CN110069143B

CN110069143B - Information error correction preventing method and device and electronic equipment

Info

Publication number: CN110069143B
Application number: CN201810059547.7A
Authority: CN
Inventors: 陈小帅; 臧娇娇
Original assignee: Beijing Sogou Technology Development Co Ltd
Current assignee: Beijing Sogou Technology Development Co Ltd
Priority date: 2018-01-22
Filing date: 2018-01-22
Publication date: 2024-06-07
Anticipated expiration: 2038-01-22
Also published as: CN110069143A

Abstract

The embodiment of the invention provides an information error correction prevention method, an information error correction device and electronic equipment, wherein the method comprises the following steps: after identifying that the input information needs to be corrected, determining error correction candidate information corresponding to the input information, wherein the error correction candidate information comprises: an error correction candidate and an error correction score for the error correction candidate; determining the complete probability of the sentence corresponding to the input information; determining correction scores of the error correction candidates according to the integrity probability and the error correction scores; after the correction score meets a preset condition, displaying the error correction candidate item; thereby effectively avoiding invalid error correction and improving the error correction accuracy.

Description

Information error correction preventing method and device and electronic equipment

Technical Field

The invention relates to the technical field of input methods, in particular to an information error correction prevention method, an information error correction prevention device and electronic equipment.

Background

With the development of computer technology, electronic devices such as mobile phones and tablet computers are becoming more popular, and great convenience is brought to life, study and work of people. These electronic devices are typically installed with an input method application (input method for short) so that a user can input information using the input method.

In the process of inputting information by a user using the input method, in order to ensure the accuracy and convenience of inputting information by the user, the input method provides a plurality of functions such as an error correction function, i.e., when an error in an input word is detected, candidates for error correction are displayed. However, when the user does not completely input a sentence, error correction is easily caused due to insufficient information. As shown in fig. 1, after the user inputs "you get me", you can be mistakenly corrected to "you wait me", because in the current input state, the language model of the latter is far better than the former, but the user does not finish inputting, and the user may expect to continue inputting "qq number" and the like, and at this time, it is meaningless to reveal invalid error correction, and the input of the user is plagued.

Disclosure of Invention

The embodiment of the invention provides an information error correction prevention method, which is used for improving the error correction accuracy.

Correspondingly, the embodiment of the invention also provides an information error correction prevention device and electronic equipment, which are used for ensuring the realization and application of the method.

In order to solve the above problems, an embodiment of the present invention discloses an information error correction method, which specifically includes: after identifying that the input information needs to be corrected, determining error correction candidate information corresponding to the input information, wherein the error correction candidate information comprises: an error correction candidate and an error correction score for the error correction candidate; determining the complete probability of the sentence corresponding to the input information; determining correction scores of the error correction candidates according to the integrity probability and the error correction scores; and after the correction score meets a preset condition, displaying the error correction candidate item.

Optionally, the identifying the input information requires error correction, including: inputting the input information into a language model, and determining a reference score of the input information; and if the reference score is smaller than the error correction threshold value, determining that the input information needs error correction.

Optionally, the determining the complete probability of the sentence corresponding to the input information includes: obtaining statement identification information according to the input information, and determining the complete probability of a corresponding statement according to the statement identification information, wherein the statement identification information comprises at least one of the following: punctuation marks, sentence tail words, association information corresponding to input information and input intervals corresponding to the input information.

Optionally, the analyzing the sentence identification information according to the input information, and determining the complete probability of the corresponding sentence according to the sentence identification information includes: acquiring punctuation marks at the tail of the input information; matching the punctuation mark with a set punctuation mark; if the punctuation mark is matched with the set punctuation mark, determining a first numerical value as the complete probability; and if the punctuation mark is not matched with the set punctuation mark, determining a second numerical value as the complete probability.

Optionally, the analyzing the sentence identification information according to the input information, and determining the complete probability of the corresponding sentence according to the sentence identification information includes: identifying sentence tail words from the input information; matching the sentence tail words with set identification words; if the sentence tail word is matched with the set identification word, determining the complete probability according to the sentence tail probability of the set identification word matched with the sentence tail word, wherein the sentence tail probability is the probability of taking the set identification word as the sentence tail of the sentence; and if the sentence tail word is not matched with the set identification word, determining a third numerical value as the complete probability.

Optionally, the analyzing the sentence identification information according to the input information, and determining the complete probability of the corresponding sentence according to the sentence identification information includes: determining corresponding association information according to the input information, wherein the association information comprises association words and association probability of the association words; determining the total number of the associated words and the maximum association probability, and calculating the ratio of the total number to a set value; and determining the complete probability according to the ratio and the maximum association probability.

Optionally, the determining the complete probability according to the ratio and the maximum association probability includes: and determining the maximum value of the ratio and the maximum association probability, and determining the maximum value as the complete probability.

Optionally, the analyzing the sentence identification information according to the input information, and determining the complete probability of the corresponding sentence according to the sentence identification information includes: determining the input information of the upper screen and the input interval of the next input operation; judging whether the input interval is larger than an average input interval or not; if the input interval is greater than the average input interval, determining a fourth value as the complete probability; and if the input interval is smaller than the average input interval, determining a fifth value as the complete probability.

Optionally, determining the correction score of the error correction candidate item according to the integrity probability and the error correction score comprises: determining a penalty score according to the integrity probability and the penalty weight; and adjusting the error correction score of the error correction candidate item by adopting the penalty score, and determining the correction score.

Optionally, the error correction score is determined by inputting error correction candidates into a language model, and before the presenting of the error correction candidates, the method further includes: judging whether the correction score is larger than a reference score of the input information, wherein the reference score is used for judging whether the input information has errors or not; and if the correction score is larger than the reference score, determining that the correction score meets a preset condition.

Optionally, after determining the complete probability of the sentence corresponding to the input information, the method further includes: judging whether the integrity probability is larger than an integrity threshold value or not; if the integrity probability is greater than an integrity threshold, executing the step of adjusting the error correction score of the error correction candidate item according to the integrity probability; and if the integrity probability is smaller than the integrity threshold value, displaying the error correction candidate items in the error correction candidate information.

The embodiment of the invention also discloses an information error prevention and correction device, which specifically comprises: the information determining module is used for determining error correction candidate information corresponding to the input information after identifying that the input information needs error correction, and the error correction candidate information comprises: an error correction candidate and an error correction score for the error correction candidate; the probability determining module is used for determining the complete probability of the sentence corresponding to the input information; the score determining module is used for determining the correction score of the error correction candidate item according to the complete probability and the error correction score; and the display module is used for displaying the error correction candidate items after the correction score meets a preset condition.

Optionally, the information determining module is specifically configured to input the input information into a language model, and determine a reference score of the input information; and if the reference score is smaller than the error correction threshold value, determining that the input information needs error correction.

