Input method, device and electronic device

By receiving the input sequence and using the candidate information returned by the server to display the target long word candidates, the problem of the existing input method being too short is solved, and the input efficiency and accuracy of candidates are improved.

CN112214114BActive Publication Date: 2024-11-01BEIJING SOGOU TECHNOLOGY DEVELOPMENT CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN201910631926.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2019-07-12
Publication Date
2024-11-01
Estimated Expiration
2039-07-12

AI Technical Summary

Technical Problem

When the existing input method is used to enter the pinyin string, the candidates are too short, resulting in low input efficiency. The user needs to continue to enter to obtain longer candidates, which cannot meet the needs.

Method used

By receiving the input sequence, the target long word candidate information is determined based on the first long word candidate information returned by the server, and displayed it to the user, using the preset model to improve the accuracy and number of candidates combined with historical input information.

Benefits of technology

Long term candidates that meet the needs can be obtained without the user inputting a long input sequence, which improves input efficiency and provides more accurate and diverse candidates.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112214114B_ABST
    Figure CN112214114B_ABST
Patent Text Reader

Abstract

An embodiment of the present invention provides an input method, device and electronic device, wherein the method includes: receiving an input sequence; determining target long word candidate information corresponding to the input sequence based on received first long word candidate information, wherein the first long word candidate information is determined by a server based on a preset model and returned in response to a received long word association request; displaying the target long word candidate information; and then being able to provide long word candidates that meet user needs without the user having to input a long input sequence, thereby improving input efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an input method, device and electronic equipment. Background Art

[0002] With the development of computer technology, electronic devices such as mobile phones and tablet computers are becoming more and more popular, bringing great convenience to people's life, study and work. These electronic devices are usually installed with input method applications (input methods for short), so that users can use the input methods to input information.

[0003] In the process of inputting using the input method, in order to facilitate user input, when the user enters the Pinyin string after entering the previous text, the input method usually uses the Pinyin string to generate candidates for the user to choose; for example, when the previous text is "what time are you", and the Pinyin string is "x", the input method provides candidates such as "next", "want", and "first". Due to limited client resources, the current input method can only provide candidates that match the number of syllables in the Pinyin string, and thus when the Pinyin string entered by the user is shorter, the provided candidates will also be shorter; so that when the short candidate cannot meet the user's needs, the user needs to continue to enter the Pinyin string to obtain a longer candidate, resulting in low input efficiency; for example, when the above candidates cannot meet the user's needs, if the user needs to enter "get off work", the user needs to enter the Pinyin string "xiab" before the input method can provide candidates such as "get off work" for the user to choose. Summary of the invention

[0004] The embodiment of the present invention provides an input method to improve input efficiency.

[0005] Correspondingly, an embodiment of the present invention further provides an input device and an electronic device to ensure the implementation and application of the above method.

[0006] In order to solve the above problems, an embodiment of the present invention discloses an input method, which specifically includes: receiving an input sequence; determining target long word candidate information corresponding to the input sequence based on the received first long word candidate information, wherein the first long word candidate information is determined by the server based on a preset model and returned in response to the received long word association request; and displaying the target long word candidate information.

[0007] Optionally, after obtaining the input sequence, the method further includes: obtaining input association information, the input association information including: above information and / or input environment information; generating a first long word association request based on the input sequence and the input association information and sending it to the server, so that the server returns the corresponding first long word candidate information.

[0008] Optionally, the method further includes: obtaining input association information, the input association information including: previous information and / or input environment information; generating a second long word association request based on the input association information and sending it to the server, so that the server returns the corresponding first long word candidate information.

[0009] Optionally, the first long word candidate information includes a predicted input sequence, a first long word candidate item corresponding to the preset input sequence, and candidate association information corresponding to the first long word candidate item; determining the target long word candidate information corresponding to the input sequence based on the received first long word candidate information includes: searching for a predicted input sequence that matches the input sequence; and determining the first long word candidate item corresponding to the matching preset input sequence and the candidate association information corresponding to the first long word candidate item as the target long word candidate information.

[0010] Optionally, the method further comprises: performing local long word association to determine second long word candidate information corresponding to the input sequence.

[0011] Optionally, the performing local long word association to determine the second long word candidate information corresponding to the input sequence includes: performing local long word association based on candidate association information corresponding to the target long word candidate item to determine the second long word candidate information corresponding to the input sequence.

[0012] Optionally, the target long word candidate information includes a target long word candidate item and corresponding candidate association information, and the second long word candidate information includes a second long word candidate item and corresponding candidate association information; the display of the target long word candidate information includes: sorting the target long word candidate item and the second long word candidate item according to the candidate association information corresponding to the target long word candidate item and the candidate association information corresponding to the second long word candidate item; and displaying the target long word candidate item and the second long word candidate item according to the sorting result.

[0013] Optionally, the target long word candidate information includes target long word candidate items and corresponding candidate association information; the displaying of the target long word candidate information includes: sorting the target long word candidate items according to the candidate association information corresponding to the target long word candidate items; and displaying the target long word candidate items according to the sorting result.

[0014] Optionally, displaying the target long word candidate information includes: retaining the target long word candidate information in a display area for a set time period.

[0015] Optionally, the method further includes: receiving an interactive instruction, and performing a corresponding interactive response operation on the target long word candidate information corresponding to the interactive instruction; the interactive response operation includes at least one of the following: screen display, splitting and reorganization, display update, and candidate expansion.

[0016] Optionally, the candidate association information of the target candidate item includes at least one of the following: the candidate score corresponding to the target long word candidate item, the length of the context corresponding to the target long word candidate item, the identifier of the optimal long word candidate item that fully matches the input sequence, and the type of input word corresponding to the input sequence.

[0017] Optionally, the input environment information includes at least one of the following: location information, time information, weather information, and application-related information.

[0018] An embodiment of the present invention also discloses an input method, which specifically includes: receiving a long word association request; determining first long word candidate information corresponding to the long word association request based on a preset model, wherein the preset model is trained based on historical input information; and returning the first long word candidate information.

[0019] Optionally, the preset model includes a preset long word association model, and determining the first long word candidate information corresponding to the long word association request based on the preset model includes: if the long word association request is a first long word association request, obtaining an input sequence and input association information from the first long word association request; inputting the input sequence and input association information into the preset long word association model to obtain the first long word candidate item and candidate association information of the first long word candidate item; generating the first long word candidate information corresponding to the first long word association request based on the first long word candidate item and the candidate association information of the first long word candidate item.

[0020] Optionally, the preset model includes a preset long word association model and a preset input sequence prediction model, and determining the first long word candidate information corresponding to the long word association request based on the preset model includes: if the long word association request is a second long word association request, obtaining input association information from the second long word association request; inputting the input association information into the preset input sequence prediction model to obtain a predicted input sequence; inputting the predicted input sequence and the corresponding input association information into the preset long word association model to obtain the first long word candidate corresponding to the predicted input sequence and the candidate association information of the first long word candidate; generating the first long word candidate information based on the predicted input sequence, the first long word candidate corresponding to the predicted input sequence and the candidate association information of the first long word candidate.

[0021] Optionally, the method further comprises: screening out long word candidates that meet preset conditions based on the candidate association information.

[0022] Optionally, the method also includes the step of training the preset long word association model: collecting multiple groups of historical input information, a group of historical input information includes a historical input sequence, historical input association information corresponding to the historical input sequence, and corresponding reference long word candidate information; for a group of the historical input information, inputting the historical input sequence in the historical input information and the historical input association information corresponding to the historical input sequence into the preset long word association model to obtain first long word candidate information; adjusting the weight of the preset long word association model based on comparing the first long word candidate information with the reference long word candidate information in the group of historical input information.

[0023] An embodiment of the present invention also discloses an input device, specifically comprising: a sequence receiving module, used to receive an input sequence; a candidate determination module, used to determine target long word candidate information corresponding to the input sequence based on the received first long word candidate information, wherein the first long word candidate information is determined by the server based on a preset model and returned in response to the received long word association request; and a candidate display module, used to display the target long word candidate information.

[0024] Optionally, the device also includes: a first request sending module, which is used to obtain input association information after obtaining the input sequence, and the input association information includes: above information and / or input environment information; based on the input sequence and the input association information, generate a first long word association request and send it to the server, so that the server returns the corresponding first long word candidate information.

[0025] Optionally, the device also includes: an information acquisition module, used to acquire input association information, the input association information including: above information and / or input environment information; a second request sending module, used to generate a second long word association request based on the input association information and send it to the server, so that the server returns the corresponding first long word candidate information.

[0026] Optionally, the first long word candidate information includes a predicted input sequence, a first long word candidate item corresponding to the preset input sequence, and candidate association information corresponding to the first long word candidate item; the candidate determination module is used to find a predicted input sequence that matches the input sequence; and determine the first long word candidate item corresponding to the matching preset input sequence and the candidate association information corresponding to the first long word candidate item as the target long word candidate information.

[0027] Optionally, the device further comprises: a local association module, configured to perform local long word association and determine second long word candidate information corresponding to the input sequence.

[0028] Optionally, the local association module is used to perform local long word association based on the candidate association information corresponding to the target long word candidate item, and determine the second long word candidate information corresponding to the input sequence.

[0029] Optionally, the target long word candidate information includes target long word candidate items and corresponding candidate association information, and the second long word candidate information includes second long word candidate items and corresponding candidate association information; the candidate display module includes: a first sorting submodule, used to sort the target long word candidate items and the second long word candidate items according to the candidate association information corresponding to the target long word candidate items and the candidate association information corresponding to the second long word candidate items; a first candidate item display submodule, used to display the target long word candidate items and the second long word candidate items according to the sorting result.

[0030] Optionally, the target long word candidate information includes target long word candidate items and corresponding candidate association information; the candidate display module includes: a second sorting submodule, used to sort the target long word candidate items according to the candidate association information corresponding to the target long word candidate items; a second candidate item display submodule, used to display the target long word candidate items according to the sorting results.

[0031] Optionally, the candidate display module includes: a display retention submodule, which is used to retain the target long word candidate information in the display area for a set time period.

