CN114035693A - Pinyin input method and device, computer equipment and storage medium - Google Patents
Pinyin input method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN114035693A CN114035693A CN202111350347.5A CN202111350347A CN114035693A CN 114035693 A CN114035693 A CN 114035693A CN 202111350347 A CN202111350347 A CN 202111350347A CN 114035693 A CN114035693 A CN 114035693A
- Authority
- CN
- China
- Prior art keywords
- target
- target candidate
- candidate text
- initial
- frequency
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000012216 screening Methods 0.000 claims abstract description 49
- 230000002238 attenuated effect Effects 0.000 claims description 15
- 238000012163 sequencing technique Methods 0.000 claims description 14
- 238000004590 computer program Methods 0.000 claims description 13
- 238000011156 evaluation Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000012913 prioritisation Methods 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/02—Input arrangements using manually operated switches, e.g. using keyboards or dials
- G06F3/023—Arrangements for converting discrete items of information into a coded form, e.g. arrangements for interpreting keyboard generated codes as alphanumeric codes, operand codes or instruction codes
- G06F3/0233—Character input methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/451—Execution arrangements for user interfaces
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Input From Keyboards Or The Like (AREA)
- Document Processing Apparatus (AREA)
Abstract
The application belongs to the technical field of intelligent recommendation and provides a pinyin input method, a pinyin input device, a pinyin input computer device and a pinyin storage medium, wherein the pinyin input method comprises the following steps: determining all next first target initial letters associated with the first target initial letters input by the user and the frequency of the first target initial letters according to historical input records, comparing each next first target initial letter with the first target candidate text, screening out second target candidate texts, screening out third target candidate texts with adjacent initial letters of each initial string in the keyboard from the second target candidate texts, attenuating the frequency of the third target candidate texts as the frequency of the third target candidate texts, sorting the second target candidate texts according to the sequence of the frequencies from high to low, screening out the second target candidate texts with the top M positions as target candidate texts to be recommended, so as to reduce the recommendation of the second target candidate texts possibly with input errors and improve the recommendation accuracy.
Description
Technical Field
The application relates to the technical field of intelligent recommendation, in particular to a pinyin input method, a pinyin input device, computer equipment and a storage medium.
Background
When a user needs to use pinyin for input, in order to help the user to find a text needing to be input more quickly, an automatic semantic association recommendation function based on the pinyin is needed, for example, the user starts to type 'LX', a recommendation list needs to be fed back in a popup window and contains texts such as 'interest amount' and 'interest too high', the user can select a text which the user wants from the recommended list, if the text does not exist, the user continues to type, and meanwhile, the recommendation list refreshes the recommendation list according to the typing of the user.
Disclosure of Invention
The present application is directed to a pinyin input method, apparatus, computer device, and storage medium, so as to improve recommendation accuracy.
In order to achieve the above object, the present application provides a pinyin input method, which includes the following steps:
receiving a first target initial input by a user in a blank input box;
acquiring all pre-stored candidate texts, and screening out a first target candidate text from the candidate texts based on the first target initial; the first letter of the pinyin corresponding to the first character of the first target candidate text is consistent with the first target first letter;
acquiring a historical input record of the user during keyboard input, determining all next first target initials associated with the first target initials according to the historical input record, and calculating the frequency of each next first target initials in the historical input record;
comparing each next first target initial letter with the first target candidate text, and screening out a second target candidate text from the first target candidate text; wherein the first letter of the pinyin corresponding to the second word of the second target candidate text is consistent with one of the next first target first letters;
acquiring an initial corresponding to each word in each second target candidate text to obtain an initial string of each second target candidate text;
screening out a second target candidate text which is adjacent to the first letter of each first letter string and also exists at the position of the keyboard to obtain a third target candidate text;
calculating the logarithm of the adjacent first letter of the third target candidate text, and attenuating the frequency of the third target candidate text according to the logarithm to be used as the frequency of the third target candidate text; wherein the attenuated frequency of the third target candidate text is inversely proportional to the logarithm;
sequencing the second target candidate texts according to the sequence of the frequency from high to low to obtain the second target candidate texts arranged at the top M positions, so as to obtain target candidate texts to be recommended; wherein M is a positive integer;
and displaying the target candidate text to be recommended above the blank input box.
Preferably, the sorting the second target candidate texts according to the order of the frequencies from high to low includes:
calculating the real-time input heat of the second target candidate text;
acquiring a first weight preset for the real-time input heat;
acquiring a second weight preset for the frequency;
calculating an evaluation value of the second target candidate text based on the frequency, the real-time input heat, the first weight and the second weight;
and sorting the second target candidate texts in the order of the evaluation values from high to low.
Preferably, the calculating the real-time input heat of the second target candidate text comprises:
acquiring the frequency of each second target candidate text selected in the past;
calculating a total frequency of all the second target candidate texts selected in the past based on the frequency of each of the second target candidate texts being selected;
and calculating the real-time input heat of each second target candidate text according to the selected frequency and the total frequency of each second target candidate text.
Further, the receiving the first target initial input by the user in the blank input box further includes:
responding to a click instruction of a user for clicking a blank input box, and acquiring a candidate text with the top N-bit input popularity in real time to obtain a reference candidate text; wherein N is a positive integer;
displaying the reference candidate text above the blank input box;
the receiving the first target initial input by the user in the blank input box further comprises:
hiding the reference candidate text displayed above the blank input box.
Further, after the displaying the target candidate text to be recommended above the blank input box, the method further includes:
receiving a second target initial input by a user in the blank input box;
screening out new first target candidate texts from the first target candidate texts based on the second target initial letters; the first letter of the pinyin corresponding to the second word of the new first target candidate text is consistent with the first letter of the second target;
taking the second target initial as the first target initial, and taking the new first target candidate text as the first target candidate text;
returning to execute the step of obtaining the historical input record of the user during keyboard input and determining all next first target initials associated with the first target initials according to the historical input record until a new target candidate text to be recommended is obtained;
and updating the candidate text of the target to be recommended displayed above the blank input box by using the new candidate text of the target to be recommended.
