is used for determining coordinate information of the click position;

and the second target adjacent key determining submodule is used for searching and obtaining a target adjacent key corresponding to the coordinate information in a mapping relation between the coordinate information and the target adjacent key, which is established in advance.

In another optional embodiment of the present invention, the determining module 602 may specifically include:

the third target adjacent key determining submodule is used for determining a target adjacent key corresponding to the clicking position according to an error correction model; the error correction model is established according to collected historical click information of the user, and the historical click information may specifically include: and inputting the click position of the key corresponding to the character in the string.

In yet another optional embodiment of the present invention, the error correction module 603 may specifically include:

a mapping probability determination submodule for determining a mapping probability between the click position and the target adjacent key;

and the error correction submodule is used for carrying out error correction processing on the input string according to the mapping probability.

In yet another optional embodiment of the present invention, the mapping probability determining sub-module may specifically include:

the first mapping probability determining submodule is used for searching and obtaining the mapping probability between the click position and the target adjacent key in the mapping relation among the pre-established click position, key and mapping probability; and/or

And the first mapping probability determining submodule is used for determining the mapping probability between the clicking position and the target adjacent key according to an error correction model.

In yet another optional embodiment of the present invention, the error correction sub-module may specifically include:

the error correction string determining unit is used for determining an error correction string corresponding to the input string according to the target adjacent key;

the error correction string score determining unit is used for obtaining an error correction string score corresponding to the error correction string according to the mapping probability between the click position and the target adjacent key;

and the error correction processing unit is used for carrying out error correction processing on the input string according to the error correction string when the score of the error correction string accords with a preset score condition.

In yet another optional embodiment of the invention, the error correction model comprises a first error correction model and/or a second error correction model;

In yet another alternative embodiment of the present invention, the step of collecting historical click information of different and/or same users includes:

For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.

The embodiments in the present specification are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.

With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.

Fig. 7 is a block diagram illustrating an apparatus 800 for input error correction according to an example embodiment. For example, the apparatus 800 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, an exercise device, a personal digital assistant, and the like.

Referring to fig. 7, the apparatus 800 may include one or more of the following components: processing component 802, memory 804, power component 806, multimedia component 808, audio component 810, input/output (I/O) interface 812, sensor component 814, and communication component 816.

The processing component 802 generally controls overall operation of the device 800, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing elements 802 may include one or more processors 820 to execute instructions to perform all or a portion of the steps of the methods described above. Further, the processing component 802 can include one or more modules that facilitate interaction between the processing component 802 and other components. For example, the processing component 802 can include a multimedia module to facilitate interaction between the multimedia component 808 and the processing component 802.

The memory 804 is configured to store various types of data to support operation at the device 800. Examples of such data include instructions for any application or method operating on device 800, contact data, phonebook data, messages, pictures, videos, and so forth. The memory 804 may be implemented by any type or combination of volatile or non-volatile memory devices such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.

Power components 806 provide power to the various components of device 800. The power components 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the apparatus 800.

The multimedia component 808 includes a screen that provides an output interface between the device 800 and a user. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front facing camera and/or a rear facing camera. The front-facing camera and/or the rear-facing camera may receive external multimedia data when the device 800 is in an operating mode, such as a shooting mode or a video mode. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

The audio component 810 is configured to output and/or input audio signals. For example, the audio component 810 includes a Microphone (MIC) configured to receive external audio signals when the apparatus 800 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may further be stored in the memory 804 or transmitted via the communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

The I/O interface 812 provides an interface between the processing component 802 and peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to: a home button, a volume button, a start button, and a lock button.

The sensor assembly 814 includes one or more sensors for providing various aspects of state assessment for the device 800. For example, the sensor assembly 814 may detect the open/closed state of the device 800, the relative positioning of the components, such as a display and keypad of the apparatus 800, the sensor assembly 814 may also detect a change in position of the apparatus 800 or a component of the apparatus 800, the presence or absence of user contact with the apparatus 800, orientation or acceleration/deceleration of the apparatus 800, and a change in temperature of the apparatus 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. The sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

The communication component 816 is configured to facilitate communications between the apparatus 800 and other devices in a wired or wireless manner. The device 800 may access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 further includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

In an exemplary embodiment, the apparatus 800 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors or other electronic components for performing the above-described methods.

In an exemplary embodiment, a non-transitory computer-readable storage medium comprising instructions, such as the memory 804 comprising instructions, executable by the processor 820 of the device 800 to perform the above-described method is also provided. For example, the non-transitory computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

A non-transitory computer readable storage medium, instructions in which, when executed by a processor of a smart terminal, enable the smart terminal to perform a method of input error correction, the method comprising: acquiring click operation of keys corresponding to characters in an input string and click positions of the click operation in a key area; the key area comprises a plurality of sub-areas, and the click position corresponds to the sub-areas; determining a target adjacent key corresponding to the clicking position; and carrying out error correction processing on the input string according to the target adjacent key.

Fig. 8 is a schematic structural diagram of a server in an embodiment of the present invention. The server 1900 may vary widely by configuration or performance and may include one or more Central Processing Units (CPUs) 1922 (e.g., one or more processors) and memory 1932, one or more storage media 1930 (e.g., one or more mass storage devices) storing applications 1942 or data 1944. Memory 1932 and storage medium 1930 can be, among other things, transient or persistent storage. The program stored in the storage medium 1930 may include one or more modules (not shown), each of which may include a series of instructions operating on a server. Still further, a central processor 1922 may be provided in communication with the storage medium 1930 to execute a series of instruction operations in the storage medium 1930 on the server 1900.

The server 1900 may also include one or more power supplies 1926, one or more wired or wireless network interfaces 1950, one or more input-output interfaces 1958, one or more keyboards 1956, and/or one or more operating systems 1941, such as WindowsServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.

Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.

It will be understood that the invention is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the invention is only limited by the appended claims

The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.

The input error correction method, the input error correction device and the device for input error correction provided by the present invention are introduced in detail, and specific examples are applied in this text to explain the principle and the implementation of the present invention, and the description of the above embodiments is only used to help understanding the method of the present invention and the core idea thereof; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.

Claims

1. An input error correction method, comprising:

acquiring click operation of keys corresponding to characters in an input string and click positions of the click operation in a key area; the key area comprises n multiplied by n sub-areas, n is an odd number which is larger than or equal to 3, the click position corresponds to the sub-areas, and the click position is determined according to a coordinate system in the key area;

determining a target adjacent key corresponding to the clicking position;

carrying out error correction processing on the input string according to the target adjacent key;

wherein the performing error correction processing on the input string according to the target adjacent key comprises:

obtaining an error correction string score corresponding to the error correction string according to the mapping probability between the click position and the target adjacent key; the mapping probability is obtained according to an error correction model, the error correction model is established according to collected historical click information of the user, and the historical click information comprises: inputting the click position of a key corresponding to the character in the string;

2. The method of claim 1, wherein the step of determining the target neighboring key corresponding to the click position comprises:

3. The method of claim 1, wherein the click position comprises coordinate information, and the step of determining the target neighboring key corresponding to the click position comprises:

determining coordinate information of the click position;

4. The method of claim 1, wherein the step of determining the target neighboring key corresponding to the click position comprises:

and determining the target adjacent key corresponding to the click position according to an error correction model.

5. The method according to claim 1, characterized in that the error correction model comprises a first error correction model and/or a second error correction model;

6. The method of claim 5, wherein the step of collecting historical click information of different and/or same users comprises:

7. An input error correction apparatus, comprising:

the device comprises an acquisition module, a display module and a display module, wherein the acquisition module is used for acquiring the click operation of a key corresponding to a character in an input string and the click position of the click operation in a key area; the key area comprises n multiplied by n sub-areas, n is an odd number which is larger than or equal to 3, and the click position corresponds to the sub-areas;

the error correction module is used for carrying out error correction processing on the input string according to the target adjacent key;

the error correction string score determining unit is used for obtaining an error correction string score corresponding to the error correction string according to the mapping probability between the click position and the target adjacent key; the mapping probability is obtained according to an error correction model, the error correction model is established according to collected historical click information of the user, and the historical click information comprises: inputting the click position of a key corresponding to the character in the string;

8. The apparatus of claim 7, wherein the determining module comprises:

9. The apparatus of claim 7, wherein the click location comprises coordinate information, and wherein the determining module comprises:

10. The apparatus of claim 7, wherein the determining module comprises:

and the third target adjacent key determining submodule is used for determining a target adjacent key corresponding to the clicking position according to the error correction model.

11. The apparatus according to claim 7, wherein the error correction model comprises a first error correction model and/or a second error correction model;

12. The apparatus of claim 11, further comprising:

the first collection module is used for counting the click times of keys corresponding to input characters and the click positions in the key areas corresponding to the input characters aiming at the screen-on operation of candidate items corresponding to input strings; and/or

The second collection module is used for counting the number of clicks of keys corresponding to correction characters and the click positions in the key areas corresponding to the correction characters aiming at the corrected screen-on operation of the input string; and/or

And the third collection module is used for counting the click times of the keys corresponding to the error correction characters and the click positions in the key areas corresponding to the error correction characters aiming at the screen-on operation of the error correction candidate items corresponding to the error correction strings.

13. An apparatus for input error correction comprising a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by one or more processors the one or more programs including instructions for:

determining a target adjacent key corresponding to the clicking position;

14. The apparatus of claim 13, wherein the determining that the target neighboring key corresponding to the click position comprises:

15. The apparatus of claim 13, wherein the click position comprises coordinate information, and the determining the target neighboring key corresponding to the click position comprises:

determining coordinate information of the click position;

16. The apparatus of claim 13, wherein the determining that the target neighboring key corresponding to the click position comprises:

17. The apparatus according to claim 13, wherein the error correction model comprises a first error correction model and/or a second error correction model;

18. The apparatus of claim 17, wherein collecting historical click information of different and/or same users comprises:

19. One or more machine readable media having instructions stored thereon that, when executed by one or more processors, cause an apparatus to perform the input error correction method of any of claims 1-6.