[0032] Optionally, the device also includes: an interactive module, used to receive interactive instructions, and perform corresponding interactive response operations on the target long word candidate information corresponding to the interactive instructions; the interactive response operations include at least one of the following: screen display, splitting and reorganization, display update, and candidate expansion.

[0033] Optionally, the candidate association information of the target candidate item includes at least one of the following: the candidate score corresponding to the target long word candidate item, the length of the context corresponding to the target long word candidate item, the identifier of the optimal long word candidate item that fully matches the input sequence, and the type of input word corresponding to the input sequence.

[0034] Optionally, the input environment information includes at least one of the following: location information, time information, weather information, and application-related information.

[0035] An embodiment of the present invention also discloses an input device, which specifically includes: a request receiving module, used to receive a long word association request; a candidate generating module, used to determine the first long word candidate information corresponding to the long word association request according to a preset model, wherein the preset model is trained according to historical input information; and a candidate returning module, used to return the first long word candidate information.

[0036] Optionally, the preset model includes a preset long word association model, and the candidate generation module includes: a first candidate information generation submodule, which is used to obtain an input sequence and input association information from the first long word association request if the long word association request is a first long word association request; input the input sequence and input association information into the preset long word association model to obtain the first long word candidate item and the candidate association information of the first long word candidate item; and generate the first long word candidate information corresponding to the first long word association request based on the first long word candidate item and the candidate association information of the first long word candidate item.

[0037] Optionally, the preset model includes a preset long word association model and a preset input sequence prediction model, and the candidate generation module includes: a second candidate information generation submodule, which is used to obtain input association information from the second long word association request if the long word association request is a second long word association request; input the input association information into the preset input sequence prediction model to obtain a predicted input sequence; input the predicted input sequence and the corresponding input association information into the preset long word association model to obtain the first long word candidate item corresponding to the predicted input sequence and the candidate association information of the first long word candidate item; generate the first long word candidate information based on the predicted input sequence, the first long word candidate item corresponding to the predicted input sequence and the candidate association information of the first long word candidate item.

[0038] Optionally, the device further comprises: a screening module, configured to screen out long word candidates that meet preset conditions based on the candidate association information.

[0039] Optionally, the device also includes: a training module, used to collect multiple groups of historical input information, a group of historical input information includes a historical input sequence, historical input association information corresponding to the historical input sequence, and corresponding reference long word candidate information; for a group of historical input information, the historical input sequence in the historical input information and the historical input association information corresponding to the historical input sequence are input into the preset long word association model to obtain first long word candidate information; based on the comparison between the first long word candidate information and the reference long word candidate information in the group of historical input information, the weight of the preset long word association model is adjusted.

[0040] The embodiment of the present invention further discloses a readable storage medium. When instructions in the storage medium are executed by a processor of a terminal device, the terminal device can execute the input method as described in any one of the embodiments of the present invention.

[0041] The embodiment of the present invention further discloses a readable storage medium. When the instructions in the storage medium are executed by a processor of a server, the server can execute the input method as described in any one of the embodiments of the present invention.

[0042] An embodiment of the present invention also discloses a terminal device, including a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors, wherein the one or more programs include instructions for performing the following operations: receiving an input sequence; determining target long word candidate information corresponding to the input sequence based on the received first long word candidate information, wherein the first long word candidate information is determined by the server based on a preset model and returned in response to the received long word association request; and displaying the target long word candidate information.

[0043] Optionally, after obtaining the input sequence, it also includes instructions for performing the following operations: obtaining input association information, the input association information including: above information and / or input environment information; generating a first long word association request based on the input sequence and the input association information and sending it to the server, so that the server returns the corresponding first long word candidate information.

[0044] Optionally, it also includes instructions for performing the following operations: obtaining input association information, the input association information including: previous information and / or input environment information; generating a second long word association request based on the input association information and sending it to the server, so that the server returns the corresponding first long word candidate information.

[0045] Optionally, the first long word candidate information includes a predicted input sequence, a first long word candidate item corresponding to the preset input sequence, and candidate association information corresponding to the first long word candidate item; determining the target long word candidate information corresponding to the input sequence based on the received first long word candidate information includes: searching for a predicted input sequence that matches the input sequence; and determining the first long word candidate item corresponding to the matching preset input sequence and the candidate association information corresponding to the first long word candidate item as the target long word candidate information.

[0046] Optionally, it also includes instructions for performing the following operations: performing local long word association to determine second long word candidate information corresponding to the input sequence.

[0047] Optionally, the performing local long word association to determine the second long word candidate information corresponding to the input sequence includes: performing local long word association based on candidate association information corresponding to the target long word candidate item to determine the second long word candidate information corresponding to the input sequence.

[0048] Optionally, the target long word candidate information includes a target long word candidate item and corresponding candidate association information, and the second long word candidate information includes a second long word candidate item and corresponding candidate association information; the display of the target long word candidate information includes: sorting the target long word candidate item and the second long word candidate item according to the candidate association information corresponding to the target long word candidate item and the candidate association information corresponding to the second long word candidate item; and displaying the target long word candidate item and the second long word candidate item according to the sorting result.

[0049] Optionally, the target long word candidate information includes target long word candidate items and corresponding candidate association information; the displaying of the target long word candidate information includes: sorting the target long word candidate items according to the candidate association information corresponding to the target long word candidate items; and displaying the target long word candidate items according to the sorting result.

[0050] Optionally, displaying the target long word candidate information includes: retaining the target long word candidate information in a display area for a set time period.

[0051] Optionally, the terminal device further includes: receiving an interactive instruction, and performing a corresponding interactive response operation on the target long word candidate information corresponding to the interactive instruction; the interactive response operation includes at least one of the following: screen display, splitting and reorganization, display update, and candidate expansion.

[0052] Optionally, the candidate association information of the target candidate item includes at least one of the following: the candidate score corresponding to the target long word candidate item, the length of the context corresponding to the target long word candidate item, the identifier of the optimal long word candidate item that fully matches the input sequence, and the type of input word corresponding to the input sequence.

[0053] Optionally, the input environment information includes at least one of the following: location information, time information, weather information, and application-related information.

[0054] An embodiment of the present invention also provides a server, comprising a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors, wherein the one or more programs include instructions for performing the following operations: receiving a long word association request; determining first long word candidate information corresponding to the long word association request based on a preset model, wherein the preset model is trained based on historical input information; and returning the first long word candidate information.

[0055] Optionally, the preset model includes a preset long word association model, and determining the first long word candidate information corresponding to the long word association request based on the preset model includes: if the long word association request is a first long word association request, obtaining an input sequence and input association information from the first long word association request; inputting the input sequence and input association information into the preset long word association model to obtain the first long word candidate item and candidate association information of the first long word candidate item; generating the first long word candidate information corresponding to the first long word association request based on the first long word candidate item and the candidate association information of the first long word candidate item.

[0056] Optionally, the preset model includes a preset long word association model and a preset input sequence prediction model, and determining the first long word candidate information corresponding to the long word association request based on the preset model includes: if the long word association request is a second long word association request, obtaining input association information from the second long word association request; inputting the input association information into the preset input sequence prediction model to obtain a predicted input sequence; inputting the predicted input sequence and the corresponding input association information into the preset long word association model to obtain the first long word candidate corresponding to the predicted input sequence and the candidate association information of the first long word candidate; generating the first long word candidate information based on the predicted input sequence, the first long word candidate corresponding to the predicted input sequence and the candidate association information of the first long word candidate.

[0057] Optionally, it also includes instructions for performing the following operations: based on the candidate association information, screening out long word candidates that meet preset conditions.

[0058] Optionally, it also includes instructions for performing the following operations to train the preset long word association model: collecting multiple groups of historical input information, a group of historical input information includes a historical input sequence, historical input association information corresponding to the historical input sequence, and corresponding reference long word candidate information; for a group of the historical input information, inputting the historical input sequence in the historical input information and the historical input association information corresponding to the historical input sequence into the preset long word association model to obtain first long word candidate information; adjusting the weight of the preset long word association model based on comparing the first long word candidate information with the reference long word candidate information in the group of historical input information.

[0059] The embodiments of the present invention include the following advantages:

[0060] After receiving an input sequence, the client of the embodiment of the present invention can determine the target long word candidate information corresponding to the input sequence based on the received first long word candidate information, and then display the target long word candidate information; wherein, the first long word candidate information is determined and returned by the server based on a preset model, thereby being able to provide long word candidates that meet user needs without the user having to input a long input sequence, thereby improving input efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 is a flowchart of the steps of an embodiment of a client-side input method of the present invention;

[0062] Figure 2 is a flow chart of steps of an embodiment of a server-side input method of the present invention;

[0063] Figure 3 is a flow chart of steps of an input method embodiment of the present invention;

[0064] Figure 4 This is a flow chart of steps of an optional embodiment of an input method of the present invention.

[0065] Figure 5 is a structural block diagram of an embodiment of a client-side input device of the present invention;

[0066] Figure 6 is a structural block diagram of an optional embodiment of a client-side input device of the present invention;

[0067] Figure 7 is a structural block diagram of an embodiment of a server-side input device of the present invention;

[0068] Figure 8 This is a structural block diagram of an optional embodiment of a server-side input device of the present invention.

[0069] Fig. 9 A structural block diagram of an electronic device for input according to an exemplary embodiment;

[0070] Fig.10 It is a structural schematic diagram of an electronic device for input according to another exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0071] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0072] One of the core concepts of an embodiment of the present invention is that after receiving an input sequence, the client determines and displays the target long word candidate information corresponding to the input sequence based on the long word candidate information received from the server; wherein the long word candidate information is determined by the server based on a preset model trained using historical input information, and then the long word candidate information corresponding to the input sequence is determined by combining the historical input information, so that long word candidates that meet user needs can be provided, thereby improving input efficiency.

[0073] The client in the embodiment of the present invention refers to the client of the input method, and the server refers to the server of the input method.

[0074] The following describes an input method on the client side.

[0075] Reference Figure 1 , shows a flowchart of a client-side input method embodiment of the present invention, which may specifically include the following steps:

[0076] Step 102: Receive an input sequence.

[0077] Step 104: Determine target long word candidate information corresponding to the input sequence based on the received first long word candidate information, wherein the first long word candidate information is determined by the server based on a preset model and returned in response to the received long word association request.

[0078] Step 106: Display the target long word candidate information.

[0079] In the embodiment of the present invention, when a user uses an input method, the input method can directly receive user input content, and then convert the input content into word candidates of the input text, and provide them to the user for selection. The user input content directly received by the input method can be called an input sequence.

[0080] In actual processing, users can use a variety of input methods, such as pinyin input, stroke input, Wubi input, voice input, and handwriting input. Users can complete the input of the input sequence through any of the above input methods. For input methods such as pinyin input, stroke input, and Wubi input, the input sequence is usually a coded string entered by the user through a keyboard, touch screen, etc.; for handwriting input, the input sequence can be a movement trajectory entered by the user through a handwriting board, touch screen, etc. This application does not limit the user's input method, and the user can use any input method.

[0081] In an embodiment of the present invention, after receiving an input sequence, the client can match the corresponding target long word candidate information for the input sequence, and then display the target long word candidate information for user selection. In one implementation of the embodiment of the present invention, the client can send a long word association request to the server, and the server determines the first long word candidate information corresponding to the long word association request and returns it; then after the client receives the first long word candidate information returned by the server, it can match the corresponding target long word candidate information for the input sequence based on the first long word candidate information. In the embodiment, the client can send a long word association request to the server after receiving the input sequence to receive the first long word candidate information returned by the server; or before receiving the input sequence, the client can send a long word association request to the server to receive the first long word candidate information returned by the server, so as to avoid the problem that the input method client cannot display the long word candidate information corresponding to the input sequence in time due to network delay; the embodiment of the present invention does not limit this. In the embodiment, the server can determine the first long word candidate information based on a preset model, and the preset model is trained based on historical input information, which will be described in detail later.

[0082] In an embodiment of the present invention, the display area corresponding to the target long word candidate information may include multiple types, such as the keyboard area of ​​the input method such as the local candidate bar, the cloud candidate bar, and the designated area of ​​the application where the input method is located. The embodiment of the present invention does not limit this.

[0083] In an example of the present invention, after the user inputs the above "what time are you" and enters the pinyin "x", the client can receive an input sequence such as "x"; then, based on the received first long word candidate information, determine the target long word candidate information corresponding to the input sequence, for example, the first long word candidate information (hereinafter, the first long word candidate information including the first long word candidate items and candidate scores are used as an example for explanation) includes: "get off work-0.92", "get off get out of class-0.88", "get off the plane-0.52" and "get off the high-speed train-0.54"; the first long word candidate information can all be used as target long word candidate information, and then the target long word candidate information can be displayed, for example, target long word candidate items such as "get off work", "get off get out of class", "get off the plane" and "get off the high-speed train" are displayed.

[0084] In summary, after the client of the embodiment of the present invention receives an input sequence, it can determine the target long word candidate information corresponding to the input sequence based on the received first long word candidate information, and then display the target long word candidate information; wherein, the first long word candidate information is determined and returned by the server based on a preset model, thereby being able to provide long word candidates that meet user needs without the user having to input a long input sequence, thereby improving input efficiency.

[0085] The following describes an input method on the server side.

[0086] Reference Figure 2 , shows a flow chart of the steps of an embodiment of a server-side input method of the present invention, which may specifically include the following steps:

[0087] Step 202: Receive a long word association request.

[0088] Step 204: Determine first long word candidate information corresponding to the long word association request according to a preset model, wherein the preset model is trained according to historical input information.

[0089] Step 206: Return the first long word candidate information.

[0090] In an embodiment of the present invention, the server may collect training data in advance, and then use the training data to train the preset model; wherein the training data may include historical input information, and the historical input information may include information related to the input, such as historical input sequence, historical long word candidate information corresponding to the historical input sequence, and historical input association information, etc., which is not limited in the embodiment of the present invention. wherein the historical input association information may include multiple types, such as input environment information such as time information, location information, etc., and such as the above information, etc., which is not limited in the embodiment of the present invention. then, after the client sends a long word association request, the server may receive the long word association request, and then use the trained preset model to determine the first long word candidate information corresponding to the long word association request, and then return the first long word candidate information to the corresponding client according to the long word association request. after the client receives the first long word candidate information, it may determine the target long word candidate information corresponding to the input sequence according to the first long word candidate information and display it for user selection.

[0091] In summary, in an embodiment of the present invention, after receiving a long word association request, the server can determine the first long word candidate information corresponding to the long word association request based on a preset model, and then return it to the client, so that the client can determine and display the target long word candidate information corresponding to the input sequence; wherein the preset model is trained based on historical input information, so that the determined first long word candidate information is more accurate, thereby improving the accuracy of the target long word candidate information, and can better meet user needs.

[0092] The following is an example of a client sending a long word association request to a server after receiving an input sequence.

[0093] Reference Figure 3 , shows a flow chart of steps of an input method embodiment of the present invention, which may specifically include the following steps:

[0094] Step 302: The client receives an input sequence.

[0095] Step 304, the client obtains input association information, where the input association information includes: previous information and / or input environment information.

[0096] Step 306, the client generates a first long word association request based on the input sequence and the input association information, and sends it to the server.

[0097] In an embodiment of the present invention, after receiving an input sequence, the client can generate a long word association request and send the long word association request to the server to obtain first long word candidate information; wherein, the input sequence can be a single character such as a pinyin "x", an English letter such as "t", a stroke such as "丿", or a string of characters, such as a string of pinyin "xiak", a string of English letters such as "tha", a string of strokes such as "丿丨", and the embodiments of the present invention do not limit this. Among them, a long word association request can be sent to the server every time a character is received, or a long word association request can be sent to the server after determining that the user has completed entering the input sequence, and the embodiments of the present invention do not limit this.

[0098] In an implementation manner of an embodiment of the present invention, after receiving an input sequence, the client can obtain input association information, and then generate a long word association request (subsequently referred to as a first long word association request) based on the input sequence and the input association information to improve the accuracy of the server in determining the first long word candidate information. Wherein, the input association information can include previous information and / or input environment information, the previous information can refer to the information in the information editing box, and the input environment information can include at least one of the following: location information, time information, weather information, association information of application programs; the embodiments of the present invention do not limit this. Among them, the application programs can include various application programs on the terminal, such as instant messaging application programs, video application programs, music application programs, game application programs, memo, address book, itinerary, etc., and the association information of the application programs can also include various types, such as usage information of entertainment application programs such as video application programs, music application programs, game application programs, etc., memo information in the memo, communication information in the address book, itinerary information in the itinerary; when the current application program is a chat application, the association information of the application program can include chat records, grouping information of the peer user, etc.; the present invention does not limit this.

[0099] Of course, after receiving the input sequence, it is also possible to generate a first long word association request based only on the input association information and send it to the server, and the embodiments of the present invention do not limit this.

[0100] Step 308, the server receives the first long word association request.

[0101] Step 310, the server obtains the input sequence and the input association information from the first long word association request.

[0102] Step 312: The server inputs the input sequence and the input association information into a preset long word association model to obtain a first long word candidate item and candidate association information of the first long word candidate item.

[0103] Step 314: The server generates first long word candidate information corresponding to the first long word association request based on the first long word candidate item and the candidate association information of the first long word candidate item.

[0104] Step 316: The server returns the first long word candidate information.

[0105] Correspondingly, the server can receive the first long word association request; when the first long word association request is generated based on the input sequence and input association information, the server can obtain the input sequence and input association information from the first long word association request; and then determine the first long word candidate information corresponding to the first long word association request based on the input sequence and input association information.

[0106] In an embodiment of the present invention, the preset model may include a preset long word association model. The server may pre-train the preset long word association model, and then input the input sequence and input association information into the preset long word association model to obtain the first long word candidate item and the candidate association information of the first long word candidate item; then use the first long word candidate item and the candidate association information of the first long word candidate item to generate the first long word candidate information corresponding to the first long word association request and return it to the client.

[0107] The training process of the preset long word association model may include the following sub-steps:

[0108] Sub-step 22: Collect multiple groups of historical input information, where a group of historical input information includes a historical input sequence, historical input association information corresponding to the historical input sequence, and corresponding reference long word candidate information.

[0109] Sub-step 24: for a group of the historical input information, input the historical input sequence in the historical input information and the historical input association information corresponding to the historical input sequence into the preset long word association model to obtain first long word candidate information.

[0110] Sub-step 26: adjusting the weight of the preset long word association model based on the comparison between the first long word candidate information and the reference long word candidate information in the group of historical input information.

[0111] In an embodiment of the present invention, historical input information of users across the entire network can be collected, and a set of historical input information can include a historical input sequence, historical input association information corresponding to the historical input sequence, and corresponding reference long word candidate information. The historical input association information corresponding to the historical input sequence can include context information and / or input environment information, the reference long word candidate information includes reference long word candidate items and candidate association information corresponding to the reference long word candidate items, and the candidate association information corresponding to the reference long word candidate items can include: reference candidate scores of reference long word candidate items, reference context length corresponding to the reference long word candidate items and reference probability corresponding to the reference context length, reference type corresponding to the reference long word candidate items and reference probability corresponding to the reference type; of course, the candidate association information corresponding to the reference long word candidate items can also include other information, which is not limited in the embodiment of the present invention.

[0112] Among them, under the premise of the same historical input sequence and the same historical input association information, there may be multiple reference long word candidates, and the reference candidate score of the reference long word candidate, such as probability, can be determined according to the frequency of each reference long word candidate. The reference type corresponding to the reference long word candidate may include multiple types, such as the part of speech of the reference long word candidate, such as verb, noun, etc.; for example, if the reference long word candidate is a noun, the noun category to which the reference long word candidate belongs, such as name of a person, place name, dish name, etc.; for example, the emotional category to which the reference long word candidate belongs, such as sadness, happiness, sadness, etc., and the embodiment of the present invention does not limit this.

[0113] In addition, the embodiments of the present invention can also collect data through web crawling, manual operation, hot spot discovery, etc., and then convert the collected data into historical input data. For example, the hot topic "The Twelve Hours in Chang'an" can be converted into a set of historical input information: the input association information corresponding to the historical input sequence, such as the above information "Chang'an", the historical input sequence, such as the pinyin "s", the reference long word candidate, such as "Twelve Hours", the reference above length corresponding to the reference long word candidate, such as "2", and the reference type corresponding to the reference long word candidate, such as "time".

[0114] Then, the preset long word association model can be trained using the collected multiple groups of historical input information, and at least one group of historical input information can be input into the preset long word association model for training each time; the following uses the input of a group of historical input information into the preset long word association model for training for illustration. In an embodiment of the present invention, a historical input sequence in a group of historical input information and the historical input association information corresponding to the historical input sequence can be input into the preset long word association model for training, and the preset long word association model can output multiple groups of first long word candidate information; a group of first long word candidate information can include a first long word candidate and candidate association information corresponding to the first long word candidate, and the candidate association information corresponding to the first long word candidate can include: the candidate score of the first long word candidate, the preceding text length corresponding to the first long word candidate and the probability of the preceding text length, the type of input word corresponding to the input sequence and the probability corresponding to the type. Among them, the type of input word corresponding to the input sequence may refer to the type of input word predicted by a preset long word association model for the input sequence input therein. For example, the above information is "my friend", the input sequence is "1", and the predicted type of input word is a person's name; for another example, the above information is "I want to go", the input sequence is "1", and the predicted type of the word to be input is a place name.

[0115] Then, the first long word candidate information output by the preset long word association model is compared with the reference long word candidate information in the group of historical input information, and the preset long word association model is reversely trained to adjust the weight of the preset long word association model. Among the multiple groups of first long word candidate information output by the preset long word association model, a group of long word candidate information matching the reference long word candidate information in the group of historical input information can be searched, wherein the long word candidate information matching the first long word candidate item with the reference long word candidate item, the context length matching the reference context length, and the type matching the reference type can be determined as the first long word candidate information matching the reference long word candidate information. Then, the weight of the preset long word association model is adjusted with the goal of making the candidate score approach the reference candidate score, the probability of the context length approach the reference probability of the reference context length, and the probability of the type approach the reference probability of the reference type.

[0116] In an optional embodiment of the present invention, in order to make the preset long word association model converge quickly, the preset long word association model can be trained by using the historical input sequence, the historical input association information corresponding to the historical input sequence, the reference long word candidate corresponding to the historical input sequence, and a part of the candidate association information corresponding to the reference long word candidate (for example, only one kind of information, such as the reference candidate score corresponding to the reference long word candidate). Then, the historical input sequence, the historical input association information corresponding to the historical input sequence, the reference long word candidate corresponding to the historical input sequence, and another part of the candidate association information corresponding to the reference long word candidate are used to train other preset models. Among them, one kind of information in the other part of the candidate association information corresponding to the reference long word candidate can be used to train another preset model, and multiple other preset models can be obtained accordingly; or all the information in the other part of the candidate association information corresponding to the reference long word candidate can be used to train another preset model to obtain another preset model; the embodiment of the present invention does not limit this. Then, the trained preset long word association model outputs the first long word candidate and the candidate score corresponding to the first long word candidate, and other preset models can output other candidate association information corresponding to the first long word candidate.

[0117] Furthermore, when only the preset long word association model is trained, the input sequence and the input association information can be input into the trained preset long word association model to obtain the first long word candidate and the candidate association information of the first long word candidate. When the preset long word association model and other preset models are trained, the input sequence and the input association information can be respectively input into the trained preset long word association model and other preset models to obtain the first long word candidate and the candidate association information of the first long word candidate.

[0118] In an embodiment of the present invention, the first long word candidate items and the corresponding candidate association information output by the preset model can be used to generate the first long word candidate information corresponding to the first long word association request; then the first long word candidate information can be directly returned to the client; or the first long word candidate information can be screened and returned to the client to reduce bandwidth occupancy and occupied space.

[0119] In an example of the present invention, a method for screening the first long word candidate information may be to screen the first long word candidates in the first long word candidate information according to the candidate association information corresponding to the first long word candidates in the first long word candidate information. For example, the probabilities corresponding to various information in the candidate association information can be weighted, and the first long word candidates can be screened according to the corresponding weighted results; for example, the first N first long word candidates with the largest weighted results corresponding to the candidate association information are screened out, where N is a positive integer. For another example, the first long word candidates are screened according to the probability corresponding to the information with the largest weight in the candidate association information, and the first M first long word candidates with the largest probabilities are screened out, where M is a positive integer; the embodiments of the present invention do not limit this.

[0120] In addition, each first long word candidate in the screened first long word candidate information can be compared with the input sequence uploaded by the client to determine the optimal first long word candidate that completely matches the input sequence; then a first identifier such as "1" is added to the candidate association information corresponding to this first long word candidate, and the first identifier can be used to identify the optimal long word candidate that completely matches the input sequence, and a second identifier such as "0" is added to the candidate association information corresponding to other first long word candidates, and the second identifier can be used to identify long word candidates other than non-optimal long word candidates that completely match the input sequence. Among them, the complete match may mean that the input sequence corresponding to the first long word candidate completely matches the input sequence uploaded by the client. For example, if the input sequence uploaded by the client is "ya", and the first long word candidates are "要", "压", and "样子", then it can be determined that the first long word candidate "压" is the optimal first long word candidate that completely matches the input sequence. This facilitates subsequent display of the target long word candidate information based on this identifier.

[0121] When the first long word association request is generated based on the input association information, the method for the server to generate the first long word candidate information can refer to the method for the server to determine the first long word candidate information according to the second long word association request in the subsequent steps 408-414; this will not be elaborated here.

[0122] Step 318: The client receives the first long word candidate information.

[0123] Step 320: The client determines the target long word candidate information corresponding to the input sequence according to the received first long word candidate information.

[0124] Step 322: The client displays the target long word candidate information.

[0125] The client receives the first long word candidate information returned by the server, and then determines the first long word candidate information as the target long word candidate information corresponding to the input sequence, and displays the target long word candidate information, such as displaying the target long word candidate items in the target long word candidate information. The target long word candidate items can be displayed in a cloud candidate column, a local candidate column, or a designated area of ​​an application, which is not limited in the embodiment of the present invention.

[0126] Correspondingly, the target long word candidate information may include a target long word candidate item and corresponding candidate association information, and the candidate association information of the target long word candidate item includes at least one of the following: the candidate score corresponding to the target long word candidate item, the context length corresponding to the target long word candidate item, the optimal long word candidate item identifier that fully matches the input sequence, and the type of input word corresponding to the input sequence.

[0127] In an example of the present invention, a method for displaying the target long word candidate information may be to sort the target long word candidate items according to the candidate association information corresponding to the target long word candidate items; and to display the target long word candidate items according to the sorting result. Among them, one method for sorting the target long word candidate items may be that, for a target long word candidate item, the probability of various candidate association information corresponding to the target long word candidate item may be weighted; and then the target long word candidate items are sorted in descending order according to the weighted result. Another method for sorting the target long word candidate items may be to determine the candidate association information with the largest weight in the candidate association information corresponding to each target long word candidate item, and then sort the target long word candidate items according to the probability of the candidate association information with the largest weight. Of course, other sorting methods may also be included, and the embodiment of the present invention does not limit this. Among them, when the target long word candidate item with the first identifier in the candidate association information has the same sorting position as other target long word candidate items, the target long word candidate item with the first identifier in the candidate association information may be displayed before other target long word candidate items. In addition, it is also possible to display the long word candidate items with the first identifier in the candidate association information at the first screen of the candidate column or at the first position of the first screen without sorting, and the embodiments of the present invention are not limited to this.

[0128] In an optional embodiment of the present invention, the client may also perform local long word association to determine local long word candidate information corresponding to the input sequence, which may be subsequently referred to as second long word association information; the client may perform local long word association before receiving the input sequence, or may perform local long word association after receiving the input sequence, and the embodiment of the present invention does not impose any restrictions on this.

[0129] Among them, one way of associating local long words may be that the client directly performs local association based on local information (such as locally pre-stored models, vocabulary, etc.). Another way of associating local long words may be that after receiving the first long word candidate information returned by the server, the client combines the candidate association information in the first long word candidate information and the local information to perform local long word association. For example, when the type of input word corresponding to the input sequence in the candidate association information of the first long word candidate information is a name, an address book can be obtained, and then contacts matching the input sequence can be searched from the address book; for example, when the input sequence is the pinyin "l", matching contacts such as "Li**", "Liang**", etc. are screened out from the address book.

[0130] In an example of the present invention, after determining the target long word candidate information and the second long word candidate information, a display method may be to display the target long word candidate in the cloud candidate column or the designated area of ​​the application, and display the second long word candidate in the local candidate column. When displaying the second long word candidate in the local candidate column, the second long word candidate may be sorted according to the candidate association information of the second long word candidate, and then the sorted second long word candidate is displayed.

[0131] In another example of the present invention, both the target long word candidate and the second long word candidate can be displayed in the local candidate column; wherein, the target long word candidate and the second long word candidate can be sorted according to the candidate association information corresponding to the target long word candidate and the candidate association information corresponding to the second long word candidate; and according to the sorting result, the target long word candidate and the second long word candidate can be displayed.

[0132] Whether the target long word candidate and the second long word candidate are displayed separately or displayed together, the target long word candidate and the second long word candidate may be deduplicated before being displayed.

[0133] In summary, in an embodiment of the present invention, after receiving an input sequence, the client can obtain input association information, and then generate a first long word association request based on the input sequence and the input association information and send it to the server; after receiving the first long word association request, the server can determine the first long word candidate information based on the input sequence and the input association information in the first long word association request, thereby improving the accuracy of the first long word candidate information, and then return the first long word association request to the client; after receiving the first long word candidate information with high accuracy, the client can determine the target long word candidate information corresponding to the input sequence, thereby improving the accuracy of the target long word candidate information, and then display the target long word candidate information with high accuracy, so that the user can obtain long word candidates that meet the needs without entering a long input sequence, thereby improving input efficiency.

[0134] The following is an example of a client sending a long word association request to a server before receiving an input sequence.

[0135] Reference Figure 4 , shows a flow chart of steps of an optional embodiment of an input method of the present invention, which may specifically include the following steps:

[0136] Step 402: The client obtains input-related information, where the input-related information includes: the above information and / or input environment information.

[0137] Step 404: The client generates a second long word association request based on the input association information and sends it to the server.

[0138] In an embodiment of the present invention, in order to reduce network delay and the number of times long word association requests are sent, a long word association request may be sent to the server before the user inputs the input sequence to obtain the corresponding first long word candidate information. The client may obtain the input association information, and then generate a long word association request (hereinafter referred to as a second long word association request) based on the input association information and send it to the server.

[0139] Step 406: The server receives a second long word association request.

[0140] Step 408: The server obtains input association information from the second long word association request.

[0141] Step 410: The server inputs the input association information into a preset input sequence prediction model to obtain a predicted input sequence.

[0142] Step 412: Input the preset input sequence and the corresponding input association information into a preset long word association model to obtain the first long word candidate corresponding to the predicted input sequence and the candidate association information of the first long word candidate.

[0143] In the embodiment of the present invention, after receiving the second long word association request, the server may obtain input association information from the second long word association request, and then perform long word association according to the input association information to determine the corresponding first long word candidate information.

[0144] In one example of the present invention, the preset model may also include a preset input sequence prediction model, which may be pre-trained, and then the trained preset input sequence prediction model is used to predict the input sequence that the user may input based on input association information; and then the predicted input sequence and input association information are input into the trained preset long word association model to determine the corresponding first long word candidate information.

[0145] In an embodiment of the present invention, data (hereinafter referred to as sequence training data) for training a preset input sequence prediction model can be extracted from the historical input information obtained above, and the preset input sequence prediction model can be trained; wherein, a historical input sequence and corresponding historical input association information can be extracted from a set of historical input information, and then multiple sets of sequence training data can be obtained by extracting the historical input information. Then, the preset input sequence prediction model can be trained using the multiple sets of sequence training data collected, and at least one set of sequence training data can be input into the preset input sequence prediction model for training each time; the following is an explanation of inputting a set of sequence training data into a preset long word association model for training. Among them, the historical input association information in a set of sequence training data can be input into the preset input sequence prediction model to obtain multiple sets of predicted input sequences and corresponding probabilities; then, with the goal of making the probability of the predicted input sequence matching the historical input sequence in the set of sequence training data approach 1, the weight of the preset input sequence prediction model is adjusted.

[0146] Then, the input association information can be input into the trained preset input sequence prediction model to obtain multiple predicted input sequences and corresponding probabilities; and then each time the predicted input sequence and the corresponding input association information are input into the trained preset long word association model to obtain the first long word candidate corresponding to the pre-stored input sequence and the candidate association information of the first long word candidate.

[0147] In one example of the present invention, it is also possible that the prediction input sequence prediction model does not need to be pre-trained, but based on the historical input information obtained above, the mapping relationship between the historical input association information and the corresponding historical input sequence is configured; and then after obtaining the second long word association request, based on the input association information in the first long word association request, the pre-configured mapping relationship is searched to determine the corresponding historical input sequence; and then the historical input sequence determined based on the mapping relationship is determined as the predicted input sequence. Then, each time the predicted input sequence and the corresponding input association information are input into the trained preset long word association model, the first long word candidate corresponding to the pre-stored input sequence and the candidate association information of the first long word candidate are obtained.

[0148] Of course, other ways of determining the predicted input sequence may also be included, and the present invention is not limited thereto.

[0149] Step 414: Generate first long word candidate information according to the predicted input sequence, the first long word candidate corresponding to the predicted input sequence, and the candidate association information of the first long word candidate.

[0150] Step 416: Return the first long word candidate information.

[0151] In an embodiment of the present invention, the first long word candidate information can be generated by using the predicted input sequence, the probability corresponding to the predicted input sequence, the first long word candidate corresponding to the predicted input sequence, and the candidate association information of the first long word candidate; then the first long word candidate information can be directly returned to the client; or the first long word candidate information can be screened and returned to the client to reduce bandwidth occupancy and occupied space.

[0152] In one example of the present invention, a method for screening the first long word candidate information may be to screen the predicted input sequence according to the probability corresponding to the predicted input sequence in the first long word candidate information. It may be determined whether the probability corresponding to each predicted input sequence is greater than a probability threshold, and then the predicted input sequence with a probability greater than the probability threshold is screened out, and the predicted input sequence with a probability less than the probability threshold is filtered out; the probability threshold is determined according to demand, and the embodiment of the present invention does not limit this. Of course, the first long word candidate corresponding to the predicted input sequence with a probability greater than the probability threshold and the candidate relationship information corresponding to the first long word candidate are also screened out.

[0153] In addition, after filtering the predicted input sequence to obtain the first long word candidate information, the first long word candidate items corresponding to each predicted input sequence in the first long word candidate information can also be filtered. The filtering method is similar to the above-mentioned filtering method for the first long word candidate information determined based on the first long word association request, and will not be repeated here.

[0154] In addition, after the first long word candidate items corresponding to each predicted input sequence in the first long word candidate information are screened and the first long word candidate information is obtained, it is also possible to determine, for each predicted input sequence in the screened first long word candidate information, a first long word candidate item that completely matches the predicted input sequence among the long word candidate items corresponding to the predicted input sequence, and add a corresponding first identifier to the candidate association information corresponding to the first long word candidate item; the details are similar to the above and will not be repeated here.

[0155] Step 418: The client receives the first long word candidate information.

[0156] Step 420: The client receives an input sequence.

[0157] Step 422: The client determines target long word candidate information corresponding to the input sequence based on the received first long word candidate information.

[0158] In the embodiment of the present invention, since the first long word candidate in the first long word candidate information is determined based on the predicted input sequence, after the client receives the user's input sequence, it can determine the predicted input sequence that matches the input sequence, and then determine the target long word candidate information based on the matched predicted input sequence. Please refer to sub-steps 42-44:

[0159] Sub-step 42: The client searches for a predicted input sequence that matches the input sequence.

[0160] Sub-step 44: The client determines the first long word candidate corresponding to the matched preset input sequence and the candidate association information corresponding to the first long word candidate as target long word candidate information.

[0161] In an embodiment of the present invention, there may be multiple ways to search for a predicted input sequence that matches the input sequence. In one example of the present invention, one matching method may be prefix matching: each character in the predicted input sequence is matched with the characters in the input sequence from front to back, and the predicted input sequence with the longest prefix matching length with the input sequence is determined, and the predicted input sequence is determined as the predicted input sequence that matches the input sequence. For example, if the input sequence is "bu", the predicted input sequence includes: "buz", "b" and "bz"; "buz" has the longest prefix matching length with "bu", and "buz" can be determined as the predicted input sequence that matches the input sequence.

[0162] In another example of the present invention, a matching method may be syllable matching: determine the syllables corresponding to each predicted input sequence and the syllables corresponding to the input sequence, and for a predicted input sequence, match the syllables of the predicted input sequence with the syllables corresponding to the input sequence to determine the number of syllables that match the predicted input sequence with the input sequence; then determine the predicted input sequence with the largest number of syllable matches as the preset input sequence that matches the input sequence. For example, for the input sequence "buz", the predicted input sequences include: "b", "bu" and "bz", and "bz" has the largest number of syllable matches with "buz", so "bz" can be determined as the predicted input sequence that matches the input sequence.

[0163] Of course, other ways of matching the preset input sequence with the input sequence may also be included, and the embodiment of the present invention is not limited to this.

[0164] Step 424: The client displays the target long word candidate information.

[0165] This step 424 is similar to the above step 322 and will not be described again.

[0166] In summary, in an embodiment of the present invention, before receiving an input sequence, the client can obtain input association information, and then generate a second long word association request based on the input association information and send it to the server; after receiving the second long word association request, the server can generate the first long word candidate information based on the input sequence in the second long word association request, thereby improving the accuracy of the first long word candidate information, and then return the first long word association request to the client, thereby avoiding the untimely display of long word candidates due to network delays; after receiving the first long word candidate information with high accuracy, the client can determine the target long word candidate information corresponding to the input sequence, thereby improving the accuracy of the target long word candidate information, and then display the target long word candidate information with high accuracy, so that the user can obtain long word candidates that meet the needs without entering a long input sequence, thereby improving input efficiency.

[0167] In an optional embodiment of the present invention, the user often misses the long word candidate due to too fast input during the input process, so after the target long word candidate information and / or the second long word candidate information corresponding to the input sequence is determined and displayed, the target long word candidate information and / or the second long word candidate information can be retained in the display area for a set time. The set time can be set according to demand, and the embodiment of the present invention does not limit this.

[0168] In an optional embodiment of the present invention, after the client displays the target long word candidate information, the user can perform interactive operations on the displayed target long word candidate information, such as screen operation (such as clicking on the corresponding target long word candidate), operation of displaying multiple long word candidate items (such as performing a sliding operation in the candidate column), candidate splitting and reorganizing operation (such as performing a long press operation on the corresponding target long word candidate item), operation of updating long word candidates (such as clicking on a preset mark), etc. Correspondingly, the client can receive an interactive instruction and perform a corresponding interactive response operation on the target long word candidate information corresponding to the interactive instruction; the interactive response operation includes at least one of the following: screen, splitting and reorganizing, display update, and candidate expansion. The screen can refer to the target long word candidate corresponding to the interactive instruction being screened from the candidate column to the edit box or directly sent. The splitting and reorganizing can include reorganizing the displayed long word candidate items, such as selecting a part of them, or changing the order of each word, etc. The display update can refer to updating the long word candidate items displayed on the first screen, or it can refer to re-sending and updating the displayed long word candidate items according to the first long word candidate information returned by the server. The candidate expansion can refer to the operation of expanding the candidate column to view more long word candidates. Of course, the user can also perform other interactive operations to achieve other interactive functions, such as performing an operation of displaying other information of the long word candidate (such as performing a re-pressing operation (3D touch) on the corresponding target long word candidate), and the corresponding client can perform an interactive response operation to display other information corresponding to the long word candidate, such as the source, explanation, etc. Of course, when the local long word association is performed, the user can also perform an interactive operation on the displayed second long word candidate, which is not limited in the embodiment of the present invention.

[0169] It should be noted that, for the sake of simplicity, the method embodiments are described as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0170] Reference Figure 5 , shows a structural block diagram of an embodiment of a client-side input device of the present invention, which may specifically include the following modules:

[0171] A sequence receiving module 502, used for receiving an input sequence;

[0172] A candidate determination module 504 is used to determine target long word candidate information corresponding to the input sequence according to the received first long word candidate information, wherein the first long word candidate information is determined by the server according to a preset model and returned in response to the received long word association request;

[0173] The candidate display module 506 is used to display the target long word candidate information.

[0174] Reference Figure 6 , showing a structural block diagram of an optional embodiment of a client-side input device of the present invention.

[0175] In an optional embodiment of the present invention, the device further comprises:

[0176] The first request sending module 508 is used to obtain input association information after obtaining the input sequence, and the input association information includes: previous information and / or input environment information; based on the input sequence and the input association information, generate a first long word association request and send it to the server, so that the server returns the corresponding first long word candidate information.

[0177] In an optional embodiment of the present invention, the device further comprises:

[0178] The information acquisition module 510 is used to acquire input-related information, where the input-related information includes: the above information and / or input environment information;

[0179] The second request sending module 512 is used to generate a second long word association request according to the input association information and send the request to the server, so that the server returns the corresponding first long word candidate information.

[0180] In an optional embodiment of the present invention, the first long word candidate information includes a predicted input sequence, a first long word candidate corresponding to the preset input sequence, and candidate association information corresponding to the first long word candidate;

[0181] The candidate determination module 504 is used to search for a predicted input sequence matching the input sequence; and determine the first long word candidate corresponding to the matching preset input sequence and the candidate association information corresponding to the first long word candidate as target long word candidate information.

[0182] In an optional embodiment of the present invention, the device further comprises:

[0183] The local association module 514 is used to perform local long word association to determine the second long word candidate information corresponding to the input sequence.

[0184] In an optional embodiment of the present invention, the local association module 514 is used to perform local long word association based on the candidate association information corresponding to the target long word candidate item, and determine the second long word candidate information corresponding to the input sequence.

[0185] In an optional embodiment of the present invention, the target long word candidate information includes target long word candidate items and corresponding candidate association information, and the second long word candidate information includes second long word candidate items and corresponding candidate association information; the candidate display module 506 includes:

[0186] A first sorting submodule 5062 is used to sort the target long word candidate and the second long word candidate according to the candidate association information corresponding to the target long word candidate and the candidate association information corresponding to the second long word candidate;

[0187] The first candidate item display submodule 5064 is used to display the target long word candidate items and the second long word candidate items according to the sorting result.

[0188] In an optional embodiment of the present invention, the target long word candidate information includes target long word candidate items and corresponding candidate association information; the candidate display module 506 includes:

[0189] The second sorting submodule 5066 is used to sort the target long word candidate items according to the candidate association information corresponding to the target long word candidate items;

[0190] The second candidate item display submodule 5068 is used to display the target long word candidate items according to the sorting result.

[0191] In an optional embodiment of the present invention, the candidate display module 506 includes:

[0192] The display retention submodule 50610 is used to retain the target long word candidate information in the display area for a set time period.

[0193] In an optional embodiment of the present invention, the device further comprises:

[0194] The interactive module 516 is used to receive interactive instructions and perform corresponding interactive response operations on the target long word candidate information corresponding to the interactive instructions; the interactive response operations include at least one of the following: screen display, splitting and reorganization, display update, and candidate expansion.

[0195] In an optional embodiment of the present invention, the candidate association information of the target candidate item includes at least one of the following: the candidate score corresponding to the target long word candidate item, the length of the previous context corresponding to the target long word candidate item, the identifier of the optimal long word candidate item that completely matches the input sequence, and the type of input word corresponding to the input sequence.

[0196] In an optional embodiment of the present invention, the input environment information includes at least one of the following: location information, time information, weather information, and application-related information.

[0197] Reference Figure 7 , shows a structural block diagram of an embodiment of a server-side input device of the present invention, which may specifically include the following modules:

[0198] A request receiving module 702, for receiving a long word association request;

[0199] A candidate generation module 704 is used to determine first long word candidate information corresponding to the long word association request according to a preset model, wherein the preset model is trained according to historical input information;

[0200] The candidate returning module 706 is used to return the first long word candidate information.

[0201] Reference Figure 8 , showing a structural block diagram of an optional embodiment of a server-side input device of the present invention.

[0202] In an optional embodiment of the present invention, the preset model includes a preset long word association model, and the candidate generation module 704 includes:

[0203] The first candidate information generating submodule 7042 is used for obtaining an input sequence and input association information from the first long word association request if the long word association request is a first long word association request; inputting the input sequence and input association information into a preset long word association model to obtain the first long word candidate item and the candidate association information of the first long word candidate item; and generating the first long word candidate information corresponding to the first long word association request based on the first long word candidate item and the candidate association information of the first long word candidate item.

[0204] In an optional embodiment of the present invention, the preset model includes a preset long word association model and a preset input sequence prediction model, and the candidate generation module 704 includes:

[0205] The second candidate information generating submodule 7044 is used for obtaining input association information from the second long word association request if the long word association request is a second long word association request; inputting the input association information into a preset input sequence prediction model to obtain a predicted input sequence; inputting the predicted input sequence and the corresponding input association information into a preset long word association model to obtain the first long word candidate corresponding to the predicted input sequence and the candidate association information of the first long word candidate; generating the first long word candidate information based on the predicted input sequence, the first long word candidate corresponding to the predicted input sequence and the candidate association information of the first long word candidate.

[0206] In an optional embodiment of the present invention, the device further comprises:

[0207] The screening module 708 is used to screen out long word candidates that meet preset conditions based on the candidate association information.

[0208] In an optional embodiment of the present invention, the device further comprises:

[0209] The training module 710 is used to collect multiple groups of historical input information, where a group of historical input information includes a historical input sequence, historical input association information corresponding to the historical input sequence, and corresponding reference long word candidate information; for a group of the historical input information, the historical input sequence in the historical input information and the historical input association information corresponding to the historical input sequence are input into the preset long word association model to obtain first long word candidate information; and the weight of the preset long word association model is adjusted based on the comparison between the first long word candidate information and the reference long word candidate information in the group of historical input information.

[0210] After receiving an input sequence, the client of the embodiment of the present invention can determine the target long word candidate information corresponding to the input sequence based on the received first long word candidate information, and then display the target long word candidate information; wherein, the first long word candidate information is determined and returned by the server based on a preset model, thereby being able to provide long word candidates that meet user needs without the user having to input a long input sequence, thereby improving input efficiency.

[0211] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0212] Fig. 9 1 is a block diagram of an electronic device 900 for input according to an exemplary embodiment. For example, the electronic device 900 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0213] Reference Fig. 9 , the electronic device 900 may include one or more of the following components: a processing component 902 , a memory 904 , a power component 906 , a multimedia component 908 , an audio component 910 , an input / output (I / O) interface 912 , a sensor component 914 , and a communication component 916 .

[0214] The processing component 902 generally controls the overall operation of the electronic device 900, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 902 may include one or more processors 920 to execute instructions to complete all or part of the steps of the above-mentioned method. In addition, the processing component 902 may include one or more modules to facilitate the interaction between the processing component 902 and other components. For example, the processing component 902 may include a multimedia module to facilitate the interaction between the multimedia component 908 and the processing component 902.

[0215] The memory 904 is configured to store various types of data to support operations on the device 900. Examples of such data include instructions for any application or method operating on the electronic device 900, contact data, phone book data, messages, pictures, videos, etc. The memory 904 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0216] The power component 906 provides power to the various components of the electronic device 900. The power component 906 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the electronic device 900.

[0217] The multimedia component 908 includes a screen that provides an output interface between the electronic device 900 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 908 includes a front camera and / or a rear camera. When the electronic device 900 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0218] The audio component 910 is configured to output and / or input audio signals. For example, the audio component 910 includes a microphone (MIC), and when the electronic device 900 is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 904 or sent via the communication component 916. In some embodiments, the audio component 910 also includes a speaker for outputting audio signals.

[0219] I / O interface 912 provides an interface between processing component 902 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: a home button, a volume button, a start button, and a lock button.

[0220] The sensor assembly 914 includes one or more sensors for providing various aspects of status assessment for the electronic device 900. For example, the sensor assembly 914 can detect the open / closed state of the device 900, the relative positioning of components, such as the display and keypad of the electronic device 900, and the sensor assembly 914 can also detect the position change of the electronic device 900 or a component of the electronic device 900, the presence or absence of user contact with the electronic device 900, the orientation or acceleration / deceleration of the electronic device 900, and the temperature change of the electronic device 900. The sensor assembly 914 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 914 may also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 914 may also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0221] The communication component 916 is configured to facilitate wired or wireless communication between the electronic device 900 and other devices. The electronic device 900 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 914 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 914 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0222] In an exemplary embodiment, the electronic device 900 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above methods.

[0223] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 904 including instructions, and the instructions can be executed by a processor 920 of an electronic device 900 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0224] A non-temporary computer-readable storage medium, when the instructions in the storage medium are executed by a processor of a terminal device, enables the terminal device to execute an input method, the method comprising: receiving an input sequence; determining target long word candidate information corresponding to the input sequence based on received first long word candidate information, wherein the first long word candidate information is determined by a server based on a preset model and returned in response to a received long word association request; and displaying the target long word candidate information.

[0225] Optionally, after obtaining the input sequence, it also includes instructions for performing the following operations: obtaining input association information, the input association information including: above information and / or input environment information; generating a first long word association request based on the input sequence and the input association information and sending it to the server, so that the server returns the corresponding first long word candidate information.

[0226] Optionally, it also includes instructions for performing the following operations: obtaining input association information, the input association information including: previous information and / or input environment information; generating a second long word association request based on the input association information and sending it to the server, so that the server returns the corresponding first long word candidate information.

[0227] Optionally, the first long word candidate information includes a predicted input sequence, a first long word candidate item corresponding to the preset input sequence, and candidate association information corresponding to the first long word candidate item; determining the target long word candidate information corresponding to the input sequence based on the received first long word candidate information includes: searching for a predicted input sequence that matches the input sequence; and determining the first long word candidate item corresponding to the matching preset input sequence and the candidate association information corresponding to the first long word candidate item as the target long word candidate information.

[0228] Optionally, it also includes instructions for performing the following operations: performing local long word association to determine second long word candidate information corresponding to the input sequence.

[0229] Optionally, the performing local long word association to determine the second long word candidate information corresponding to the input sequence includes: performing local long word association based on candidate association information corresponding to the target long word candidate item to determine the second long word candidate information corresponding to the input sequence.

[0230] Optionally, the target long word candidate information includes a target long word candidate item and corresponding candidate association information, and the second long word candidate information includes a second long word candidate item and corresponding candidate association information; the display of the target long word candidate information includes: sorting the target long word candidate item and the second long word candidate item according to the candidate association information corresponding to the target long word candidate item and the candidate association information corresponding to the second long word candidate item; and displaying the target long word candidate item and the second long word candidate item according to the sorting result.

[0231] Optionally, the target long word candidate information includes target long word candidate items and corresponding candidate association information; the displaying of the target long word candidate information includes: sorting the target long word candidate items according to the candidate association information corresponding to the target long word candidate items; and displaying the target long word candidate items according to the sorting result.

[0232] Optionally, displaying the target long word candidate information includes: retaining the target long word candidate information in a display area for a set time period.

[0233] Optionally, the terminal device further includes: receiving an interactive instruction, and performing a corresponding interactive response operation on the target long word candidate information corresponding to the interactive instruction; the interactive response operation includes at least one of the following: screen display, splitting and reorganization, display update, and candidate expansion.

[0234] Optionally, the candidate association information of the target candidate item includes at least one of the following: the candidate score corresponding to the target long word candidate item, the length of the context corresponding to the target long word candidate item, the identifier of the optimal long word candidate item that fully matches the input sequence, and the type of input word corresponding to the input sequence.

[0235] Optionally, the input environment information includes at least one of the following: location information, time information, weather information, and application-related information.

[0236] Fig.10 1 is a schematic diagram of the structure of an electronic device 1000 for input according to another exemplary embodiment of the present invention. The electronic device 1000 may be a server, which may have relatively large differences due to different configurations or performances, and may include one or more central processing units (CPU) 1022 (for example, one or more processors) and a memory 1032, and one or more storage media 1030 (for example, one or more mass storage devices) storing application programs 1042 or data 1044. Among them, the memory 1032 and the storage medium 1030 may be temporary storage or permanent storage. The program stored in the storage medium 1030 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the server. Furthermore, the central processing unit 1022 may be configured to communicate with the storage medium 1030 and execute a series of instruction operations in the storage medium 1030 on the server.

[0237] The server may also include one or more power supplies 1026, one or more wired or wireless network interfaces 1050, one or more input and output interfaces 1058, one or more keyboards 1056, and / or one or more operating systems 1041, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.

[0238] A non-temporary computer-readable storage medium, when the instructions in the storage medium are executed by a processor of a server, enables the server to execute an input method, the method comprising: receiving a long word association request; determining first long word candidate information corresponding to the long word association request based on a preset model, wherein the preset model is trained based on historical input information; and returning the first long word candidate information.

[0239] Optionally, the preset model includes a preset long word association model, and determining the first long word candidate information corresponding to the long word association request based on the preset model includes: if the long word association request is a first long word association request, obtaining an input sequence and input association information from the first long word association request; inputting the input sequence and input association information into the preset long word association model to obtain the first long word candidate item and candidate association information of the first long word candidate item; generating the first long word candidate information corresponding to the first long word association request based on the first long word candidate item and the candidate association information of the first long word candidate item.

[0240] Optionally, the preset model includes a preset long word association model and a preset input sequence prediction model, and determining the first long word candidate information corresponding to the long word association request based on the preset model includes: if the long word association request is a second long word association request, obtaining input association information from the second long word association request; inputting the input association information into the preset input sequence prediction model to obtain a predicted input sequence; inputting the predicted input sequence and the corresponding input association information into the preset long word association model to obtain the first long word candidate corresponding to the predicted input sequence and the candidate association information of the first long word candidate; generating the first long word candidate information based on the predicted input sequence, the first long word candidate corresponding to the predicted input sequence and the candidate association information of the first long word candidate.

[0241] Optionally, it also includes instructions for performing the following operations: based on the candidate association information, screening out long word candidates that meet preset conditions.

[0242] Optionally, it also includes instructions for performing the following operations to train the preset long word association model: collecting multiple groups of historical input information, a group of historical input information includes a historical input sequence, historical input association information corresponding to the historical input sequence, and corresponding reference long word candidate information; for a group of the historical input information, inputting the historical input sequence in the historical input information and the historical input association information corresponding to the historical input sequence into the preset long word association model to obtain first long word candidate information; adjusting the weight of the preset long word association model based on comparing the first long word candidate information with the reference long word candidate information in the group of historical input information.

[0243] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0244] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0245] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0246] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0247] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0248] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0249] The above is a detailed introduction to an input method, an input device and an electronic device provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. An input method, characterized in that: include: Receive an input sequence; Acquire input-related information, where the input-related information includes: the above information and / or input environment information; According to the input association information, a second long word association request is generated and sent to the server, so that the server returns the corresponding first long word candidate information; Determining target long word candidate information corresponding to the input sequence according to the received first long word candidate information, wherein the first long word candidate information is determined by the server according to a preset model and returned in response to the received long word association request; Displaying the target long word candidate information; The preset model includes a preset long word association model and a preset input sequence prediction model, and the server determines and returns the first long word candidate information according to the preset model for the received long word association request, including: If the long word association request is a second long word association request, acquiring input association information from the second long word association request; Inputting the input association information into a preset input sequence prediction model to obtain a predicted input sequence; Inputting the predicted input sequence and the corresponding input association information into a preset long word association model to obtain a first long word candidate corresponding to the predicted input sequence and candidate association information of the first long word candidate; The first long word candidate information is generated and returned according to the predicted input sequence, the first long word candidate corresponding to the predicted input sequence and the candidate association information of the first long word candidate.

2. The method according to claim 1, characterized in that The first long word candidate information includes a predicted input sequence, a first long word candidate corresponding to the predicted input sequence, and candidate association information corresponding to the first long word candidate; The step of determining target long word candidate information corresponding to the input sequence based on the received first long word candidate information includes: Finding a predicted input sequence that matches the input sequence; The first long word candidate corresponding to the matched predicted input sequence and the candidate associated information corresponding to the first long word candidate are determined as target long word candidate information.

3. The method according to claim 2, characterized in that The method further comprises: Perform local long word association to determine second long word candidate information corresponding to the input sequence.

4. The method according to claim 3, characterized in that The performing of local long word association to determine second long word candidate information corresponding to the input sequence includes: Perform local long word association based on the candidate association information corresponding to the first long word candidate item to determine the second long word candidate information corresponding to the input sequence.

5. The method according to claim 3 or 4, characterized in that: The target long word candidate information includes target long word candidates and corresponding candidate association information, and the second long word candidate information includes second long word candidates and corresponding candidate association information; The displaying of the target long word candidate information includes: sorting the target long word candidate and the second long word candidate according to the candidate association information corresponding to the target long word candidate and the candidate association information corresponding to the second long word candidate; According to the ranking result, the target long word candidate items and the second long word candidate items are displayed.

6. The method according to claim 1, characterized in that The target long word candidate information includes target long word candidates and corresponding candidate association information; The displaying of the target long word candidate information includes: sorting the target long word candidate items according to the candidate association information corresponding to the target long word candidate items; According to the sorting result, the target long word candidates are displayed.

7. The method according to claim 1, characterized in that The displaying of the target long word candidate information includes: The target long word candidate information is retained in the display area for a set period of time.

8. The method according to claim 1, characterized in that The method further comprises: receiving an interactive instruction, and performing a corresponding interactive response operation on the target long word candidate information corresponding to the interactive instruction; The interactive response operation includes at least one of the following: screen display, splitting and reorganization, display update, and candidate expansion.

9. The method according to claim 5, characterized in that The candidate association information of the target long word candidate includes at least one of the following: the candidate score corresponding to the target long word candidate, the context length corresponding to the target long word candidate, the identifier of the optimal long word candidate that fully matches the input sequence, and the type of input word corresponding to the input sequence.

10. The method according to claim 1, characterized in that The input environment information includes at least one of the following: location information, time information, weather information, and application-related information.

11. An input method, characterized in that: include: Receive long word association requests; Determining first long word candidate information corresponding to the long word association request according to a preset model, wherein the preset model is trained according to historical input information; Return the first long word candidate information; The preset model includes a preset long word association model and a preset input sequence prediction model, and determining the first long word candidate information corresponding to the long word association request according to the preset model includes: If the long word association request is a second long word association request, acquiring input association information from the second long word association request; Inputting the input association information into a preset input sequence prediction model to obtain a predicted input sequence; Inputting the predicted input sequence and the corresponding input association information into a preset long word association model to obtain a first long word candidate corresponding to the predicted input sequence and candidate association information of the first long word candidate; First long word candidate information is generated according to the predicted input sequence, the first long word candidate corresponding to the predicted input sequence, and candidate association information of the first long word candidate.

12. The method according to claim 11, characterized in that The method further comprises: Based on the candidate association information, long word candidates that meet preset conditions are screened out.

13. The method according to claim 11, characterized in that The method further comprises the step of training the preset long word association model: Collecting multiple groups of historical input information, where a group of historical input information includes a historical input sequence, historical input association information corresponding to the historical input sequence, and corresponding reference long word candidate information; For a set of the historical input information, inputting the historical input sequence in the historical input information and the historical input association information corresponding to the historical input sequence into the preset long word association model to obtain first long word candidate information; The weight of the preset long word association model is adjusted based on the comparison between the first long word candidate information and the reference long word candidate information in the group of historical input information.

14. An input device, characterized in that: include: A sequence receiving module, used for receiving an input sequence; An information acquisition module, used to acquire input-related information, wherein the input-related information includes: the above information and / or input environment information; A second request sending module, used to generate a second long word association request according to the input association information and send it to the server, so that the server returns the corresponding first long word candidate information; A candidate determination module, configured to determine target long word candidate information corresponding to the input sequence according to the received first long word candidate information, wherein the first long word candidate information is determined by the server according to a preset model and returned in response to the received long word association request; A candidate display module, used to display the candidate information of the target long word; The preset model includes a preset long word association model and a preset input sequence prediction model, and the server determines and returns the first long word candidate information according to the preset model for the received long word association request, including: If the long word association request is a second long word association request, acquiring input association information from the second long word association request; Inputting the input association information into a preset input sequence prediction model to obtain a predicted input sequence; Inputting the predicted input sequence and the corresponding input association information into a preset long word association model to obtain a first long word candidate corresponding to the predicted input sequence and candidate association information of the first long word candidate; The first long word candidate information is generated and returned according to the predicted input sequence, the first long word candidate corresponding to the predicted input sequence and the candidate association information of the first long word candidate.

15. The device according to claim 14, characterized in that The first long word candidate information includes a predicted input sequence, a first long word candidate corresponding to the predicted input sequence, and candidate association information corresponding to the first long word candidate; The candidate determination module is used to find a predicted input sequence matching the input sequence; and determine the first long word candidate corresponding to the matched predicted input sequence and the candidate association information corresponding to the first long word candidate as target long word candidate information.

16. The device according to claim 15, characterized in that The device also includes: The local association module is used to perform local long word association and determine the second long word candidate information corresponding to the input sequence.

17. The device according to claim 16, characterized in that The local association module is used to perform local long word association based on the candidate association information corresponding to the first long word candidate item, and determine the second long word candidate information corresponding to the input sequence.

18. The device according to claim 16 or 17, characterized in that The target long word candidate information includes target long word candidates and corresponding candidate association information, and the second long word candidate information includes second long word candidates and corresponding candidate association information; The candidate display module includes: A first sorting submodule, configured to sort the target long word candidate and the second long word candidate according to the candidate association information corresponding to the target long word candidate and the candidate association information corresponding to the second long word candidate; The first candidate item display submodule is used to display the target long word candidate items and the second long word candidate items according to the sorting result.

19. The device according to claim 14, characterized in that The target long word candidate information includes target long word candidates and corresponding candidate association information; The candidate display module includes: A second sorting submodule, used to sort the target long word candidate items according to the candidate association information corresponding to the target long word candidate items; The second candidate item display submodule is used to display the target long word candidate items according to the sorting result.

20. The device according to claim 14, characterized in that The candidate display module includes: The display and retention submodule is used to retain the target long word candidate information in the display area for a set time period.

21. The device according to claim 14, characterized in that The device also includes: An interactive module, used for receiving an interactive instruction and performing a corresponding interactive response operation on the target long word candidate information corresponding to the interactive instruction; The interactive response operation includes at least one of the following: screen display, splitting and reorganization, display update, and candidate expansion.

22. The device according to claim 18, characterized in that The candidate association information of the target long word candidate includes at least one of the following: the candidate score corresponding to the target long word candidate, the context length corresponding to the target long word candidate, the identifier of the optimal long word candidate that fully matches the input sequence, and the type of input word corresponding to the input sequence.

23. The device according to claim 14, characterized in that The input environment information includes at least one of the following: location information, time information, weather information, and application-related information.

24. An input device, characterized in that: include: A request receiving module, used for receiving a long word association request; A candidate generation module, used to determine first long word candidate information corresponding to the long word association request according to a preset model, wherein the preset model is trained according to historical input information; A candidate returning module, used for returning the first long word candidate information; The preset model includes a preset long word association model and a preset input sequence prediction model, and the candidate generation module includes: The second candidate information generating submodule is used for obtaining input association information from the second long word association request if the long word association request is a second long word association request; inputting the input association information into a preset input sequence prediction model to obtain a predicted input sequence; inputting the predicted input sequence and the corresponding input association information into a preset long word association model to obtain a first long word candidate item corresponding to the predicted input sequence and candidate association information of the first long word candidate item; generating first long word candidate information based on the predicted input sequence, the first long word candidate item corresponding to the predicted input sequence and the candidate association information of the first long word candidate item.

25. The device according to claim 24, characterized in that The device also includes: The screening module is used to screen out long word candidates that meet preset conditions based on the candidate association information.

26. The device according to claim 24, characterized in that The device also includes: A training module is used to collect multiple groups of historical input information, where a group of historical input information includes a historical input sequence, historical input association information corresponding to the historical input sequence, and corresponding reference long word candidate information; for a group of the historical input information, the historical input sequence in the historical input information and the historical input association information corresponding to the historical input sequence are input into the preset long word association model to obtain first long word candidate information; based on the comparison between the first long word candidate information and the reference long word candidate information in the group of historical input information, the weight of the preset long word association model is adjusted.

27. A readable storage medium, characterized in that: When the instructions in the storage medium are executed by the processor of the terminal device, the terminal device is enabled to execute the method as described in any one of method claims 1-10.

28. A readable storage medium, characterized in that: When the instructions in the storage medium are executed by the processor of the server, the server is enabled to execute the method as described in any one of method claims 11-13.

29. A terminal device, characterized in that: The system comprises a memory and at least one program, wherein the at least one program is stored in the memory and is configured to be executed by at least one processor to include instructions for performing the following operations: Receive an input sequence; Acquire input-related information, where the input-related information includes: the above information and / or input environment information; According to the input association information, a second long word association request is generated and sent to the server, so that the server returns the corresponding first long word candidate information; Determining target long word candidate information corresponding to the input sequence according to the received first long word candidate information, wherein the first long word candidate information is determined by the server according to a preset model and returned in response to the received long word association request; Displaying the target long word candidate information; The preset model includes a preset long word association model and a preset input sequence prediction model, and the server determines and returns the first long word candidate information according to the preset model for the received long word association request, including: If the long word association request is a second long word association request, acquiring input association information from the second long word association request; Inputting the input association information into a preset input sequence prediction model to obtain a predicted input sequence; Inputting the predicted input sequence and the corresponding input association information into a preset long word association model to obtain a first long word candidate corresponding to the predicted input sequence and candidate association information of the first long word candidate; The first long word candidate information is generated and returned according to the predicted input sequence, the first long word candidate corresponding to the predicted input sequence and the candidate association information of the first long word candidate.

30. The terminal device according to claim 29, characterized in that: The first long word candidate information includes a predicted input sequence, a first long word candidate corresponding to the predicted input sequence, and candidate association information corresponding to the first long word candidate; The step of determining target long word candidate information corresponding to the input sequence based on the received first long word candidate information includes: Finding a predicted input sequence that matches the input sequence; The first long word candidate corresponding to the matched predicted input sequence and the candidate associated information corresponding to the first long word candidate are determined as target long word candidate information.

31. The terminal device according to claim 30, characterized in that: Also contains instructions for: Perform local long word association to determine second long word candidate information corresponding to the input sequence.

32. The terminal device according to claim 31, characterized in that: The performing of local long word association to determine second long word candidate information corresponding to the input sequence includes: Perform local long word association based on the candidate association information corresponding to the first long word candidate item to determine the second long word candidate information corresponding to the input sequence.

33. The terminal device according to claim 31 or 32, characterized in that: The target long word candidate information includes target long word candidates and corresponding candidate association information, and the second long word candidate information includes second long word candidates and corresponding candidate association information; The displaying of the target long word candidate information includes: sorting the target long word candidate and the second long word candidate according to the candidate association information corresponding to the target long word candidate and the candidate association information corresponding to the second long word candidate; According to the ranking result, the target long word candidate items and the second long word candidate items are displayed.

34. The terminal device according to claim 29, characterized in that: The target long word candidate information includes target long word candidates and corresponding candidate association information; The displaying of the target long word candidate information includes: sorting the target long word candidate items according to the candidate association information corresponding to the target long word candidate items; According to the sorting result, the target long word candidates are displayed.

35. The terminal device according to claim 29, characterized in that: The displaying of the target long word candidate information includes: The target long word candidate information is retained in the display area for a set period of time.

36. The terminal device according to claim 29, characterized in that: The terminal device further includes: receiving an interactive instruction, and performing a corresponding interactive response operation on the target long word candidate information corresponding to the interactive instruction; The interactive response operation includes at least one of the following: screen display, splitting and reorganization, display update, and candidate expansion.

37. The terminal device according to claim 33, characterized in that: The candidate association information of the target long word candidate includes at least one of the following: the candidate score corresponding to the target long word candidate, the context length corresponding to the target long word candidate, the identifier of the optimal long word candidate that fully matches the input sequence, and the type of input word corresponding to the input sequence.

38. The terminal device according to claim 29, characterized in that: The input environment information includes at least one of the following: location information, time information, weather information, and application-related information.

39. A server, characterized in that: The system comprises a memory and at least one program, wherein the at least one program is stored in the memory and is configured to be executed by at least one processor to include instructions for performing the following operations: Receive long word association requests; Determining first long word candidate information corresponding to the long word association request according to a preset model, wherein the preset model is trained according to historical input information; Return the first long word candidate information; The preset model includes a preset long word association model and a preset input sequence prediction model, and determining the first long word candidate information corresponding to the long word association request according to the preset model includes: If the long word association request is a second long word association request, acquiring input association information from the second long word association request; Inputting the input association information into a preset input sequence prediction model to obtain a predicted input sequence; Inputting the predicted input sequence and the corresponding input association information into a preset long word association model to obtain a first long word candidate corresponding to the predicted input sequence and candidate association information of the first long word candidate; First long word candidate information is generated according to the predicted input sequence, the first long word candidate corresponding to the predicted input sequence, and candidate association information of the first long word candidate.

40. The server according to claim 39, characterized in that Also contains instructions for: Based on the candidate association information, long word candidates that meet preset conditions are screened out.

41. The server according to claim 39, characterized in that It also includes instructions for performing the following operations of training the preset long word association model: Collecting multiple groups of historical input information, where a group of historical input information includes a historical input sequence, historical input association information corresponding to the historical input sequence, and corresponding reference long word candidate information; For a set of the historical input information, inputting the historical input sequence in the historical input information and the historical input association information corresponding to the historical input sequence into the preset long word association model to obtain first long word candidate information; The weight of the preset long word association model is adjusted based on the comparison between the first long word candidate information and the reference long word candidate information in the group of historical input information.