Preferably, the screening out that there is also an adjacent second target candidate text at the position of the adjacent initial of each initial string in the keyboard, and obtaining a third target candidate text includes:
determining all adjacent initials of each initial string; wherein each of said adjacent initials consists of two adjacent initials in said initial string;
acquiring the position information of the first letters of all adjacent first letters in the keyboard;
judging whether the keyboard contains adjacent first letters with adjacent positions in the keyboard or not according to the position information;
if so, screening out the initial strings corresponding to the adjacent initial also adjacent to the position in the keyboard to obtain a target initial string;
and taking the second target candidate text corresponding to the target initial string as a third target candidate text.
Preferably, the attenuating the frequency of the third target candidate text according to the logarithm is used as the frequency of the third target candidate text, and includes the following formula:
Y＝ye-ωt；
wherein Y is the frequency of the third target candidate text before attenuation, Y is the frequency of the third target candidate text after attenuation, e is a constant, t is a logarithm, and ω is an attenuation factor.
The present application further provides a pinyin input device, which includes:
the receiving module is used for receiving a first target initial letter input by a user in the blank input box;
the acquisition module is used for acquiring all pre-stored candidate texts and screening out a first target candidate text from the candidate texts based on the first target initial; the first letter of the pinyin corresponding to the first character of the first target candidate text is consistent with the first target first letter;
the determining module is used for acquiring a historical input record of the user during keyboard input, determining all next first target initials associated with the first target initials according to the historical input record, and calculating the frequency of each next first target initials in the historical input record;
the comparison module is used for comparing each next first target initial letter with the first target candidate text and screening out a second target candidate text from the first target candidate text; wherein the first letter of the pinyin corresponding to the second word of the second target candidate text is consistent with one of the next first target first letters;
the obtaining module is further configured to obtain an initial corresponding to each word in each second target candidate text, and obtain an initial string of each second target candidate text;
the screening module is used for screening out that adjacent second target candidate texts also exist at the positions of the adjacent first letters of each first letter string in the keyboard to obtain third target candidate texts;
the calculation module is used for calculating the logarithm of the adjacent initial of the third target candidate text, and attenuating the frequency of the third target candidate text according to the logarithm to be used as the frequency of the third target candidate text; wherein the attenuated frequency of the third target candidate text is inversely proportional to the logarithm;
the sorting module is used for sorting the second target candidate texts according to the sequence of the frequency from high to low, acquiring the second target candidate texts arranged at the top M positions, and obtaining target candidate texts to be recommended; wherein M is a positive integer;
and the display module is used for displaying the candidate text of the target to be recommended above the blank input box.
The present application further provides a computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the steps of any of the above methods when executing the computer program.
The present application also provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of any of the methods described above.
When a first target initial letter input by a user in a blank input box is received, screening a corresponding first target candidate text from the candidate texts based on the first target initial letter, determining all next first target initial letters associated with the first target initial letter and the frequency thereof according to a historical input record, comparing each next first target initial letter with the first target candidate text, screening out second target candidate texts to obtain texts containing characters which are likely to be input next by the user, then obtaining the corresponding initial letter of each character in each second target candidate text to obtain the initial string of each second target candidate text, screening out the position of the adjacent initial letter of each initial string in a keyboard and also having adjacent second target candidate texts, and obtaining a third target candidate text to obtain the third target candidate text with wrong input caused by the fact that the user may touch the keyboard by mistake during typing, attenuating the frequency of the third target candidate text according to the logarithm of adjacent initial letters to be used as the frequency of the third target candidate text, finally sequencing the second target candidate text according to the sequence from high frequency to low frequency, screening out the second target candidate text with top M positions as the target candidate text to be recommended, and displaying the target candidate text above a blank input box to reduce the recommendation of the second target candidate text with wrong input, so that the displayed second target candidate text is more consistent with the text expected by the user, and the recommendation accuracy is improved.
Drawings
FIG. 1 is a schematic flow chart illustrating a pinyin input method according to an embodiment of the present application;
FIG. 2 is a block diagram illustrating a structure of a pinyin input device according to an embodiment of the present application;
fig. 3 is a block diagram illustrating a structure of a computer device according to an embodiment of the present application.
The implementation, functional features and advantages of the objectives of the present application will be further explained with reference to the accompanying drawings.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
The pinyin input method is used for solving the technical problems that when a user mistakenly touches and inputs wrong pinyin on a keyboard, the user still preferentially recommends a text corresponding to the wrong pinyin, the recommended text is not a result wanted by the user, and the recommendation accuracy is low. Referring to fig. 1, in one embodiment, the pinyin input method includes the following steps:
s11, receiving a first target initial letter input by a user in the blank input box;
s12, acquiring all pre-stored candidate texts, and screening out a first target candidate text from the candidate texts based on the first target initial; the first letter of the pinyin corresponding to the first character of the first target candidate text is consistent with the first target first letter;
s13, acquiring a history input record of the user during keyboard input, determining all next first target initials associated with the first target initials according to the history input record, and calculating the frequency of each next first target initial in the history input record;
s14, comparing each next first target initial letter with the first target candidate text, and screening out a second target candidate text from the first target candidate text; wherein the first letter of the pinyin corresponding to the second word of the second target candidate text is consistent with one of the next first target first letters;
s15, obtaining the initial letter corresponding to each word in each second target candidate text to obtain the initial string of each second target candidate text;
s16, screening out that adjacent second target candidate texts also exist at the positions of the adjacent initial letters of each initial string in the keyboard to obtain third target candidate texts;
s17, calculating the logarithm of the adjacent initial of the third target candidate text, and attenuating the frequency of the third target candidate text according to the logarithm to be used as the frequency of the third target candidate text; wherein the attenuated frequency of the third target candidate text is inversely proportional to the logarithm;
s18, sorting the second target candidate texts according to the sequence of the frequencies from high to low, and acquiring the second target candidate texts arranged at the top M positions to obtain target candidate texts to be recommended; wherein M is a positive integer;
and S19, displaying the candidate text of the target to be recommended above the blank input box.
As described in the above step S11, the blank input box may be a chat input box in which the user does not input any content, and when it is detected that the user inputs a simple pinyin in the blank input box of the chat interface, the first target initial input entered in the blank input box by the user is received.
In order to be compatible with wrongly written characters or spoken language problems, the method supports simple spelling of initials, can use a hidden Markov chain for exhausting possible combinations of the initials in the candidate text, and can calculate the possible combination frequency of the next initials according to a frequency formula.
Specifically, the first target candidate text is found according to the initial letter input by the user in the blank space, such as 'interest', the initial letter is 'LX', and all the first target candidate texts are sorted according to the frequency formula of the hidden Markov chain. Wherein the frequency formula includes P (the next pinyin is i | the client input first LX) — (the frequency number of the next pinyin first i when the client inputs the first LX)/all text sets of the client input first LX. And by analogy, calculating the frequency of each first target candidate text occupying all the first target candidate texts, sequencing all the first target candidate texts according to the frequency to obtain a sequencing result, and displaying the first target candidate texts above the blank input box according to the sequencing result.
In an embodiment, a certain threshold may also be set, for example, the threshold is set to 0.5, among all the first target candidate texts, the first target candidate text with the frequency exceeding the threshold is reserved, and the frequency is recorded as p (p), which is used as the first priority ranking index ranking, and the reserved first target candidate text is ranked according to the first priority ranking, so as to filter out the first target candidate text desired by the user, thereby improving the recommendation accuracy.
In addition, a chinese character search recommendation may also be performed on the basis of the first letter of the pinyin of the first target candidate text, such as interest, as a 2_ gram, and a complete matching of the first letter is performed in all the first target candidate texts, and a value p (h) is set as a sorting criterion, and if the first target candidate text completely matches all the first letters input by the user in the blank input box, the first target candidate text is prioritized in the front, such as where the complete matching is denoted as p (h) ═ 1, and if the complete matching is not performed, the value p (h) ═ 0, so as to sort the first target candidate text by the second prioritization index, and preferentially recommend the completely matched first target candidate text, so as to facilitate the user to find a desired result.
As described in the above step S12, when the user inputs the first target initial, a candidate text having the same initial as the first target initial input by the user is selected from all the current candidate texts as the first target candidate text. The candidate text can be a text corresponding to the simple spelling with the input hot degree of the user in the front.
As described in the above step S13, the present embodiment obtains the history input record of the user when the keyboard, which may be a physical keyboard or a virtual keyboard, inputs letters in the history data record including the frequency of each letter input, the frequency of the last initial appearing simultaneously with the next initial, so as to calculate the frequency of all the next first target initial associated with the first target initial appearing in the history input record from the history input record, i.e., obtain the first initial that is likely to be input next to the first target initial.
As described in step S14, each next first target initial is compared with the first target candidate text one by one, specifically, each next first target initial is compared with the initial of the pinyin corresponding to the second word of the first target candidate text, and the first target candidate text containing the first target initial and the next first target initial in the first target candidate text is screened out and used as the second target candidate text. For example, when the initial is W, then the second target candidate text may be a word, a literature, a text, with the next initial of the initial being Z, X, B, respectively, and the corresponding frequency may be 80%, 70%, 60%.
As described in the above step S15, the present embodiment traverses the initial letter corresponding to each word in each second target candidate text to obtain the initial string of each second target candidate text, i.e., the initial string is composed of the initial letter corresponding to each word in the second target candidate text.
As described in the above steps S16-S17, it is then determined whether the adjacent initials of each initial string are also adjacent in the keyboard, and the logarithm of the adjacent initials of the third target candidate text is calculated, and the frequency of the third target candidate text is attenuated according to the logarithm to be the frequency of the third target candidate text. Wherein the attenuated frequency of the third target candidate text is inversely proportional to the logarithm.
For example, if the second target candidate text is "text match" or "literary connotation", the initial string of "text match" is "WZYZ", and since the corresponding positions of every two adjacent initial letters "WZ", "ZY", and "YZ" are not adjacent in the keyboard, the "text match" is that there is no first letter adjacent to the keyboard, the second target candidate text is marked as not having an adjacent second target text, and the frequency of the absence of the adjacent "text match" is taken as the target frequency of the "text match"; and the initial string of the "literary connotation" is "WXHY", and since the corresponding positions of the "WX" and the "XH" in every two adjacent initial letters are not adjacent, and the corresponding positions of the "HY" in the keyboard are adjacent, the "literary connotation" is the initial letter of the adjacent keyboard, is marked as the second target candidate text, and the adjacent logarithm isIf the number of the pairs is one, the frequency of the literature maintenance is attenuated according to the number of the pairs to obtain the target frequency, for example, the frequency of the literature maintenance is multiplied by 0.9bAnd b is the adjacent logarithm, and the target frequency of the literature connotation is obtained. Because two adjacent initial letters adjacent to the keyboard are generally letters which are easy for a user to input errors, for example, when the user inputs, two letters adjacent to the keyboard are easy to touch between fingers, the target frequency is obtained by attenuating the frequency of the second target candidate text with the two adjacent initial letters adjacent to the keyboard.
As described in the above steps S18-S19, the present embodiment ranks the second target candidate texts based on frequency to obtain a ranking result. If the adjacent first letters but the adjacent second target candidate texts do not exist in the keyboard, the original frequency is adopted for sequencing without attenuation; and for the position of the adjacent initial in the keyboard, the adjacent second target candidate texts also exist, the attenuated frequency is adopted for sorting, and then the second target candidate texts are displayed above the blank input box according to the sorting result, so that the second target candidate texts with input errors possibly exist, and the displayed second target candidate texts are more in line with the candidate texts expected by the user.
The pinyin input method includes screening corresponding first target candidate texts from the candidate texts based on first target initial letters when the first target initial letters input by a user in a blank input box are received, determining all next first target initial letters associated with the first target initial letters and frequency thereof according to historical input records, comparing each next first target initial letter with the first target candidate texts, screening second target candidate texts to obtain texts containing characters which are likely to be input next by the user, then obtaining the initial letter corresponding to each character in each second target candidate text to obtain the initial string of each second target candidate text, screening adjacent second target candidate texts of the adjacent initial letters of each initial string in a keyboard to obtain third target candidate texts, the method comprises the steps of obtaining a third target candidate text which is input wrongly and possibly caused by the fact that a user touches a keyboard mistakenly when typing, attenuating the frequency of the third target candidate text according to the logarithm of adjacent initial letters to serve as the frequency of the third target candidate text, finally sorting the second target candidate text according to the sequence from high to low in frequency, screening the second target candidate text which is ranked at the top M as a target candidate text to be recommended, and displaying the target candidate text above a blank input box to reduce the recommendation of the second target candidate text which is possibly input wrongly, so that the displayed second target candidate text is more consistent with the text expected by the user, and the recommendation accuracy is improved.
In an embodiment, the sorting the second target candidate texts according to the order from high to low of the frequency may specifically include:
calculating the real-time input heat of the second target candidate text;
acquiring a first weight preset for the real-time input heat;
acquiring a second weight preset for the frequency;
calculating an evaluation value of the second target candidate text based on the frequency, the real-time input heat, the first weight and the second weight;
and sorting the second target candidate texts in the order of the evaluation values from high to low.
And the real-time input heat is used for representing the selected heat of the second target candidate text. The present embodiment may calculate the real-time input heat of the second target candidate text by the frequency of the second target candidate text being selected by the user, and the real-time input heat is automatically updated once based on each time the user selects the second target candidate text, then obtain the first weight set in advance for the real-time input heat, and obtain the second weight set in advance for the frequency, generally, the first weight is smaller than the second weight, to prioritize the second target candidate text that is most likely to be desired by the user, finally calculate the evaluation value of the second target candidate text based on the frequency, the real-time input heat, the first weight, and the second weight, and rank the second target candidate text based on the level of the evaluation value, such as ranking the second target text with high evaluation value in the front and ranking the second target text with low evaluation value in the back.
The evaluation value calculation formula is as follows:
S＝P1*A+P2*B；
wherein, P1 is the first weight, A is the real-time input heat, P2 is the second weight, and B is the frequency.
In an embodiment, the calculating the real-time input heat of the second target candidate text may specifically include:
acquiring the frequency of each second target candidate text selected in the past;
calculating a total frequency of all the second target candidate texts selected in the past based on the frequency of each of the second target candidate texts being selected;
and calculating the real-time input heat of each second target candidate text according to the selected frequency and the total frequency of each second target candidate text.
In this embodiment, the frequency of selecting each second target candidate text is obtained, after each second target candidate text is selected in a history period, the frequency is accumulated once on the original basis, then the frequency of selecting all the second target candidate texts is accumulated to obtain a total frequency, the ratio of the frequency of selecting each second target candidate text to the total frequency is respectively calculated, the real-time input heat of each second target candidate text is calculated, then the second target candidate texts are ranked based on the real-time input heat, and the second target candidate texts ranked in the front are screened out.
In an embodiment, the receiving the first target initial input by the user in the blank input box may further include:
responding to a click instruction of a user for clicking a blank input box, and acquiring a candidate text with the top N-bit input popularity in real time to obtain a reference candidate text; wherein N is a positive integer;
displaying the reference candidate text above the blank input box;
the receiving the first target initial input by the user in the blank input box further comprises:
hiding the reference candidate text displayed above the blank input box.
In this embodiment, when a user clicks a blank input box, a click instruction for clicking the blank input box is generated, at this time, a candidate text with the top N-th input popularity is obtained, a reference candidate text is obtained, and the reference candidate text is displayed above the blank input box, so that before the user inputs an initial, a reference candidate text with a high popularity in real time is recommended for the user, so as to recommend a text that the user may input. When a first target initial input by a user in the blank input box is received, hiding the reference candidate text displayed above the blank input box, and obtaining a target candidate text to be recommended according to the first target initial.
In an embodiment, after the displaying the target candidate text to be recommended above the blank input box, the method may further include:
receiving a second target initial input by a user in the blank input box;
screening out new first target candidate texts from the first target candidate texts based on the second target initial letters; the first letter of the pinyin corresponding to the second word of the new first target candidate text is consistent with the first letter of the second target;
taking the second target initial as the first target initial, and taking the new first target candidate text as the first target candidate text;
returning to execute the step of obtaining the historical input record of the user during keyboard input and determining all next first target initials associated with the first target initials according to the historical input record until a new target candidate text to be recommended is obtained;
and updating the candidate text of the target to be recommended displayed above the blank input box by using the new candidate text of the target to be recommended.
In this embodiment, when a second target first letter input by the user in the blank input box is received, a new first target candidate text is screened out from the first target candidate text based on the second target first letter, where the second letter of the new first target candidate text corresponds to the new first target candidate text whose pinyin first letter is consistent with the second target first letter, that is, the first letter of the pinyin corresponding to the first word is consistent with the first target first letter, and the second word corresponds to the new first target candidate text whose pinyin first letter is consistent with the second target first letter. Taking the second target initial as the first target initial, taking the new first target candidate text as the first target candidate text, returning to obtain a historical input record of the user during keyboard input, determining all next first target initial associated with the first target initial according to the historical input record and subsequent steps until obtaining a new target candidate text to be recommended, screening the new target candidate text to be recommended based on the first target initial and the second target initial, and finally updating the target candidate text to be recommended displayed above the blank input box by using the new target candidate text to be recommended so as to update the target candidate text to be recommended in real time based on the target initial input by the user and ensure the accuracy of the recommended target candidate text to be recommended.
In an embodiment, the screening out that there is also an adjacent second target candidate text at a position of an adjacent initial of each of the initial strings in the keyboard, and obtaining a third target candidate text may specifically include:
determining all adjacent initials of each initial string; wherein each of said adjacent initials consists of two adjacent initials in said initial string;
acquiring the position information of the first letters of all adjacent first letters in the keyboard;
judging whether the keyboard contains adjacent first letters with adjacent positions in the keyboard or not according to the position information;
if so, screening out the initial strings corresponding to the adjacent initial also adjacent to the position in the keyboard to obtain a target initial string;
and taking the second target candidate text corresponding to the target initial string as a third target candidate text.
In the embodiment, all adjacent first letters of each first string are determined, the position information of the first letters of all the adjacent first letters in the keyboard is acquired, whether the keyboard contains the adjacent first letters with the adjacent positions in the keyboard is judged according to the position information, if the keyboard contains the adjacent first letters with the adjacent positions in the keyboard, the first strings corresponding to the adjacent first letters with the adjacent positions in the keyboard are screened out to obtain the target first strings, and the second target candidate text corresponding to the target first strings is used as the third target candidate text. For example, the first letter string of "literary composition" is "WXHY", and since the corresponding location information of "WX", "XH" in the keyboard is not adjacent and the corresponding location information of "HY" in the keyboard is adjacent in every two adjacent first letters, "literary composition" is the first letter of existence of keyboard adjacent, and is marked as existence of the adjacent second target candidate text.
In an embodiment, the attenuating the frequency of the third target candidate text according to the logarithm as the frequency of the third target candidate text includes the following formula:
Y＝ye-ωt；
wherein Y is the frequency of the third target candidate text before attenuation, Y is the frequency of the third target candidate text after attenuation, e is a constant, t is a logarithm, and ω is an attenuation factor.
Referring to fig. 2, an embodiment of the present application further provides a pinyin input device, including:
the receiving module 11 is configured to receive a first target initial input by a user in a blank input box;
the obtaining module 12 is configured to obtain all pre-stored candidate texts, and screen out a first target candidate text from the candidate texts based on the first target initial; the first letter of the pinyin corresponding to the first character of the first target candidate text is consistent with the first target first letter;
a determining module 13, configured to obtain a historical input record of the user during keyboard input, determine, according to the historical input record, all next first target initials associated with the first target initials, and calculate a frequency of occurrence of each next first target initials in the historical input record;
a comparison module 14, configured to compare each next first target initial with the first target candidate text, and screen a second target candidate text from the first target candidate text; wherein the first letter of the pinyin corresponding to the second word of the second target candidate text is consistent with one of the next first target first letters;
the obtaining module 12 is further configured to obtain an initial corresponding to each word in each second target candidate text, so as to obtain an initial string of each second target candidate text;
the screening module 15 is configured to screen out that an adjacent second target candidate text also exists at a position of an adjacent initial of each initial string in the keyboard, so as to obtain a third target candidate text;
the calculating module 16 is configured to calculate a logarithm of adjacent initials of the third target candidate text, and attenuate the frequency of the third target candidate text according to the logarithm to obtain the frequency of the third target candidate text; wherein the attenuated frequency of the third target candidate text is inversely proportional to the logarithm;
the sorting module 17 is configured to sort the second target candidate texts according to the sequence from high to low of the frequency, obtain the second target candidate texts ranked at top M, and obtain target candidate texts to be recommended; wherein M is a positive integer;
and the display module 18 is used for displaying the target candidate text to be recommended above the blank input box.
The blank input box may be a chat input box in which the user does not input any content, and when it is detected that the user inputs a simple pinyin in the blank input box of the chat interface, the first target initial input in the blank input box by the user is received.
In order to be compatible with wrongly written characters or spoken language problems, the method supports simple spelling of initials, can use a hidden Markov chain for exhausting possible combinations of the initials in the candidate text, and can calculate the possible combination frequency of the next initials according to a frequency formula.
Specifically, the first target candidate text is found according to the initial letter input by the user in the blank space, such as 'interest', the initial letter is 'LX', and all the first target candidate texts are sorted according to the frequency formula of the hidden Markov chain. Wherein the frequency formula includes P (the next pinyin is i | the client input first LX) — (the frequency number of the next pinyin first i when the client inputs the first LX)/all text sets of the client input first LX. And by analogy, calculating the frequency of each first target candidate text occupying all the first target candidate texts, sequencing all the first target candidate texts according to the frequency to obtain a sequencing result, and displaying the first target candidate texts above the blank input box according to the sequencing result.
In an embodiment, a certain threshold may also be set, for example, the threshold is set to 0.5, among all the first target candidate texts, the first target candidate text with the frequency exceeding the threshold is reserved, and the frequency is recorded as p (p), which is used as the first priority ranking index ranking, and the reserved first target candidate text is ranked according to the first priority ranking, so as to filter out the first target candidate text desired by the user, thereby improving the recommendation accuracy.
In addition, a chinese character search recommendation may also be performed on the basis of the first letter of the pinyin of the first target candidate text, such as interest, as a 2_ gram, and a complete matching of the first letter is performed in all the first target candidate texts, and a value p (h) is set as a sorting criterion, and if the first target candidate text completely matches all the first letters input by the user in the blank input box, the first target candidate text is prioritized in the front, such as where the complete matching is denoted as p (h) ═ 1, and if the complete matching is not performed, the value p (h) ═ 0, so as to sort the first target candidate text by the second prioritization index, and preferentially recommend the completely matched first target candidate text, so as to facilitate the user to find a desired result.
When the user inputs the first target initial letter, candidate texts with the initial letters of the first letter same as the first target initial letter input by the user are screened from all current candidate texts and are used as first target candidate texts. The candidate text can be a text corresponding to the simple spelling with the input hot degree of the user in the front.
The present embodiment obtains the history input record of the user when the keyboard is used for inputting letters, the keyboard may be a physical keyboard or a virtual keyboard, the history data record includes the frequency of inputting each letter, and the frequency of the occurrence of the last initial and the next initial at the same time, so as to calculate the frequency of the occurrence of all the next first target initial associated with the first target initial in the history input record according to the history input record, that is, obtain the first initial which is likely to be input next to the first target initial.
And comparing each next first target initial with the first target candidate text one by one, specifically comparing each next first target initial with the first letter of the pinyin corresponding to the second character of the first target candidate text, screening out the first target candidate text containing the first target initial and the next first target initial in the first target candidate text, and taking the first target candidate text as the second target candidate text. For example, when the initial is W, then the second target candidate text may be a word, a literature, a text, with the next initial of the initial being Z, X, B, respectively, and the corresponding frequency may be 80%, 70%, 60%.
In this embodiment, the initials corresponding to each word in each second target candidate text are traversed to obtain the initials of each second target candidate text, that is, the initials are composed of the initials corresponding to each word in the second target candidate text.
And then judging whether the adjacent initial of each initial string is adjacent in the position of the keyboard, calculating the logarithm of the adjacent initial of the third target candidate text, and attenuating the frequency of the third target candidate text according to the logarithm to be used as the frequency of the third target candidate text. Wherein the attenuated frequency of the third target candidate text is inversely proportional to the logarithm.
For example, if the second target candidate text is "text match" or "literary connotation", the initial string of "text match" is "WZYZ", and since the corresponding positions of every two adjacent initial letters "WZ", "ZY", and "YZ" are not adjacent in the keyboard, the "text match" is that there is no first letter adjacent to the keyboard, the second target candidate text is marked as not having an adjacent second target text, and the frequency of the absence of the adjacent "text match" is taken as the target frequency of the "text match"; and the first letter string of the "literary support" is "WXHY", because the corresponding positions of the "WX" and the "XH" in the keyboard are not adjacent in every two adjacent first letters, and the corresponding positions of the "HY" in the keyboard are adjacent, the "literary support" is the first letter with the keyboard adjacent, the character is marked as that the second target candidate text adjacent exists, and the adjacent logarithm is a pair, the frequency of the "literary support" is attenuated according to the pair to obtain the target frequency, for example, the frequency of the "literary support" is multiplied by 0.9bAnd b is the adjacent logarithm, and the target frequency of the literature connotation is obtained. Because two adjacent initial letters adjacent to the keyboard are generally letters which are easy for a user to input errors, for example, when the user inputs, two letters adjacent to the keyboard are easy to touch between fingers, the target frequency is obtained by attenuating the frequency of the second target candidate text with the two adjacent initial letters adjacent to the keyboard.
The second target candidate text is ranked based on the frequency to obtain a ranking result. If the adjacent first letters but the adjacent second target candidate texts do not exist in the keyboard, the original frequency is adopted for sequencing without attenuation; and for the position of the adjacent initial in the keyboard, the adjacent second target candidate texts also exist, the attenuated frequency is adopted for sorting, and then the second target candidate texts are displayed above the blank input box according to the sorting result, so that the second target candidate texts with input errors possibly exist, and the displayed second target candidate texts are more in line with the candidate texts expected by the user.
As described above, it can be understood that the components of the pinyin input device provided in the present application may implement the functions of any one of the pinyin input methods described above, and the detailed structure is not described again.
Referring to fig. 3, an embodiment of the present application further provides a computer device, and an internal structure of the computer device may be as shown in fig. 3. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the computer designed processor is used to provide computational and control capabilities. The memory of the computer device comprises a storage medium and an internal memory. The storage medium stores an operating system, a computer program, and a database. The memory provides an environment for the operation of the operating system and computer programs in the storage medium. The database of the computer device is used for storing candidate texts, historical input records and other data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a pinyin input method.
The processor executes the pinyin input method, and the method comprises the following steps:
receiving a first target initial input by a user in a blank input box;
acquiring all pre-stored candidate texts, and screening out a first target candidate text from the candidate texts based on the first target initial; the first letter of the pinyin corresponding to the first character of the first target candidate text is consistent with the first target first letter;
acquiring a historical input record of the user during keyboard input, determining all next first target initials associated with the first target initials according to the historical input record, and calculating the frequency of each next first target initials in the historical input record;
comparing each next first target initial letter with the first target candidate text, and screening out a second target candidate text from the first target candidate text; wherein the first letter of the pinyin corresponding to the second word of the second target candidate text is consistent with one of the next first target first letters;
acquiring an initial corresponding to each word in each second target candidate text to obtain an initial string of each second target candidate text;
screening out a second target candidate text which is adjacent to the first letter of each first letter string and also exists at the position of the keyboard to obtain a third target candidate text;
calculating the logarithm of the adjacent first letter of the third target candidate text, and attenuating the frequency of the third target candidate text according to the logarithm to be used as the frequency of the third target candidate text; wherein the attenuated frequency of the third target candidate text is inversely proportional to the logarithm;
sequencing the second target candidate texts according to the sequence of the frequency from high to low to obtain the second target candidate texts arranged at the top M positions, so as to obtain target candidate texts to be recommended; wherein M is a positive integer;
and displaying the target candidate text to be recommended above the blank input box.
An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, the computer program, when executed by a processor, implementing a pinyin input method, including the steps of:
receiving a first target initial input by a user in a blank input box;
acquiring all pre-stored candidate texts, and screening out a first target candidate text from the candidate texts based on the first target initial; the first letter of the pinyin corresponding to the first character of the first target candidate text is consistent with the first target first letter;
acquiring a historical input record of the user during keyboard input, determining all next first target initials associated with the first target initials according to the historical input record, and calculating the frequency of each next first target initials in the historical input record;
comparing each next first target initial letter with the first target candidate text, and screening out a second target candidate text from the first target candidate text; wherein the first letter of the pinyin corresponding to the second word of the second target candidate text is consistent with one of the next first target first letters;
acquiring an initial corresponding to each word in each second target candidate text to obtain an initial string of each second target candidate text;
screening out a second target candidate text which is adjacent to the first letter of each first letter string and also exists at the position of the keyboard to obtain a third target candidate text;
calculating the logarithm of the adjacent first letter of the third target candidate text, and attenuating the frequency of the third target candidate text according to the logarithm to be used as the frequency of the third target candidate text; wherein the attenuated frequency of the third target candidate text is inversely proportional to the logarithm;
sequencing the second target candidate texts according to the sequence of the frequency from high to low to obtain the second target candidate texts arranged at the top M positions, so as to obtain target candidate texts to be recommended; wherein M is a positive integer;
and displaying the target candidate text to be recommended above the blank input box.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and can include the processes of the embodiments of the methods described above when the computer program is executed. Any reference to memory, storage, database, or other medium provided herein and used in the examples may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), double-rate SDRAM (SSRSDRAM), Enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and bus dynamic RAM (RDRAM).
To sum up, the most beneficial effect of this application lies in:
when a first target initial letter input by a user in a blank input box is received, screening a corresponding first target candidate text from the candidate texts based on the first target initial letter, determining all next first target initial letters associated with the first target initial letter and the frequency thereof according to a historical input record, comparing each next first target initial letter with the first target candidate text, screening out second target candidate texts to obtain texts containing characters which are likely to be input next by the user, then obtaining the corresponding initial letter of each character in each second target candidate text to obtain the initial string of each second target candidate text, screening out the position of the adjacent initial letter of each initial string in a keyboard and also having adjacent second target candidate texts, and obtaining a third target candidate text to obtain the third target candidate text with wrong input caused by the fact that the user may touch the keyboard by mistake during typing, attenuating the frequency of the third target candidate text according to the logarithm of adjacent initial letters to be used as the frequency of the third target candidate text, finally sequencing the second target candidate text according to the sequence from high frequency to low frequency, screening out the second target candidate text with top M positions as the target candidate text to be recommended, and displaying the target candidate text above a blank input box to reduce the recommendation of the second target candidate text with wrong input, so that the displayed second target candidate text is more consistent with the text expected by the user, and the recommendation accuracy is improved.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, apparatus, article, or method 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, apparatus, article, or method. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, apparatus, article, or method that includes the element.
The above description is only a preferred embodiment of the present application, and not intended to limit the scope of the present application, and all modifications of equivalent structures and equivalent processes, which are made by the contents of the specification and the drawings of the present application, or which are directly or indirectly applied to other related technical fields, are also included in the scope of the present application.
Claims (10)
1. A pinyin input method, comprising:
receiving a first target initial input by a user in a blank input box;
acquiring all pre-stored candidate texts, and screening out a first target candidate text from the candidate texts based on the first target initial; the first letter of the pinyin corresponding to the first character of the first target candidate text is consistent with the first target first letter;
acquiring a historical input record of the user during keyboard input, determining all next first target initials associated with the first target initials according to the historical input record, and calculating the frequency of each next first target initials in the historical input record;
comparing each next first target initial letter with the first target candidate text, and screening out a second target candidate text from the first target candidate text; wherein the first letter of the pinyin corresponding to the second word of the second target candidate text is consistent with one of the next first target first letters;
acquiring an initial corresponding to each word in each second target candidate text to obtain an initial string of each second target candidate text;
screening out a second target candidate text which is adjacent to the first letter of each first letter string and also exists at the position of the keyboard to obtain a third target candidate text;
calculating the logarithm of the adjacent first letter of the third target candidate text, and attenuating the frequency of the third target candidate text according to the logarithm to be used as the frequency of the third target candidate text; wherein the attenuated frequency of the third target candidate text is inversely proportional to the logarithm;
sequencing the second target candidate texts according to the sequence of the frequency from high to low to obtain the second target candidate texts arranged at the top M positions, so as to obtain target candidate texts to be recommended; wherein M is a positive integer;
and displaying the target candidate text to be recommended above the blank input box.
2. The method of claim 1, wherein the ranking the second target candidate texts in order of the frequency from high to low comprises:
calculating the real-time input heat of the second target candidate text;
acquiring a first weight preset for the real-time input heat;
acquiring a second weight preset for the frequency;
calculating an evaluation value of the second target candidate text based on the frequency, the real-time input heat, the first weight and the second weight;
and sorting the second target candidate texts in the order of the evaluation values from high to low.
3. The method of claim 2, wherein calculating the real-time input heat of the second target candidate text comprises:
acquiring the frequency of each second target candidate text selected in the past;
calculating a total frequency of all the second target candidate texts selected in the past based on the frequency of each of the second target candidate texts being selected;
and calculating the real-time input heat of each second target candidate text according to the selected frequency and the total frequency of each second target candidate text.
4. The method of claim 1, wherein the receiving the first target initial input by the user in the blank input box further comprises:
responding to a click instruction of a user for clicking a blank input box, and acquiring a candidate text with the top N-bit input popularity in real time to obtain a reference candidate text; wherein N is a positive integer;
displaying the reference candidate text above the blank input box;
the receiving the first target initial input by the user in the blank input box further comprises:
hiding the reference candidate text displayed above the blank input box.
5. The method according to claim 1, wherein after displaying the target candidate text to be recommended above the blank input box, the method further comprises:
receiving a second target initial input by a user in the blank input box;
screening out new first target candidate texts from the first target candidate texts based on the second target initial letters; the first letter of the pinyin corresponding to the second word of the new first target candidate text is consistent with the first letter of the second target;
taking the second target initial as the first target initial, and taking the new first target candidate text as the first target candidate text;
returning to execute the step of obtaining the historical input record of the user during keyboard input and determining all next first target initials associated with the first target initials according to the historical input record until a new target candidate text to be recommended is obtained;
and updating the candidate text of the target to be recommended displayed above the blank input box by using the new candidate text of the target to be recommended.
6. The method of claim 1, wherein the filtering out that there is also a second target candidate text adjacent to the adjacent initial of each initial string in the keyboard to obtain a third target candidate text comprises:
determining all adjacent initials of each initial string; wherein each of said adjacent initials consists of two adjacent initials in said initial string;
acquiring the position information of the first letters of all adjacent first letters in the keyboard;
judging whether the keyboard contains adjacent first letters with adjacent positions in the keyboard or not according to the position information;
if so, screening out the initial strings corresponding to the adjacent initial also adjacent to the position in the keyboard to obtain a target initial string;
and taking the second target candidate text corresponding to the target initial string as a third target candidate text.
7. The method according to claim 1, wherein the attenuating the frequency of the third target candidate text according to the logarithm is taken as the frequency of the third target candidate text, and comprises the following formula:
Y＝ye-ωt；
wherein Y is the frequency of the third target candidate text before attenuation, Y is the frequency of the third target candidate text after attenuation, e is a constant, t is a logarithm, and ω is an attenuation factor.
8. A pinyin input device, comprising:
the receiving module is used for receiving a first target initial letter input by a user in the blank input box;
the acquisition module is used for acquiring all pre-stored candidate texts and screening out a first target candidate text from the candidate texts based on the first target initial; the first letter of the pinyin corresponding to the first character of the first target candidate text is consistent with the first target first letter;
the determining module is used for acquiring a historical input record of the user during keyboard input, determining all next first target initials associated with the first target initials according to the historical input record, and calculating the frequency of each next first target initials in the historical input record;
the comparison module is used for comparing each next first target initial letter with the first target candidate text and screening out a second target candidate text from the first target candidate text; wherein the first letter of the pinyin corresponding to the second word of the second target candidate text is consistent with one of the next first target first letters;
the obtaining module is further configured to obtain an initial corresponding to each word in each second target candidate text, and obtain an initial string of each second target candidate text;
the screening module is used for screening out that adjacent second target candidate texts also exist at the positions of the adjacent first letters of each first letter string in the keyboard to obtain third target candidate texts;
the calculation module is used for calculating the logarithm of the adjacent initial of the third target candidate text, and attenuating the frequency of the third target candidate text according to the logarithm to be used as the frequency of the third target candidate text; wherein the attenuated frequency of the third target candidate text is inversely proportional to the logarithm;
the sorting module is used for sorting the second target candidate texts according to the sequence of the frequency from high to low, acquiring the second target candidate texts arranged at the top M positions, and obtaining target candidate texts to be recommended; wherein M is a positive integer;
and the display module is used for displaying the candidate text of the target to be recommended above the blank input box.
9. A computer device, comprising:
a processor;
a memory;
wherein the memory stores a computer program, and the processor implements the pinyin input method of any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, wherein the computer-readable storage medium has stored thereon a computer program which, when executed by a processor, implements the pinyin input method as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111350347.5A CN114035693A (en) | 2021-11-15 | 2021-11-15 | Pinyin input method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111350347.5A CN114035693A (en) | 2021-11-15 | 2021-11-15 | Pinyin input method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114035693A true CN114035693A (en) | 2022-02-11 |
Family
ID=80137677
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111350347.5A Pending CN114035693A (en) | 2021-11-15 | 2021-11-15 | Pinyin input method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114035693A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101079060A (en) * | 2007-03-26 | 2007-11-28 | 腾讯科技（深圳）有限公司 | Chinese character input simple 'pinyin' implementation method and system |
| CN109032375A (en) * | 2018-06-29 | 2018-12-18 | 北京百度网讯科技有限公司 | Candidate text sort method, device, equipment and storage medium |
| CN111090341A (en) * | 2019-12-24 | 2020-05-01 | 科大讯飞股份有限公司 | Input method candidate result display method, related equipment and readable storage medium |
| CN111488466A (en) * | 2020-04-16 | 2020-08-04 | 清华大学 | Chinese language error corpus generating method, calculating device and storage medium |
| CN112684910A (en) * | 2020-12-29 | 2021-04-20 | 维沃移动通信有限公司 | Input method candidate word display method and device and electronic equipment |
| CN112748811A (en) * | 2021-01-21 | 2021-05-04 | 李博林 | English word input method and device |
-
2021
- 2021-11-15 CN CN202111350347.5A patent/CN114035693A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101079060A (en) * | 2007-03-26 | 2007-11-28 | 腾讯科技（深圳）有限公司 | Chinese character input simple 'pinyin' implementation method and system |
| CN109032375A (en) * | 2018-06-29 | 2018-12-18 | 北京百度网讯科技有限公司 | Candidate text sort method, device, equipment and storage medium |
| CN111090341A (en) * | 2019-12-24 | 2020-05-01 | 科大讯飞股份有限公司 | Input method candidate result display method, related equipment and readable storage medium |
| CN111488466A (en) * | 2020-04-16 | 2020-08-04 | 清华大学 | Chinese language error corpus generating method, calculating device and storage medium |
| CN112684910A (en) * | 2020-12-29 | 2021-04-20 | 维沃移动通信有限公司 | Input method candidate word display method and device and electronic equipment |
| CN112748811A (en) * | 2021-01-21 | 2021-05-04 | 李博林 | English word input method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11621940B2 (en) | Systems, devices, and methods for generating a domain name using a user in interface | |
| US8639684B2 (en) | Dynamic search box for web browser | |
| CN101292243B (en) | Removing documents from search results | |
| CA2153684C (en) | Combined dictionary based and likely character string method of handwriting recognition | |
| US20080294982A1 (en) | Providing relevant text auto-completions | |
| US9898464B2 (en) | Information extraction supporting apparatus and method | |
| KR20070047304A (en) | Named url entry | |
| CN110716991B (en) | Method for displaying entity associated information based on electronic book and electronic equipment | |
| CN104036004B (en) | Search for error correction method and search error correction device | |
| CN104102704B (en) | System control methods of exhibiting and device | |
| CN110968789A (en) | Electronic book pushing method, electronic equipment and computer storage medium | |
| EP3682350A1 (en) | Method for analysing digital documents | |
| CN106919593B (en) | Searching method and device | |
| CN111008519B (en) | Display method of reading page, electronic equipment and computer storage medium | |
| JP2002202984A (en) | Automatic text information sorter based on rule base model | |
| CN114035693A (en) | Pinyin input method and device, computer equipment and storage medium | |
| CN110059243B (en) | Data engine optimization method, device, equipment and computer readable storage medium | |
| CN110728113A (en) | Information screening method and device of electronic forms and terminal equipment | |
| EP3413210A1 (en) | Information search method, information search device and information search system | |
| US9087119B2 (en) | Method for searching in a plurality of data sets and search engine | |
| AU2020218504A1 (en) | Free text model explanation heat map | |
| CN111913646A (en) | Information sending method and device | |
| JP4496797B2 (en) | Document management apparatus and method | |
| JP2020113048A (en) | Information processing apparatus and program | |
| US11615155B1 (en) | Method of providing user interface for retrieving information on e-book and server using the same |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |