KR101349967B1 - Method of Improving Logic to Propose Query for Mobile Keyboard Typo Pattern and the Device Thereof - Google Patents

Abstract

Disclosed are a method and apparatus for calibrating a user query including a duplicate character based on statistical data. The user query correction method may include determining whether an input user query is a typo query, extracting at least one calibration candidate string when the user query is a typo query, and calculating candidate scores for the at least one calibration candidate string. And weighting a string corresponding to a case in which the first character or the back character of the character to be corrected among the correction candidate strings is the same as the character to be corrected, and determining a correction query among the correction candidate strings according to candidate scores and weights. can do.

Description

Method of Improving Logic to Propose Query for Mobile Keyboard Typo Pattern and the Device Thereof}

The present invention relates to a method and apparatus for correcting a user query based on statistical data, and more particularly, to a method and apparatus for correcting a user query determined as a typo query, including overlapping syllables or letters.

The user can perform a search to get the desired information. In this case, the user may perform a search by inputting a query in the query input window of the search page. In this case, the user may incorrectly press a key existing on the keyboard, or may input a wrong query if the user presses the key repeatedly.

As such, when a user inputs and searches a typo query, a search result that is completely unrelated to the intended search result may be derived, and thus the search quality may be deteriorated.

In particular, with the widespread use of smartphones, mobile devices may incorrectly select peripheral keyboards or convert consonants incorrectly. Mobile typos are mostly caused by Hangul-Hangul typos rather than Korean-English typos. The Hangul-Hangul typo type on mobile can be used to accidentally click on the surrounding keyboard to make the alphabet unnecessarily, or to delete certain letters, which causes sperm to become sperm. Type of character conversion error.

Specifically, the most frequently used type of error type includes a consonant or a vowel replacement error. In the case of a keyboard user who has a problem, a typo with a dot or a period is introduced.

sperm typo substitute
error melon Mel Ulala session Ulala cushion weather ㅆ sat In order Naver Nay. Naver Movies Naver Youngho

Types of character conversion errors include excessive conversion of consonant sounds into consonants, or typos that enter the wrong support due to no conversion. Examples are shown in Table 2 below.

sperm typo excess
conversion Flower Me too Pension Lottery Alchemy Infinite Infiniz beast Vischen Wonder Girls Wonder Girl Deep-rooted tree Deep wood Vampire test Van Fire Inspection next Yes

Therefore, since the user may unknowingly input the wrong person query, there is a need for a method of changing the wrong person query into a sperm query originally intended by the user when the wrong person query is input. However, there is a problem that a system such as a search engine cannot easily identify a sperm query that a user originally intended to input. Rather, the sperm query proposed by the system may cause an erratic result.

Therefore, when a user inputs a wrong person query, a method of reflecting the user's intention and providing a high accuracy sperm query is required.

According to an embodiment of the present invention, when correcting a user query determined as a typo, a user score calibration method for correcting a typo query more accurately by calculating candidate scores based on probabilities based on dictionary data composed of a typo-sperm query pair. Provide a device.

According to an embodiment, when correcting a user query determined to be a typo, a user query calibration method for correcting a typo query more accurately by weighting a candidate score if the typo in a typo-sperm query pair includes a middle verse Provide a device.

According to an embodiment of the present invention, when correcting a user query determined as a typo, a user query calibration method for correcting a typo query more accurately by weighting a candidate score if the typo in a typo-sperm query pair includes duplicate children and Provide a device.

According to an embodiment, a method of retouching a user query may include determining whether an input user query is a typo query, extracting at least one calibration candidate string when the user query is a typo query, and at least one calibration candidate string. Calculating a candidate score for the step, weighting a string corresponding to a case in which the front character or the rear character of the character to be corrected among the correction candidate strings is the same as the character to be corrected, and according to the candidate score and the weight And determining a calibration query among the candidate candidate strings.

In another embodiment, the method of calculating a candidate score for the at least one candidate string for correction may include calculating a first probability for the sperm query based on dictionary data composed of a typo-sperm query pair. The method may further include calculating a second probability for the user query based on the dictionary data of the sperm-sperm query pair, and calculating the candidate score according to the first probability and the second probability. Can be.

In another embodiment, the determining of a correction query among the candidate candidate strings based on the candidate scores and the weights comprises: adding the weights to the candidate scores to obtain the largest one among the candidate candidate strings. The correction candidate string having the candidate score may be determined as a correction query.

In another method of correcting a user query, a character string corresponding to a case in which a character before or after the character to be corrected is the same as the character to be corrected is a letter of the character before or after the character to be corrected. The character string corresponding to the letter of the character to be corrected, and the step of correcting the user query according to the correction query, characterized in that the step of deleting the letter of the character to be corrected in the typo according to the correction query. have.

In another method of correcting a user query, a string corresponding to a case in which a character before or after the character to be corrected is the same as the character to be corrected is a syllable of a syllable of a character before or after the character to be corrected. A character string corresponding to a syllable of a character to be corrected, and the correcting of the user query according to the correction query may include deleting a syllable of the character corrected in the typographer according to the correction query. have.

According to another embodiment, in the user query correcting method, extracting at least one correction candidate string when the user query is a typo query, the user query includes a typo query of dictionary data consisting of a typo-sperm query pair. The method may include determining whether the information is registered, and extracting a correction candidate string corresponding to the user query from the dictionary data when registered in the typo query.

An apparatus for calibrating a user query according to an embodiment, comprising: a computer storage medium including a processing device and instructions for causing the processing device to perform operations when executed by the processing device, wherein the operations include a typo entered by the user query. Determining whether the query is a query, extracting at least one proof candidate string when the user query is a typo query, calculating a candidate score for the at least one proof candidate string, and correcting among the proof candidate strings The method may include weighting a string corresponding to a case in which a character before or after a character is the same as the character to be corrected, and determining a correction query among the correction candidate strings according to the candidate score and the weight.

A device for calibrating a user query according to another embodiment, wherein the calculating of a candidate score for the at least one candidate string for calibration includes calculating a first probability for the sperm query based on dictionary data composed of a sperm-sperm query pair. The method may further include calculating a second probability for the user query based on the dictionary data of the sperm-sperm query pair, and calculating the candidate score according to the first probability and the second probability. Can be.

A device for calibrating a user query according to another embodiment, wherein the determining of a calibration query from among the candidate candidate strings according to the candidate score and the weighting value comprises: adding the weight to the candidate score to increase the largest among the candidate candidate strings. The correction candidate string having the candidate score may be determined as a correction query.

In another embodiment, a user query correcting apparatus, wherein a character string corresponding to a case in which a character before or after the character to be corrected is the same as the character to be corrected is a letter of the character before or after the character to be corrected. The character string corresponding to the letter of the character to be corrected, and the step of correcting the user query according to the correction query, characterized in that the step of deleting the letter of the character to be corrected in the typo according to the correction query. have.

In another embodiment, a user query correcting apparatus, wherein a character string corresponding to a case in which the character of the character to be corrected is the same as the character to be corrected is a syllable of the syllable of the character of the character before or after the character to be corrected. A character string corresponding to a syllable of a character to be corrected, and the correcting of the user query according to the correction query may include deleting a syllable of the character corrected in the typographer according to the correction query. have.

In another embodiment, a user query correcting apparatus, wherein the extracting of at least one correction candidate string when the user query is a typo query comprises: applying the typo query to the typo query of dictionary data consisting of a typo-sperm query pair. The method may include determining whether the information is registered, and extracting a correction candidate string corresponding to the user query from the dictionary data when registered in the typo query.

According to an embodiment, a computer readable storage medium may be provided that stores one or more programs including instructions for performing a user query retouching method.

In the method and apparatus for correcting a user query according to an embodiment, when correcting a user query determined as a typo, a candidate score based on probability is calculated based on prior data composed of a typo-sperm query pair, thereby more accurately correcting the typo query can do.

In the method and apparatus for correcting a user query according to an embodiment, when correcting a user query determined as a typo, the typo query is more accurate by weighting the candidate score if the typo in the typo-sperm query pair includes the middle verse. Calibration can be made according to the user's intention.

In the method and apparatus for correcting a user query according to an embodiment, when correcting a user query that is determined to be a typo, correcting the typo query more accurately by weighting a candidate score if the typo in a typo-sperm query pair includes duplicate hairs can do.

1 is a diagram schematically illustrating a relationship between a user terminal and a user query calibrating apparatus, according to an exemplary embodiment.
2 is a flowchart for displaying a method of correcting a user query, according to an exemplary embodiment.
3 is a flowchart illustrating a method of extracting a candidate candidate string, according to an exemplary embodiment.
4 is a flowchart of a method of calculating a candidate score, according to an exemplary embodiment.
FIG. 5 is a block diagram illustrating a detailed configuration of an apparatus for correcting a user query according to an embodiment.

Hereinafter, each embodiment will be described in detail with reference to the contents described in the accompanying drawings.

1 is a diagram schematically illustrating a relationship between a user terminal 190 and a user query calibrating apparatus 100.

In the user terminal 190, a user may input a user query 110 for searching. In this case, the user query may be composed of one or more words. The input user query may be transmitted to the user query correcting apparatus 100. The user query correcting apparatus 100 may determine whether the input user query is a typo query. In addition, the user query calibration apparatus 100 may be an electronic device capable of communicating with the user terminal 190, and according to an exemplary embodiment, may be a server.

In this case, when the user query is determined to be a typo query, the user query calibration apparatus 100 may correct the typo query to provide the calibration query 120. On the contrary, if the user query calibration apparatus 100 does not determine the input user query as a typo query, it may not provide a separate calibration query.

Here, the user terminal 190 may be a device capable of communicating with the user inquiry correcting apparatus 100 and may include a PC, a notebook computer, an IPTV, a smartphone, a tablet PC, and all other electronic devices capable of wired and wireless communication. According to an embodiment, a misspelling query including a duplicate character may be a problem in a portable terminal such as a smartphone.

2 is a flowchart for displaying a method of correcting a user query, according to an exemplary embodiment.

In step S210, it is possible to determine whether the input user query is a typo query, and if it is determined that the typo query is not the typo query, the user query can be immediately terminated without performing the next step. The user query refers to a query input by the user, and may be composed of a word or a set of words input when the user searches or writes a document. Here, the typo query may refer to a query that occurs when a Korean-English key change is incorrect or a key input is incorrect. According to an embodiment, a key input may be wrong in a portable terminal such as a smart phone.

Specifically, if the input user query does not have a string in the sperm dictionary with the sperm word, or if the user query is in the list of error words commonly encountered by the users collected in advance, the number of documents retrieved by searching the web with the user query is remarkable. In a few cases, the user's query may be judged as a typo query if the user's query is low frequency or does not appear in a text composed of sperm words such as a novel or newspaper article, or if the user's query corresponds to a Korean character that is not completed.

According to an embodiment of the present invention, it is possible to determine whether the input user query and the correction candidate string are registered in the sperm query of the dictionary data composed of the sperm-sperm query pair. It may mean data including a sperm query corresponding to each query. Misspelled queries can include spaces, and sperm queries can include spaces. An example of dictionary data consisting of a dentifier-sperm query pair is shown in Table 3.

Typo Sperm vaginal Impressed An impressive book Korean Cart Korea Card Wonder Girls Study Wonder Girls Sohat Naver Zitshopin Naver Knowledge Shopping

Dictionary data composed of sperm words may refer to data containing sperm words. In detail, the sperm word may be extracted from highly accurate data such as a Korean dictionary and an encyclopedia. Dictionary data consisting of a sperm-sperm query pair provides a sperm query for the entire sperm query, while dictionary data consisting of sperm words may provide a sperm word corresponding to each word constituting the sperm query.

Subsequently, in the correction candidate string extraction step (S220), the correction candidate string may be extracted from the dictionary data composed of the sperm-sperm query pairs with respect to the user query that is determined as the sperm query. For example, a "girl's age, a boy's age" corresponding to a sperm query of a sperm-sperm query pair may be extracted as a correction candidate string for a sperm query called "one generation".

In operation S230, candidate scores may be calculated based on a probability of inputting a sperm query and a probability of inputting a user query. This is described in detail in FIG. 3.

Subsequently, in step S240, it may be determined whether the extracted candidate candidate character string has been corrected with duplicate characters. The overlapping character is the same as the character to be corrected before or after the character to be corrected, specifically, the syllable or the letter may be duplicated. For example, "car" extracted as a candidate string for "car" that duplicates "car" twice as a duplicate syllable, or "Hanna" " It may correspond to "Hana card" extracted as a candidate string for correction of "Hanna card" where "B" is duplicated twice.

In this case, if the corrected candidate string is extracted by correcting duplicate characters, the process may proceed to the next weighting step S250, and if not, the process may proceed to the correcting query step S260.

In the weighting step (S250), among the candidate candidate strings corresponding to the user query determined as the typo query, weights are applied to candidate scores of the candidate candidate strings corresponding to the case in which the first letter or the second letter of the letter to be corrected is the same as the letter to be corrected. Can be given. The weight may be calculated based on the probability of the user making a typo with duplicate syllables or duplicate letters.

For example, when "Hanna Card" is entered as a user query, there are "Hana Card" and "Hanja Card" as possible candidates. "Hana Card" is a candidate string corrected by deleting the duplicate letter "b". "Kanji card" is a candidate character string corrected by changing "b" to "j". According to an embodiment of the present invention, it is reasonable to view “n” twice as wrong in a portable terminal using a pad that is cheonjiin, so that only “one card” is weighted among the candidate candidate strings or more weighted than “kanji card”. can do.

Subsequently, in the determination of the correction query (S260), the correction query may be determined among the candidate candidate strings according to the candidate score and the weight. Specifically, a candidate candidate string having the largest candidate score is determined as a correction query by summing a candidate score according to a first probability for a sperm query, a candidate score according to a second probability for a typo query, and weights for overlapping characters. Can be.

For example, assume that there are "Hanja card" and "Hana card" as candidate correction strings for "Hanna card". First, the candidate scores are summed based on the first probability according to the frequency with which the "kanji card" is input for the "kanji card" and the second probability according to the frequency of incorrectly inputting "ja" as "I". Next, candidate scores based on the first probability according to the frequency with which "Hana Card" is input and the second probability according to the frequency of incorrectly inputting "Ha" as "one" and "B" for "Hana Card" Sum the weights according to the corrections you deleted. If the candidate scores are similar according to the probability, the "one card" has weights for the duplicate characters, and when summed, the candidate score is larger than the "Hanja card", and thus can be determined by the correction query.

3 is a flowchart illustrating a method of extracting a candidate candidate string, according to an exemplary embodiment.

In operation S321, it may be determined whether the input user query is registered in the sperm query of the dictionary data consisting of the sperm-sperm query pair, and according to an embodiment, it may be performed in advance in step 210 of FIG. 2. . If the user query input here is registered in the sperm query of the dictionary data consisting of the sperm-sperm query pair, the sperm word providing step (S322), and if not registered, the algorithm-based extraction step (S323) may be performed.

In the sperm word providing step (S322), when the user query is found in the dictionary data, the corresponding sperm query may be extracted as the correction candidate string. Here, the candidate candidate string corresponding to the user query may be at least one. According to an embodiment, dictionary data for a candidate candidate string for which a duplicate consonant or a syllable is deleted from a user query may be separately constructed as shown in Table 4 below.

Typo Sperm vaginal
(Delete Duplicate Characters) Sperm vaginal High speed Samhwa Express Samwha Express ... Hannah Card Hana Card Kanji Card ... Car car Car window ... ?example Milbon Gold price Gold price Imprisonment ...

In the algorithm-based extraction step (S323), when the user query is not found in the dictionary data, the candidate candidate string may be extracted through the algorithm. According to an embodiment of the present invention, since dictionary data consisting of a typo-sperm query pair is statistically accumulated data on the typo, the searcher may not provide a corrective query for the typo if there is no input user query. Therefore, a large list of sperm words can be made, and candidate candidate strings can be extracted using an algorithm that extracts sperm words that are similar in shape to the user's query.

4 is a flowchart of a method of calculating a candidate score, according to an exemplary embodiment.

In a first probability calculation step S431, a probability for a sperm query may be calculated. Specifically, it may be defined in Equation 1. For example, if the input user query is "Hanna card" and "Hana card" is extracted as a candidate candidate string for correction, the probability that the user inputs "Hana card" may be the first probability for the sperm query.

Subsequently, in the second probability calculation step S432, a probability for the user query may be calculated. Specifically, it may be defined in Equation 1. For example, if the entered user query is "Hannacard" and "Hanacard" is extracted as the candidate correction string for it, the probability that the user incorrectly inputs "ha" as "han" is the second probability for the user query. Can be.

According to an embodiment, the probability of the sperm query and the user query described above may be determined according to Equation 1 below, which calculates a syllable conversion probability based on different syllables.

에 대해 음절 단위의 음절 변환 확률을 이용할 수 있다. 여기서,

는 사용자가 오탈자를 정자로 잘못 알고 있다가, 사용자가 잘못 알고 있음을 인지하고 정자로 교정할 확률을 의미할 수 있다. 또는

는 사용자가 오탈자 질의를 입력한 후, 질의가 잘못 입력되었음을 인지하고, 정자 질의를 입력할 확률로 해석될 수 있다.Q may mean a user query, and Q 'may mean a sperm query corrected through dictionary data composed of a typo-sperm query pair. At this time,

The syllable conversion probability in syllable units can be used for. here,

Denotes a probability that a user incorrectly recognizes a typo as a sperm, and then corrects the sperm by recognizing that the user incorrectly recognizes the typo. or

After the user inputs a wrong person query, the user may recognize that the query is incorrectly input and may be interpreted as a probability of inputting a sperm query.

는

로 대체될 수 있다. 이 때,

는 사용자가 사용자 질의를 정자 질의로 생각하고 있지만, 타이핑 과정에서 오탈자를 생성할 확률로 해석될 수 있다.

The

≪ / RTI > At this time,

Although the user considers the user query as a sperm query, it can be interpreted as a probability of generating a typo in the typing process.

When calculating the probability of conversion for all the words that make up a user query, a data shortage problem may occur. In addition, as the number of words increases, the amount of calculation may increase rapidly. Therefore, according to an embodiment, syllable conversion probabilities in syllable units may be calculated for different syllable sequence portions of a user query and a sperm query.

는 수학식 2에 따라 결정될 수 있다.For example, in Equation 1

May be determined according to Equation 2.

는 음절간 변환 확률을 나타낸다. 단어 q_ij와 q'_ij에 대해 상이한 음절열을 기준으로 분할할 수 있다. 수학식 2에서는 k개의 분할이 이루어진 것으로 가정하여 분할된 결과 중 서로 다른 음절열에 대해서 확률을 계산할 수 있다. 예를 들어, 사용자 질의가 abcd이고, 정자 질의가 abed인 경우, 음절간 변환 확률

는

가 될 수 있다.At this time, in equation (2)

Denotes the probability of intersyllable conversion. The words q _ij and q ' _ij may be divided based on different syllable sequences. In Equation 2, assuming that k divisions are made, a probability may be calculated for different syllable sequences among the divided results. For example, if the user query is abcd and the sperm query is abed, the intersyllable conversion probability

The

.

For example, the intersyllable conversion probability may be calculated through the following process using dictionary data consisting of a sperm-sperm query pair, QC (input frequency for a user query), and QQ (input frequency of a user query pair).

(1) Assign QC and QQ to each sperm-sperm query pair included in the dictionary data. For example, abcd (qc: 10) -abed (qc: 100), qq: 5

(2) Determine different substrings (c-e) in the typos-sperm query pair.

(3) Calculate the frequency of the substring. Specifically, the sum of qc and qq is calculated for all typos-sperm query pairs in which c-e pairs appear in prior data. For example, c (qc: 50) -e (qc: 1000), qq: 20

(4) The syllable conversion probability is calculated using the calculated frequency.

=20/50

= 20/50

In operation S433, the candidate score may be calculated based on the first probability and the second probability. The candidate score may be calculated by a formula that converts the probability into a score, such as using the probability as it is or converting the probability into a frequency.

FIG. 5 is a block diagram illustrating a detailed configuration of the user query correcting apparatus 500 according to an exemplary embodiment.

The user query calibration apparatus 500 may include a processing apparatus 501 and a storage medium 502. The user query calibration apparatus 500 may be a server as an electronic device that receives a user query 510 from the user terminal 590 and transmits a calibration query 520, but is not limited thereto.

The processing device 501 may include a CPU and other microprocessors as a device for processing operations according to a program stored in the storage medium 502.

The storage medium 502 determines, by the processing device 501, whether the input user query is a typo query, extracting at least one calibration candidate string when the user query is a typo query, and the at least one Calculating a candidate score for a proof candidate string, weighting a string corresponding to a case in which the front letter or the back letter of the letter to be corrected among the proof candidate strings is the same as the letter to be corrected, and the candidate score and the weight The program may include a program including an instruction to perform an operation according to the step of determining a calibration query among the calibration candidate strings. The storage medium 502 may be, for example, hardware capable of storing a hard disk, an SSD, an SD card, and other programs.

The apparatus described above may be implemented as a hardware component, a software component, and / or a combination of hardware components and software components. For example, the apparatus and components described in the embodiments may be implemented within a computer system, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA) A programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For ease of understanding, the processing apparatus may be described as being used singly, but those skilled in the art will recognize that the processing apparatus may have a plurality of processing elements and / As shown in FIG. For example, the processing unit may comprise a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of the above, and configure the processing unit to operate as desired, or independently or collectively. The processing device can be commanded. The software and / or data may be in the form of any type of machine, component, physical device, virtual equipment, computer storage media, or device , Or may be permanently or temporarily embodied in a transmitted signal wave. The software may be distributed over a networked computer system and stored or executed in a distributed manner. The software and data may be stored on one or more computer readable recording media.

The method according to an embodiment may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

500: user quality correction device
501 processing device
502: storage medium
510: user query
520: correction query
590: user terminal

Claims (11)

Determining whether the input user query is a typo query;
Extracting at least one correction candidate string when the user query is a typo query;
Calculating a candidate score for the at least one calibration candidate string;
Determining whether the at least one proof candidate string corrects duplicate characters in the user query;
And based on the determination result, if the at least one proof candidate string includes correcting duplicate characters in the user query, the correct candidate candidate string corresponding to correcting the duplicate characters in the user query is higher than the remaining proof candidate string. Assigning a relatively high weight; And
Determining a calibration query among the at least one calibration candidate string according to the candidate score and the weight
User query correction method comprising a. The method of claim 1,
Computing a candidate score for the at least one correction candidate string,
Calculating a first probability for the sperm query based on dictionary data consisting of a sperm-sperm query pair;
Calculating a second probability for the user query based on dictionary data consisting of the sperm-sperm query pair; And
Calculating the candidate score according to the first probability and the second probability
User query correction method further comprising. 3. The method of claim 2,
Determining a calibration query from the at least one calibration candidate string according to the candidate score and the weight,
Adding the weights to the candidate scores to determine a calibration candidate string having the largest candidate score among the at least one calibration candidate strings as a calibration query.
User query correction method characterized in that. The method of claim 1,
The character string corresponding to the case where the character before or after the character to be corrected is the same as the character to be corrected,
A character string corresponding to the case where the letter or syllable of the letter or the letter of the letter to be corrected is the same as the letter or syllable of the letter to be corrected,
Correcting the user query according to the correction query,
Deleting the alphabet or syllable of the character to be corrected in the typo in accordance with the correction query
User query correction method characterized in that. The method of claim 1,
If the user query is a typo query, extracting at least one correction candidate string,
Determining whether the user query is registered in a spelling query of dictionary data consisting of a spelling-sperm query pair; And
Extracting a candidate candidate string corresponding to the user query from the dictionary data when registered in the typo query
User query correction method comprising a. As an electronic device,
Processing device; And
A computer storage medium comprising instructions for causing the processing device to perform operations when executed by the processing device, the operations comprising:
Determining whether the input user query is a typo query;
Extracting at least one correction candidate string when the user query is a typo query;
Calculating a candidate score for the at least one calibration candidate string;
Determining whether the at least one proof candidate string corrects duplicate characters in the user query;
And based on the determination result, if the at least one proof candidate string includes correcting duplicate characters in the user query, the correct candidate candidate string corresponding to correcting the duplicate characters in the user query is higher than the remaining proof candidate string. Assigning a relatively high weight; And
Determining a calibration query among the at least one calibration candidate string according to the candidate score and the weight
User query correction apparatus comprising a. The method according to claim 6,
Computing a candidate score for the at least one correction candidate string,
Calculating a first probability for the sperm query based on dictionary data consisting of a sperm-sperm query pair;
Calculating a second probability for the user query based on dictionary data consisting of the sperm-sperm query pair; And
Calculating the candidate score according to the first probability and the second probability
The user query correction apparatus further comprising. The method of claim 7, wherein
Determining a calibration query from the at least one calibration candidate string according to the candidate score and the weight,
Adding the weights to the candidate scores to determine a calibration candidate string having the largest candidate score among the at least one calibration candidate strings as a calibration query.
User query correction apparatus characterized in that. The method according to claim 6,
The character string corresponding to the case where the character before or after the character to be corrected is the same as the character to be corrected,
A character string corresponding to the case where the letter or syllable of the letter or the letter of the letter to be corrected is the same as the letter or syllable of the letter to be corrected,
Correcting the user query according to the correction query,
Deleting the letter or syllable of the letter to be corrected in the typo in accordance with the correction query
User query correction apparatus characterized in that. The method according to claim 6,
If the user query is a typo query, extracting at least one correction candidate string,
Determining whether the user query is registered in a spelling query of dictionary data consisting of a spelling-sperm query pair; And
Extracting a candidate candidate string corresponding to the user query from the dictionary data when registered in the typo query
User query correction apparatus comprising a. A computer readable storage medium having stored thereon one or more programs comprising instructions for performing the method of any one of claims 1 to 5.

Images