Optionally, the probability determining module is specifically configured to obtain statement identification information according to the input information, and determine a complete probability of a corresponding statement according to the statement identification information, where the statement identification information includes at least one of the following: punctuation marks, sentence tail words, association information corresponding to input information and input intervals corresponding to the input information.

Optionally, the probability determining module includes: the first determining submodule is used for acquiring punctuation marks after the input information; matching the punctuation mark with a set punctuation mark; if the punctuation mark is matched with the set punctuation mark, determining a first numerical value as the complete probability; and if the punctuation mark is not matched with the set punctuation mark, determining a second numerical value as the complete probability.

Optionally, the probability determining module includes: the second determining submodule is used for identifying sentence tail words from the input information; matching the sentence tail words with set identification words; if the sentence tail word is matched with the set identification word, determining the complete probability according to the sentence tail probability of the set identification word matched with the sentence tail word, wherein the sentence tail probability is the probability of taking the set identification word as the sentence tail of the sentence; and if the sentence tail word is not matched with the set identification word, determining a third numerical value as the complete probability.

Optionally, the probability determining module includes: a third determining sub-module, configured to determine corresponding association information according to the input information, where the association information includes an association word and an association probability of the association word; determining the total number of the associated words and the maximum association probability, and calculating the ratio of the total number to a set value; and determining the complete probability according to the ratio and the maximum association probability.

Optionally, the third determining submodule is configured to determine a maximum value of the ratio and the maximum association probability, and determine the maximum value as the complete probability.

Optionally, the probability determining module includes: a fourth determining submodule for determining an input interval of the input information and the subsequent input information; judging whether the input interval is larger than an average input interval or not; if the input interval is greater than the average input interval, determining a fourth value as the complete probability; and if the input interval is smaller than the average input interval, determining a fifth value as the complete probability.

Optionally, the score determining module is configured to determine a penalty score according to the integrity probability and the penalty weight; and adjusting the error correction score of the error correction candidate item by adopting the penalty score, and determining the correction score.

Optionally, the error correction score is determined by inputting error correction candidates into a language model, and further comprising: the score judgment module is used for judging whether the correction score is larger than a reference score of the input information, and the reference score is used for judging whether the input information is wrong or not; and if the correction score is larger than the reference score, determining that the correction score meets a preset condition.

Optionally, the method further comprises: the threshold judging module is used for judging whether the integrity probability is larger than an integrity threshold or not; if the integrity probability is greater than an integrity threshold, executing the step of adjusting the error correction score of the error correction candidate item according to the integrity probability; and if the integrity probability is smaller than the integrity threshold value, displaying the error correction candidate items in the error correction candidate information.

The embodiment of the invention also provides a readable storage medium, which is characterized in that when the instructions in the storage medium are executed by a processor of the electronic device, the electronic device can execute the information error correction method according to the embodiment of the invention.

An embodiment of the present invention also provides an electronic device, including a memory, and one or more programs, where 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: after identifying that the input information needs to be corrected, determining error correction candidate information corresponding to the input information, wherein the error correction candidate information comprises: an error correction candidate and an error correction score for the error correction candidate; determining the complete probability of the sentence corresponding to the input information; adjusting the error correction scores of the error correction candidates according to the complete probability, and determining the correction scores of the error correction candidates; and after the correction score meets a preset condition, displaying the error correction candidate item.

Optionally, the error correction score is determined by inputting error correction candidates into a language model, and before the presenting of the error correction candidates, further comprising instructions for: judging whether the correction score is larger than a reference score of the input information, wherein the reference score is used for judging whether the input information has errors or not; and if the correction score is larger than the reference score, determining that the correction score meets a preset condition.

Optionally, after determining the complete probability of the sentence corresponding to the input information, instructions for: judging whether the integrity probability is larger than an integrity threshold value or not; if the integrity probability is greater than an integrity threshold, executing the step of adjusting the error correction score of the error correction candidate item according to the integrity probability; and if the integrity probability is smaller than the integrity threshold value, displaying the error correction candidate items in the error correction candidate information.

The embodiment of the invention has the following advantages:

After the input information is identified to need error correction, determining error correction candidates corresponding to the input information and error correction scores of the error correction candidates; when the sentence corresponding to the input information is incomplete, determining that the error judgment rate of the input information requiring error correction is higher, determining the complete probability of the sentence corresponding to the input information, determining the correction score of the error correction candidate item according to the complete probability and the error correction score, and determining whether to display the error correction candidate item according to the error correction score; and further, the error correction probability can be reduced. And if the correction score meets the preset condition, displaying the error correction candidate item, and if the correction score does not meet the preset condition, not displaying the error correction candidate item, so that invalid error correction is effectively avoided, and the error correction accuracy is improved.

Drawings

FIG. 1 is a schematic diagram of an input interface of the prior art;

FIG. 2 is a flow chart of steps of an embodiment of an information anti-miscorrection method of the present invention;

FIG. 3 is a flow chart of steps of an alternative embodiment of an information anti-miscorrection method of the present invention;

FIG. 4 is a schematic illustration of an input interface of the present invention;

FIG. 5 is a block diagram illustrating an embodiment of an information anti-miscorrection device of the present invention;

FIG. 6 is a block diagram of an alternative embodiment of an information anti-miscorrection apparatus of the present invention;

FIG. 7 is a block diagram illustrating an electronic device for information anti-miscorrection, according to an example embodiment;

fig. 8 is a schematic structural view of an electronic device for information anti-error correction according to another exemplary embodiment of the present invention.

Detailed Description

In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.

One of the core ideas of the invention is that, when the sentence corresponding to the input information is incomplete, the error judgment rate of the input information needing error correction is determined to be higher, so after the error correction candidate item and the corresponding error correction score are determined, the correction score of the error correction candidate item is determined according to the complete probability of the sentence corresponding to the input information and the error correction score of the error correction candidate item, and whether the error correction candidate item is displayed is determined according to the correction score; therefore, error correction can be prevented, and the error correction accuracy can be improved.

Referring to fig. 2, a step flow chart of an embodiment of an information error correction method of the present invention is shown, which specifically may include the following steps:

Step 202, after identifying that the input information needs error correction, determining error correction candidate information corresponding to the input information.

In the embodiment of the invention, the input information can be information of the screen, and of course, the input information can also be a candidate item of the area to be screen, and the input information can be specifically set according to the requirement; wherein the input information may include at least one word, which is not limited to language such as chinese, english, korean, japanese, etc., and may include single words and vocabulary such as chinese, i'm, etc., and "waiting" as vocabulary. After determining input information, the input information can be analyzed to identify whether the input information has errors, and if the input information is determined to have errors, that is, the input information needs error correction, the input information can be matched with corresponding error correction candidate information; if the input information is determined to have no error, that is, the input information is determined to have no error correction, the updated input information can be subjected to error correction recognition after the subsequently input information is determined. In the embodiment of the present invention, the error correction candidate information may include: an error correction candidate and an error correction score of the error correction candidate, wherein the error correction candidate refers to a word displayed in an error correction area of an input method, such as 'etc' in fig. 1, and the error correction area of the input method can be suspended on a current interface, such as a toolbar of a current application program; the correction score is determined by scoring the correction candidates according to a language model.

And 204, determining the complete probability of the sentence corresponding to the input information.

And 206, determining the correction scores of the error correction candidates according to the integrity probability and the error correction scores.

When a user inputs the first few words of a sentence, the words possibly input are correct, but the input method may determine that the words input by the user are incorrect, and display correction candidates; for example, in fig. 1, after the user wants to input the sentence "i am you QQ", the input method will display the error correction candidates "etc" because "i am you" are more consistent with the natural language law than "i am you"; however, the user does not need to correct the input words, so that the correction is meaningless and causes trouble to the user. After the user inputs the complete sentence 'I log on your QQ', the input method does not correct errors; when the sentence integrity is relatively low, the correct probability of the sentence error identified by the input method is likely to be low, and when the sentence integrity is relatively high, the correct probability of the sentence error identified by the input method is likely to be high; therefore, after determining that the input information needs to be corrected, the embodiment of the invention determines that the accuracy rate of the input information needs to be corrected may not be high, and in order to avoid displaying invalid correction candidates, the embodiment of the invention can determine whether the statement corresponding to the input information is complete, that is, determine the completeness of the statement corresponding to the input information, so as to determine whether to display the correction candidates according to the completeness of the statement corresponding to the input information.

The embodiment of the invention can adopt the complete probability to represent the complete degree of the sentence, so that the input information can be analyzed, such as analyzing words and the like at the tail of the input information, and the complete probability of the sentence corresponding to the input information can be determined; and determining the error correction score of the error correction candidate item according to the complete probability and the error correction score, if the error correction score of the error correction candidate item is adjusted by adopting the complete probability, determining the adjusted error correction score as the correction score of the error correction candidate item, and judging whether to display the error correction candidate item according to the correction score.

And step 208, displaying the error correction candidate items after the correction score meets a preset condition.

In the embodiment of the invention, preset conditions can be preset, the preset conditions are used for judging whether error correction candidates are displayed or not, further, after the correction scores of the error correction candidates are determined, whether the correction scores meet the preset conditions can be judged, and if the correction scores meet the preset conditions, the error correction candidates can be displayed in an error correction area of an input method.

Optionally, if it is determined that the correction score does not meet the preset condition, the error correction candidate is not required to be displayed in the error correction area, so that display of invalid error correction candidate is reduced, and user experience is improved.

After the input information is identified to need error correction, determining error correction candidates corresponding to the input information and error correction scores of the error correction candidates; when the sentence corresponding to the input information is incomplete, determining that the error judgment rate of the input information requiring error correction is higher, determining the complete probability of the sentence corresponding to the input information, determining the correction score of the error correction candidate item according to the complete probability and the error correction score, and determining whether to display the error correction candidate item according to the error correction score; and if the correction score meets the preset condition, displaying the error correction candidate item, and if the correction score does not meet the preset condition, not displaying the error correction candidate item, thereby effectively avoiding displaying invalid error correction and improving the error correction accuracy.

In another embodiment of the present invention, the complete probability of the sentence corresponding to the input information may be determined according to the sentence identification information corresponding to the input information, where the sentence identification information includes punctuation marks, end words, and the like; the method of determining the complete probability of the sentence corresponding to the input information is described in detail below.

Referring to fig. 3, a flowchart illustrating steps of an alternative embodiment of an information anti-error correction method of the present invention may specifically include the following steps:

step 302, inputting the input information into a language model, and determining a reference score of the input information.

The information in the application edit box may include a plurality of words, for example, "weather today is good," and the input method may use all the words in the application edit box as input information, for example, "weather today is good," and may also use some of the words as input information, for example, "we go," and then identify whether the input information has errors. When identifying whether the input information has errors, the input information can be input into a language model, and the language model is adopted to score the input information, so that the reference score of the input information is determined; the language model is built based on natural language, and can be used for scoring the input information to determine the fluency of the sentence corresponding to the input information, and can include multiple categories, for example: may be NGram language model, neural network language model, etc. Specifically, word segmentation processing can be performed on the input information, the input information is split into word segments, then a language model is adopted to score a word segment sequence corresponding to the input information, and then a reference score of the input information is obtained through calculation.

The embodiment of the invention can preset an error correction threshold value to determine whether the input information has errors or not by comparing the reference score of the input information with the error correction threshold value, wherein the error correction threshold value can be set according to requirements.

Step 304, judging whether the reference score is smaller than an error correction threshold value.

In the embodiment of the invention, the reference score of the input information and the error correction threshold value can be compared to determine whether the input information has errors or not; specifically, whether the reference score is smaller than an error correction threshold value or not may be determined, if the reference score is smaller than the error correction threshold value, it is determined that the input information has an error, that is, the input information needs error correction, and step 306 may be executed; if the reference score is greater than the error correction threshold, it is determined that there is no error in the input information, i.e., the input information does not need error correction, step 322 is performed.

Step 306, determining error correction candidate information corresponding to the input information.

After determining that the input information needs to be corrected, the embodiment of the invention can match the correction candidate corresponding to the input information from a word stock, then screen one or more correction candidates from the matched correction candidate, and determine the correction candidate information according to the screened correction candidate; and when the error correction candidates corresponding to the input information are matched, scoring each error correction candidate, and determining the error correction score of each error correction candidate so as to screen according to the error correction score. An optional screening method of the embodiment of the invention is that according to the error correction scores of the error correction candidates, searching the error correction candidate with the highest error correction score, and judging whether the highest error correction score is larger than a reference score; if the highest error correction score is larger than the reference score, determining that error correction candidates for error correction exist, and generating error correction candidate information corresponding to the input information by adopting the error correction candidate with the highest error correction score; if the highest error correction score is less than the reference score, it is determined that no error correction candidates exist for error correction, step 322 may be performed.

In the embodiment of the invention, in order to prevent error correction, the error correction score of the error correction candidate item can be adjusted by adopting the complete probability of the statement corresponding to the input information to obtain the correction score; and further judging whether to display the error correction candidate according to the correction score, wherein the error correction candidate is specifically as follows:

And 308, obtaining statement identification information according to the input information, and determining the complete probability of the corresponding statement according to the statement identification information.

According to the embodiment of the invention, corresponding sentence identification information can be obtained according to the input information, namely, the input information is analyzed to determine the sentence identification information corresponding to the input information, wherein the sentence identification information can be information for determining the integrity degree of the input information; wherein the sentence identification information includes at least one of: punctuation marks, sentence tail words, association information corresponding to input information and input intervals corresponding to the input information; of course, the statement identification information may also include other information, which is not illustrated herein.

After determining statement identification information, the embodiment of the invention can determine the complete probability of the statement corresponding to the input information according to at least one statement identification information; the following describes in detail a method for determining the complete probability of the sentence corresponding to the input information according to different sentence identification information, specifically as follows:

1. Determining the complete probability of the sentence corresponding to the input information according to punctuation marks, wherein the method specifically comprises the following substeps:

and S11, acquiring punctuation marks at the tail of the input information.

And step S12, matching the punctuation marks with the set punctuation marks.

And a substep S13, determining the first numerical value as the complete probability.

And a substep S14, determining a second value as the complete probability.

Punctuation marks are symbols of auxiliary literal record language and are used for representing pauses, mood and the nature and effect of words; therefore, whether the sentence is complete can be determined according to the punctuation marks at the end of the sentence, and in the embodiment of the invention, the set punctuation marks can be determined in advance according to the characteristics of the punctuation marks, for example, the punctuation marks with sentence breaking effect are determined as the set punctuation marks, such as ",". ", I! "? "and the like. Therefore, when the complete probability is determined according to the punctuation marks, the punctuation marks at the tail of the input information can be obtained, and then the punctuation marks are matched with the set punctuation marks; if the punctuation mark is matched with the set punctuation mark, determining that the completeness of the sentence corresponding to the input information is higher, namely executing a substep S13; if the punctuation mark is not matched with the set punctuation mark, determining that the completeness of the sentence corresponding to the input information is lower, and executing the substep S14.

In the embodiment of the invention, if the punctuation mark is determined to be matched with the set punctuation mark, the first numerical value can be determined to be the complete probability, and if the punctuation mark is determined to be not matched with the set punctuation mark, the second numerical value can be determined to be the complete probability; wherein, the first numerical value and the second numerical value are set according to the requirement. For example, if the punctuation mark at the end of the input information is. The integrity probability can be determined to be 1, and the punctuation mark at the end of the input information is determined to be 'to' and the integrity probability can be determined to be 0.4.

2. Determining the complete probability of the sentence corresponding to the input information according to the sentence tail word, wherein the method specifically comprises the following substeps:

and a substep S21, identifying the sentence tail words from the input information.

And S22, matching the sentence tail words with the set identification words.

And S23, determining the complete probability according to the end-of-sentence probability of the set identification word matched with the end-of-sentence word.

And a substep S24, determining a third numerical value as the complete probability.

Certain words often appear at the end of a sentence, such as "do", "prama", "o", "have", etc., and when the words at the end of a sentence are those words, the sentence may be considered likely to be complete; therefore, the embodiment of the invention can determine whether the input information is complete or not according to the sentence tail words corresponding to the input information. The probability of each word as the sentence end, namely the sentence end probability, can be counted from the large-scale training corpus in advance; then, determining words with sentence end probability larger than a set threshold value as set identification words, wherein the set threshold value can be set according to requirements; for example, if the threshold is set to 0.7, the end of sentence probability of "having" is 0.99, the end of sentence probability of "having" is 0.8, and the end of sentence probability of "having" is 0.3, it is possible to determine that "having" and "o" are set as the identification words. When the embodiment of the invention determines the complete probability, the input information can be identified, and the sentence tail words are identified from the input information; then, matching the sentence tail words with the set identification words, if the sentence tail words are matched with the set identification words, determining that the sentence tail words are words frequently appearing at the tail of the sentence, determining that the completeness degree of the sentence corresponding to the input information is higher, and executing the substep S23; if the sentence tail word is not matched with the set identification word, determining that the sentence tail word is a word frequently appearing at the end of the sentence, determining that the completeness degree of the sentence corresponding to the input information is lower, and executing the substep S24.

In the embodiment of the invention, if the sentence tail word is matched with the set identification word, the complete probability can be determined according to the sentence tail probability of the set identification word matched with the sentence tail word, wherein the sentence tail probability is the probability of the set identification word as the sentence tail of the sentence. Specifically, if the integrity probability is proportional to the integrity degree of the sentence, the end probability of the set identification word matched with the end word of the sentence can be determined as the integrity probability, for example, the probability of being used as the end of the sentence is 0.87, and the integrity probability can be 0.87. If the completeness probability is inversely proportional to the completeness degree of the sentence, determining the difference value between the end probability of the set identification word matched with the end word and1 as the completeness probability; for example, the probability of "having" as the end of sentence is 0.87, and the probability of complete may be 0.13. And if the sentence tail words are not matched with the set identification words, determining a third numerical value as the complete probability, wherein the third data are set according to requirements.

3. Determining the complete probability of the sentence corresponding to the input information according to the association information corresponding to the input information, wherein the method specifically comprises the following substeps:

And a substep 31 of determining corresponding association information according to the input information, wherein the association information comprises association words and association probabilities of the association words.

Substep 32, determining the total number of associated words and the maximum associated probability, and calculating the ratio of the total number to a set value.

And a substep 33, determining the complete probability according to the ratio and the maximum association probability.

In the embodiment of the invention, the input method also comprises an association function, namely, based on the information input by the user, the information to be input by the user is predicted, and the corresponding association words are determined; if more associated words are determined based on the input information of the user, the user can be considered to be likely to continue to input at the moment, namely sentences corresponding to the input information are likely to be incomplete; if the number of the determined associated words is small based on the input information of the user, the user can be considered that the user is likely not to continue to input, namely the sentence corresponding to the input information is likely to be complete; the complete probability can be determined from the input information corresponding to the associated word.

The method comprises the steps of determining that input information is subjected to word segmentation processing to obtain word fragments, and then carrying out association according to all the word fragments to determine corresponding association information; of course, association can be performed according to the word segment at the end of the input information, and corresponding association information can be determined; wherein the association information includes an association word and an association probability of the association word. Then determining the total number and the maximum association probability of the associated words according to the association information, calculating the ratio of the total number to a set value, and determining the complete probability according to the ratio and the maximum association probability; wherein, the set value can be set as 200 according to the requirement.

In another embodiment of the present invention, the ratio and the maximum association probability may be compared, a maximum value of the ratio and the maximum association probability is determined, and then the maximum value is determined to be the complete probability; the completeness probability determined according to the association information is inversely proportional to the completeness of the sentence, namely, the higher the completeness probability is, the lower the completeness of the sentence is. For example: if the higher the completeness probability is, the more incomplete the sentence is, the input information is 180 associated words in tomorrow, the maximum association probability is 0.13, and when the set value is 200, the completeness probability=max (180/200,0.13) =0.9.

4. Determining the complete probability of the corresponding sentence according to the input interval corresponding to the input information, wherein the method specifically comprises the following substeps:

substep 41, determining the input information of the upper screen and the input interval of the next input operation.

Substep 42, determining whether the input interval is greater than an average input interval.

Substep 43, determining a fourth value as the integrity probability.

Substep 44, determining a fifth value as the integrity probability.

The input interval of two adjacent words in one sentence input by a user is relatively smaller than the input interval of two adjacent sentences input by the user; for example, the input information input by the user is "I log in you", and the interval of the subsequent input of "QQ" is smaller than the interval of the subsequent input of "you get slower; therefore, the input interval between the input information and the next input operation on the screen can be determined, wherein the next input operation refers to the first input operation on the input method keyboard after the screen operation, such as: after a user screens 'I log in you', typing a 'Q' operation in an input keyboard, wherein the time for the 'I log in you' on the upper screen is 12:20:33, the time for typing in 'Q' is 12:20:34, and the input interval is 1 second; and then determining whether the input information is complete according to the input interval. In the embodiment of the invention, the average input interval of the user can be determined in advance according to the history record, then the input information of the screen and the input interval of the next input operation are determined, and further the complete probability is determined according to the input interval corresponding to the input information and the average input interval.

Judging whether the input interval is larger than the average input interval, if the input interval is larger than the average input interval, determining that the completeness of the sentence corresponding to the input information is higher, and executing the sub-step 43; if the input interval is smaller than the average input interval, the user is likely to continue inputting, and it is determined that the completeness of the sentence corresponding to the input information is low, that is, sub-step 44 is performed. In the embodiment of the present invention, if the input interval is greater than the average input interval, the fourth value may be determined as the complete probability; if the input interval is less than the average input interval, a fifth value may be determined as the probability of completeness. Wherein, the fourth numerical value and the fifth numerical value are set according to the requirement.

In another embodiment of the present invention, the above-mentioned arbitrary methods may be used to determine the complete probability, and after each method is used to determine the corresponding complete probability, a weighted calculation may be performed on the multiple complete probabilities to obtain the final complete probability of the sentence corresponding to the input information, where weights of the complete probabilities corresponding to the various methods may be set according to requirements, so as to further improve accuracy of the complete probability, and reduce the error correction probability.

Step 310, judging whether the integrity probability is greater than an integrity threshold.

If the integrity probability is inversely proportional to the integrity degree of the sentence, judging whether the integrity probability is greater than an integrity threshold, if the integrity probability is greater than the integrity threshold, determining that the sentence corresponding to the input information is incomplete, and executing step 312; if the integrity probability is less than the integrity threshold, it is determined that the input information corresponding sentence is intact, step 318 is performed.

If the integrity probability is proportional to the integrity degree of the sentence, judging whether the integrity probability is greater than an integrity threshold value, if the integrity probability is greater than the integrity threshold value, determining that the sentence corresponding to the input information is complete, and executing step 318; if the integrity probability is less than the integrity threshold, it is determined that the input information corresponding sentence is incomplete, step 312 is performed.

After determining that the sentence corresponding to the input information is incomplete, adjusting the error correction score of the error correction candidate item according to the complete probability, and determining the correction score of the error correction candidate item, wherein the correction score is specifically as follows:

Step 312, determining a penalty score according to the integrity probability and the penalty weight.

And step 314, adjusting the error correction score of the error correction candidate item by adopting the penalty score to determine the correction score.

In the embodiment of the invention, because the sentence corresponding to the input information may be incomplete, the accuracy rate of determining that the input information needs to be corrected is low, and therefore the correction value is adjusted by adopting the complete probability of the sentence corresponding to the input information; specifically, penalty weights corresponding to the complete probabilities can be predetermined, so that scores corresponding to the complete probabilities, namely penalty scores, are determined according to the penalty weights, further, error correction scores of error correction candidates are penalized according to the penalty scores, and whether the error correction candidates are displayed or not is determined according to the penalized error correction scores, so that the error correction accuracy is improved; wherein, the punishment weight can be set according to the requirement. In the embodiment of the invention, an alternative way of determining the penalty value is that if the integrity probability is inversely proportional to the integrity degree of the sentence, the product of the integrity probability and the penalty weight can be calculated; alternatively, if the integrity probability is proportional to the integrity degree of the sentence, the difference between the integrity probability and 1 may be calculated, and then the product of the difference and the penalty weight may be calculated; the product of the two is then determined as the penalty score. Then adopting the punishment score to adjust the error correction score of the error correction candidate item, and determining the correction score; alternatively, a difference between the penalty score and the error correction score may be calculated, and the difference is determined as the correction score of the error correction candidate.

Step 316, determining whether the corrected score is greater than a reference score of the input information.

After determining the error correction score of the error correction candidate item, the embodiment of the invention can compare the correction score of the error correction candidate item with the reference score corresponding to the input information, judge whether the correction score is greater than the reference score of the input information, if the correction score is greater than the reference score, determine that the correction score meets the preset condition, namely determine that the input information needs to be subjected to error correction by adopting the error correction candidate item, and execute step 318; if the correction score is less than the reference score, it is determined that the correction score does not satisfy the preset condition, that is, it is determined that the input information does not need to be corrected by using the error correction candidate, step 320 may be performed.

And step 318, displaying the error correction candidates.

And if the correction score meets a preset condition or the input information corresponding statement is determined to be complete, displaying the error correction candidate item.

Step 320, not displaying the error correction candidates.

And if the correction score does not meet the preset condition, not displaying the error correction candidate.

Step 322, identify whether the updated input information requires error correction.

If the input information is determined to be unnecessary to correct the error, or if the input information is determined to not have the error correction candidate, the subsequent input information of the user is determined, the updated input information is identified, and whether the updated input information needs to correct the error is judged.

In one example of the present invention, if the user inputs the input information "i don you" as shown in fig. 1 in the input interface, and the corresponding reference score is 240, which is lower than the error correction threshold 400, the input information is identified that error correction is needed. Then determining that the error correction candidate is equal, the corresponding error correction score is 500, the complete probability of 'I log you' is 0.88, the penalty weight is 300, and calculating the penalty score as 264 and then calculating the difference between the penalty score and the error correction score as 236, namely, the correction score as 236. If the correction score is smaller than the reference score, the error correction candidate "etc" is not displayed, as shown in fig. 4.

In summary, the embodiment of the invention adopts the complete probability of the sentence corresponding to the input information to adjust the error correction score of the error correction candidate item and determine the corresponding correction score; further, whether the input information is corrected or not is determined according to the correction score; when the corresponding sentence of the input information is incomplete, the error judgment rate of the input information needing error correction is determined to be higher, so that whether error correction candidates are displayed or not is determined according to the correction score, and the error correction probability can be reduced; and if the correction score meets the preset condition, displaying the error correction candidate item, and if the correction score does not meet the preset condition, not displaying the error correction candidate item, so that the display of invalid error correction is effectively avoided, the error correction accuracy is improved, and the user experience is also improved.

Further, the embodiment of the invention can also determine the complete probability of the corresponding sentence according to the sentence identification information corresponding to the input information, wherein the sentence identification information comprises at least one of the following: punctuation marks, sentence tail words, association information corresponding to input information and input intervals corresponding to the input information; if the integrity probability is determined by adopting a plurality of methods, the accuracy of determining the integrity probability can be improved, thereby more effectively avoiding displaying invalid error correction and further improving the accuracy of error correction.

It should be noted that, for simplicity of description, the method embodiments are shown as a series of acts, but it should be understood by those skilled in the art that the embodiments are not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the embodiments. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred embodiments, and that the acts are not necessarily required by the embodiments of the invention.

Referring to fig. 5, a block diagram of an embodiment of an information error correction device according to the present invention is shown, and may specifically include the following modules: an information determination module 51, a probability determination module 52, a score determination module 53, and a presentation module 54, wherein,

The information determining module 51 is configured to determine error correction candidate information corresponding to input information after identifying that the input information needs error correction, where the error correction candidate information includes: an error correction candidate and an error correction score for the error correction candidate;

A probability determining module 52, configured to determine a complete probability of the sentence corresponding to the input information;

A score determining module 53, configured to determine a correction score of the error correction candidate according to the integrity probability and the error correction score;

and the display module 54 is configured to display the error correction candidate item after the correction score meets a preset condition.

Referring to fig. 6, there is shown a block diagram of an alternative embodiment of an information anti-miscorrection device of the present invention, the device further comprising: a score judgment module 55 and a threshold judgment module 56, wherein,

A score judgment module 55 for judging whether the corrected score is greater than a reference score of the input information, the reference score being used for judging whether the input information is erroneous; if the correction score is larger than the reference score, determining that the correction score meets a preset condition; wherein the correction score is determined by inputting correction candidates into a language model.

A threshold value judging module 56, configured to judge whether the integrity probability is greater than an integrity threshold value; if the integrity probability is greater than an integrity threshold, executing the step of adjusting the error correction score of the error correction candidate item according to the integrity probability; and if the integrity probability is smaller than the integrity threshold value, displaying the error correction candidate items in the error correction candidate information.

In another embodiment of the present invention, the information determining module 51 is specifically configured to input the input information into a language model, and determine a reference score of the input information; and if the reference score is smaller than the error correction threshold value, determining that the input information needs error correction.

In another embodiment of the present invention, the probability determining module 52 is specifically configured to obtain statement identification information according to the input information, and determine the complete probability of the corresponding statement according to the statement identification information, where the statement identification information includes at least one of the following: punctuation marks, sentence tail words, association information corresponding to input information and input intervals corresponding to the input information.

In another embodiment of the present invention, the probability determination module 52 includes: a first determination sub-module 521, a second determination sub-module 522, a third determination sub-module 523, and a fourth determination sub-module 524, wherein,

A first determining sub-module 521, configured to obtain a punctuation mark at the end of the input information; matching the punctuation mark with a set punctuation mark; if the punctuation mark is matched with the set punctuation mark, determining a first numerical value as the complete probability; and if the punctuation mark is not matched with the set punctuation mark, determining a second numerical value as the complete probability.

A second determining submodule 522, configured to identify a sentence-end word from the input information; matching the sentence tail words with set identification words; if the sentence tail word is matched with the set identification word, determining the complete probability according to the sentence tail probability of the set identification word matched with the sentence tail word, wherein the sentence tail probability is the probability of taking the set identification word as the sentence tail of the sentence; and if the sentence tail word is not matched with the set identification word, determining a third numerical value as the complete probability.

A third determining sub-module 523, configured to determine corresponding association information according to the input information, where the association information includes an association word and an association probability of the association word; determining the total number of the associated words and the maximum association probability, and calculating the ratio of the total number to a set value; and determining the complete probability according to the ratio and the maximum association probability.

The third determining submodule 523 is configured to determine a maximum value of the ratio and the maximum association probability, and determine the maximum value as the complete probability.

A fourth determining sub-module 524, configured to determine an input interval between the input information and a next input operation; judging whether the input interval is larger than an average input interval or not; if the input interval is greater than the average input interval, determining a fourth value as the complete probability; and if the input interval is smaller than the average input interval, determining a fifth value as the complete probability.

In another embodiment of the present invention, the score determining module 53 is configured to determine a penalty score according to the completeness probability and the penalty weight; and adjusting the error correction score of the error correction candidate item by adopting the penalty score, and determining the correction score.

After the input information is identified to need error correction, determining error correction candidates corresponding to the input information and error correction scores of the error correction candidates; when the sentence corresponding to the input information is incomplete, determining that the error judgment rate of the input information requiring error correction is higher, determining the complete probability of the sentence corresponding to the input information, adjusting the error correction score according to the complete probability, determining the correction score of the error correction candidate item, and determining whether to display the error correction candidate item according to the error correction score; and further, the error correction probability can be reduced. And if the correction score meets the preset condition, displaying the error correction candidate item, and if the correction score does not meet the preset condition, not displaying the error correction candidate item, thereby effectively avoiding displaying invalid error correction and improving the error correction accuracy.

For the device embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments for relevant points.

Fig. 7 illustrates a block diagram of an electronic device 700 for information anti-miscorrection, according to an example embodiment. For example, the electronic device 700 may be a mobile phone, computer, digital broadcast terminal, messaging device, game console, tablet device, medical device, exercise device, personal digital assistant, or the like.

Referring to fig. 7, an electronic device 700 may include one or more of the following components: a processing component 702, a memory 704, a power component 706, a multimedia component 708, an audio component 710, an input/output (I/O) interface 712, a sensor component 714, and a communication component 716.

The processing component 702 generally controls overall operation of the electronic device 700, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing element 702 may include one or more processors 720 to execute instructions to perform all or part of the steps of the methods described above. Further, the processing component 702 can include one or more modules that facilitate interaction between the processing component 702 and other components. For example, the processing component 702 may include a multimedia module to facilitate interaction between the multimedia component 708 and the processing component 702.

Memory 704 is configured to store various types of data to support operations at device 700. Examples of such data include instructions for any application or method operating on the electronic device 700, contact data, phonebook data, messages, pictures, videos, and so forth. The memory 704 may be implemented by any type or combination of volatile or nonvolatile 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 disk.

The power component 706 provides power to the various components of the electronic device 700. Power component 704 can include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for electronic device 700.

The multimedia component 708 includes a screen between the electronic device 700 and the user that provides an output interface. 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 input signals from a user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensor may sense not only the boundary of a touch or slide action, but also the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 708 includes a front-facing camera and/or a rear-facing camera. When the electronic device 700 is in an operational mode, such as a shooting mode or a video mode, the front camera and/or the rear camera may receive external multimedia data. Each front camera and rear camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

The audio component 710 is configured to output and/or input audio signals. For example, the audio component 710 includes a Microphone (MIC) configured to receive external audio signals when the electronic device 700 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may be further stored in the memory 704 or transmitted via the communication component 716. In some embodiments, the audio component 710 further includes a speaker for outputting audio signals.

The I/O interface 712 provides an interface between the processing component 702 and peripheral interface modules, which may be a keyboard, click wheel, buttons, etc. These buttons may include, but are not limited to: homepage button, volume button, start button, and lock button.

The sensor assembly 714 includes one or more sensors for providing status assessment of various aspects of the electronic device 700. For example, the sensor assembly 714 may detect an on/off state of the device 700, a relative positioning of the components, such as a display and keypad of the electronic device 700, a change in position of the electronic device 700 or a component of the electronic device 700, the presence or absence of a user's contact with the electronic device 700, an orientation or acceleration/deceleration of the electronic device 700, and a change in temperature of the electronic device 700. The sensor assembly 714 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 714 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 714 may also include an acceleration sensor, a gyroscopic sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

The communication component 716 is configured to facilitate communication between the electronic device 700 and other devices, either wired or wireless. The electronic device 700 may access a wireless network based on a communication standard, such as WiFi,2G, or 3G, or a combination thereof. In one exemplary embodiment, the communication part 714 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, the communication component 714 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 electronic device 700 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, microcontrollers, microprocessors, or other electronic elements for executing the methods described above.

In an exemplary embodiment, a non-transitory computer readable storage medium is also provided, such as memory 704, including instructions executable by processor 720 of electronic device 700 to perform the above-described method. For example, the non-transitory computer readable storage medium may be ROM, random Access Memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.

A non-transitory computer readable storage medium, which when executed by a processor of an electronic device, causes the electronic device to perform an information anti-miscorrection method, the method comprising: after identifying that the input information needs to be corrected, determining error correction candidate information corresponding to the input information, wherein the error correction candidate information comprises: an error correction candidate and an error correction score for the error correction candidate; determining the complete probability of the sentence corresponding to the input information; determining correction scores of the error correction candidates according to the integrity probability and the error correction scores; and after the correction score meets a preset condition, displaying the error correction candidate item.

Optionally, the determining the complete probability according to the ratio and the maximum association probability includes: and determining the maximum value of the ratio and the maximum association probability, and determining the maximum value as the incomplete probability.

Optionally, adjusting the error correction score of the error correction candidate according to the integrity probability, and determining the correction score of the error correction candidate includes: determining a penalty score according to the integrity probability and the penalty weight; and adjusting the error correction score of the error correction candidate item by adopting the penalty score, and determining the correction score.

Optionally, the error correction score is determined by inputting error correction candidates into a language model, and before the presenting of the error correction candidates, the method further comprises: judging whether the correction score is larger than a reference score of the input information, wherein the reference score is used for judging whether the input information has errors or not; and if the correction score is larger than the reference score, determining that the correction score meets a preset condition.

Fig. 8 is a schematic structural view of an electronic device 800 for navigation according to another exemplary embodiment of the present invention. The electronic device 800 may be a server that may vary widely in configuration or performance and may include one or more central processing units (central processing units, CPUs) 822 (e.g., one or more processors) and memory 832, one or more storage mediums 830 (e.g., one or more mass storage devices) that store applications 842 or data 844. Wherein the memory 832 and the storage medium 830 may be transitory or persistent. The program stored in the storage medium 830 may include one or more modules (not shown), each of which may include a series of instruction operations on a server. Still further, the central processor 822 may be configured to communicate with the storage medium 830 and execute a series of instruction operations in the storage medium 830 on a server.

The server(s) may also include one or more power supplies 826, one or more wired or wireless network interfaces 850, one or more input/output interfaces 858, one or more keyboards 856, and/or one or more operating systems 841 such as Windows Server, mac OS XTM, unixTM, linuxTM, freeBSDTM, and the like.

An electronic device 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 comprising instructions for: after identifying that the input information needs to be corrected, determining error correction candidate information corresponding to the input information, wherein the error correction candidate information comprises: an error correction candidate and an error correction score for the error correction candidate; determining the complete probability of the sentence corresponding to the input information; adjusting the error correction scores of the error correction candidates according to the complete probability, and determining the correction scores of the error correction candidates; and after the correction score meets a preset condition, displaying the error correction candidate item.

In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described by differences from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other.

It will be apparent to those skilled in the art that embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the invention may take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the scope of the embodiments of the invention.

Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or terminal device that comprises the element.

The above detailed description of an information error correction method, an information error correction device and an electronic device provided by the present invention applies specific examples to illustrate the principles and embodiments of the present invention, and the above description of the examples is only used to help understand the method and core ideas of the present invention; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present invention, the present description should not be construed as limiting the present invention in view of the above.

Claims

1. An information error correction method, comprising:

After identifying that the input information needs to be corrected, determining error correction candidate information corresponding to the input information, wherein the error correction candidate information comprises: an error correction candidate and an error correction score for the error correction candidate;

The complete probability of the sentence corresponding to the input information is determined, specifically: obtaining statement identification information according to the input information, and determining the complete probability of a corresponding statement according to the statement identification information;

and determining the correction score of the error correction candidate item according to the integrity probability and the error correction score, wherein the correction score is specifically: determining a penalty value according to the complete probability and the penalty weight corresponding to the complete probability, adjusting the error correction value of the error correction candidate item by adopting the penalty value, and determining the correction value; wherein, the punishment weight is set according to the requirement;

And after the correction score meets a preset condition, displaying the error correction candidate item.

2. The method of claim 1, wherein the identifying input information requires error correction, comprising:

inputting the input information into a language model, and determining a reference score of the input information;

And if the reference score is smaller than the error correction threshold value, determining that the input information needs error correction.

3. The method of claim 1, wherein the statement identification information comprises at least one of: punctuation marks, sentence tail words, association information corresponding to input information and input intervals corresponding to the input information.

4. A method according to claim 3, wherein analyzing sentence identification information based on the input information and determining the complete probability of the corresponding sentence based on the sentence identification information comprises:

Acquiring punctuation marks at the tail of the input information;

Matching the punctuation mark with a set punctuation mark;

if the punctuation mark is matched with the set punctuation mark, determining a first numerical value as the complete probability;

and if the punctuation mark is not matched with the set punctuation mark, determining a second numerical value as the complete probability.

5. A method according to claim 3, wherein analyzing sentence identification information based on the input information and determining the complete probability of the corresponding sentence based on the sentence identification information comprises:

Identifying sentence tail words from the input information;

matching the sentence tail words with set identification words;

If the sentence tail word is matched with the set identification word, determining the complete probability according to the sentence tail probability of the set identification word matched with the sentence tail word, wherein the sentence tail probability is the probability of taking the set identification word as the sentence tail of the sentence;

And if the sentence tail word is not matched with the set identification word, determining a third numerical value as the complete probability.

6. A method according to claim 3, wherein analyzing sentence identification information based on the input information and determining the complete probability of the corresponding sentence based on the sentence identification information comprises:

Determining corresponding association information according to the input information, wherein the association information comprises association words and association probability of the association words;

Determining the total number of the associated words and the maximum association probability, and calculating the ratio of the total number to a set value;

And determining the complete probability according to the ratio and the maximum association probability.

7. The method of claim 6, wherein said determining said integrity probability based on said ratio and a maximum association probability comprises:

And determining the maximum value of the ratio and the maximum association probability, and determining the maximum value as the complete probability.

8. A method according to claim 3, wherein analyzing sentence identification information based on the input information and determining the complete probability of the corresponding sentence based on the sentence identification information comprises:

Determining the input information of the upper screen and the input interval of the next input operation;

Judging whether the input interval is larger than an average input interval or not;

if the input interval is greater than the average input interval, determining a fourth value as the complete probability;

and if the input interval is smaller than the average input interval, determining a fifth value as the complete probability.

9. The method of claim 2, wherein the correction score is determined by inputting correction candidates into a language model, and further comprising, prior to the presenting the correction candidates:

Judging whether the correction score is larger than a reference score of the input information, wherein the reference score is used for judging whether the input information has errors or not;

And if the correction score is larger than the reference score, determining that the correction score meets a preset condition.

10. The method of claim 1, further comprising, after determining the complete probability of the statement corresponding to the input information:

Judging whether the integrity probability is larger than an integrity threshold value or not;

If the integrity probability is greater than the integrity threshold, executing the step of adjusting the error correction score of the error correction candidate item according to the integrity probability and determining the correction score of the error correction candidate item;

and if the integrity probability is smaller than the integrity threshold value, displaying the error correction candidate items in the error correction candidate information.

11. An information error correction device, comprising:

The information determining module is used for determining error correction candidate information corresponding to the input information after identifying that the input information needs error correction, and the error correction candidate information comprises: an error correction candidate and an error correction score for the error correction candidate;

the probability determining module is configured to determine a complete probability of the sentence corresponding to the input information, specifically: obtaining statement identification information according to the input information, and determining the complete probability of a corresponding statement according to the statement identification information;

The score determining module is configured to determine, according to the integrity probability and the error correction score, a correction score of the error correction candidate, specifically: determining a penalty value according to the complete probability and the penalty weight corresponding to the complete probability, adjusting the error correction value of the error correction candidate item by adopting the penalty value, and determining the correction value; wherein, the punishment weight is set according to the requirement;

And the display module is used for displaying the error correction candidate items after the correction score meets a preset condition.

12. The apparatus of claim 11, wherein the device comprises a plurality of sensors,

The information determining module is specifically configured to input the input information into a language model, and determine a reference score of the input information; and if the reference score is smaller than the error correction threshold value, determining that the input information needs error correction.

13. The apparatus of claim 11, wherein the statement identification information comprises at least one of: punctuation marks, sentence tail words, association information corresponding to input information and input intervals corresponding to the input information.

14. The apparatus of claim 13, wherein the probability determination module comprises:

the first determining submodule is used for acquiring punctuation marks after the input information; matching the punctuation mark with a set punctuation mark; if the punctuation mark is matched with the set punctuation mark, determining a first numerical value as the complete probability; and if the punctuation mark is not matched with the set punctuation mark, determining a second numerical value as the complete probability.

15. The apparatus of claim 13, wherein the probability determination module comprises:

The second determining submodule is used for identifying sentence tail words from the input information; matching the sentence tail words with set identification words; if the sentence tail word is matched with the set identification word, determining the complete probability according to the sentence tail probability of the set identification word matched with the sentence tail word, wherein the sentence tail probability is the probability of taking the set identification word as the sentence tail of the sentence; and if the sentence tail word is not matched with the set identification word, determining a third numerical value as the complete probability.

16. The apparatus of claim 13, wherein the probability determination module comprises:

A third determining sub-module, configured to determine corresponding association information according to the input information, where the association information includes an association word and an association probability of the association word; determining the total number of the associated words and the maximum association probability, and calculating the ratio of the total number to a set value; and determining the complete probability according to the ratio and the maximum association probability.

17. The apparatus of claim 16, wherein the device comprises a plurality of sensors,

And the third determining submodule is used for determining the maximum value of the ratio and the maximum association probability and determining the maximum value as the complete probability.

18. The apparatus of claim 13, wherein the probability determination module comprises:

a fourth determining submodule for determining an input interval of the input information and the subsequent input information; judging whether the input interval is larger than an average input interval or not; if the input interval is greater than the average input interval, determining a fourth value as the complete probability; and if the input interval is smaller than the average input interval, determining a fifth value as the complete probability.

19. The apparatus of claim 11, wherein the correction score is determined by inputting correction candidates into a language model, further comprising:

The score judgment module is used for judging whether the correction score is larger than a reference score of the input information, and the reference score is used for judging whether the input information is wrong or not; and if the correction score is larger than the reference score, determining that the correction score meets a preset condition.

20. The apparatus as recited in claim 11, further comprising:

The threshold judging module is used for judging whether the integrity probability is larger than an integrity threshold or not; if the integrity probability is greater than the integrity threshold, executing the step of adjusting the error correction score of the error correction candidate item according to the integrity probability and determining the correction score of the error correction candidate item; and if the integrity probability is smaller than the integrity threshold value, displaying the error correction candidate items in the error correction candidate information.

21. A readable storage medium, characterized in that instructions in said storage medium, when executed by a processor of an electronic device, enable the electronic device to perform the information anti-miscorrection method according to any one of the method claims 1-10.

22. An electronic device 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 comprising instructions for:

And adjusting the error correction scores of the error correction candidates according to the complete probability, and determining the correction scores of the error correction candidates, wherein the correction scores are specifically: determining a penalty value according to the complete probability and the penalty weight corresponding to the complete probability, adjusting the error correction value of the error correction candidate item by adopting the penalty value, and determining the correction value; wherein, the punishment weight is set according to the requirement;

23. The electronic device of claim 22, wherein the identification input information requires error correction, comprising:

24. The electronic device of claim 22, wherein: the sentence identification information includes at least one of: punctuation marks, sentence tail words, association information corresponding to input information and input intervals corresponding to the input information.

25. The electronic device of claim 24, wherein analyzing the sentence identification information based on the input information and determining the complete probability of the corresponding sentence based on the sentence identification information comprises:

Acquiring punctuation marks at the tail of the input information;

Matching the punctuation mark with a set punctuation mark;

26. The electronic device of claim 24, wherein analyzing the sentence identification information based on the input information and determining the complete probability of the corresponding sentence based on the sentence identification information comprises:

Identifying sentence tail words from the input information;

matching the sentence tail words with set identification words;

27. The electronic device of claim 24, wherein analyzing the sentence identification information based on the input information and determining the complete probability of the corresponding sentence based on the sentence identification information comprises:

28. The electronic device of claim 27, wherein said determining said integrity probability based on said ratio and a maximum association probability comprises:

29. The electronic device of claim 24, wherein analyzing the sentence identification information based on the input information and determining the complete probability of the corresponding sentence based on the sentence identification information comprises:

30. The electronic device of claim 23, wherein the correction score is determined by inputting correction candidates into a language model, and further comprising instructions for, prior to the presenting the correction candidates:

31. The electronic device of claim 22, further comprising instructions for, after determining the probability of completeness of the statement corresponding to the input information: