An input association method, device and device for input association

Abstract

Embodiments of the present application provide an input association method and device, and a device for input association. The method comprises: obtaining preceding information; determining an association candidate ranking result of different semantic categories corresponding to the preceding information; when the number of association candidates in the association candidate ranking result is greater than a preset display number, expanding and displaying a target association candidate in the association candidate ranking result, and folding and displaying other association candidates other than the target association candidate. Embodiments of the present application can provide more diversified association candidates to users to meet the input needs of different users in different scenarios and improve the input efficiency of users.

Description

Technical Field

[0001] The present invention relates to the technical field of input methods, and particularly to an input association method, an apparatus, and an apparatus for input association. Background Art

[0002] An input method system can provide association candidates based on the content input by the user, enabling the user to directly input the content by selecting the association candidates, which provides convenience for the user's input. For example, when the user inputs the word "what's wrong", the input method can associate candidates such as "what happened", "what's going on", "is there anything" for the user to select.

[0003] However, limited by the display strategy and display position of the association candidates, currently the input method only shows the association candidates with higher scores to the user, and the displayed association candidates usually have repetitive semantics, which cannot meet the input needs of different users and affects the user's input efficiency. Summary of the Invention

[0004] Embodiments of the present invention provide an input association method, an apparatus, and an apparatus for input association, which can provide more diversified association candidates to the user to meet the input needs of different users in different scenarios and improve the user's input efficiency.

[0005] To solve the above problems, embodiments of the present invention disclose an input association method, and the method includes:

[0006] Obtain the above-context information;

[0007] Determine the sorting results of association candidates corresponding to different semantic categories of the above-context information;

[0008] When the number of association candidates in the sorting results of association candidates is greater than a preset display number, expand and display the target association candidates in the sorting results of association candidates, and fold and display other association candidates except the target association candidates.

[0009] Optionally, the determining the sorting results of association candidates corresponding to different semantic categories of the above-context information includes:

[0010] Obtain the association candidates corresponding to the above-context information;

[0011] Divide the association candidates corresponding to the above-context information according to semantic categories to obtain association candidates corresponding to different semantic categories;

[0012] Sort the association candidates corresponding to different semantic categories respectively to obtain the sorting results of association candidates corresponding to different semantic categories of the above-context information.

[0013] Optionally, the obtaining the association candidates corresponding to the above-context information includes:

[0014] Obtain association candidates whose semantic similarity to the above information exceeds a preset value; and / or

[0015] Obtain association candidates that match the continuation writing intent corresponding to the above information.

[0016] Optionally, expanding and displaying the target association candidates in the association candidate ranking results includes:

[0017] In the ranking results of the association candidates for different semantic categories, the top N target association candidates for each semantic category are determined, where N is determined according to the preset number of displays;

[0018] Expand to display the top N target association candidates in the association candidate ranking results for each semantic category.

[0019] Optionally, the method further includes:

[0020] In response to a first triggering operation on the target association candidate, the target association candidate is style-transformed according to the target language style selected by the first triggering operation to obtain a transformed association candidate corresponding to the target language style; wherein, the target language style includes any one of court style, melancholic style, martial arts style, colloquial style, written style, dialect style, and witty style;

[0021] Display the conversion association candidates.

[0022] Optionally, the method further includes:

[0023] In response to the selection operation of the conversion association candidate, the conversion association candidate is displayed on the screen; or

[0024] In response to a second triggering operation on the conversion association candidate, the conversion association candidate is edited to obtain a custom association candidate;

[0025] In response to the selection operation of the custom association candidate, the custom association candidate is displayed on the screen.

[0026] Optionally, the above information includes any one of the following: content already displayed on the screen, content sent in the interactive scenario, and content received in the interactive scenario.

[0027] On the other hand, embodiments of the present invention disclose an input association device, the device comprising:

[0028] The above text retrieval module is used to retrieve the above text information;

[0029] The result determination module is used to determine the association candidate ranking results for different semantic categories corresponding to the above information;

[0030] The results display module is used to expand and display the target associated candidate in the associated candidate sorting results when the number of associated candidates in the associated candidate sorting results is greater than the preset number of displays, and to collapse and display other associated candidates other than the target associated candidate.

[0031] Optionally, the result determination module includes:

[0032] The candidate acquisition submodule is used to acquire the associated candidates corresponding to the above information;

[0033] The semantic segmentation submodule is used to segment the association candidates corresponding to the above information according to semantic categories to obtain association candidates of different semantic categories;

[0034] The result sorting submodule is used to sort the association candidates of different semantic categories respectively, and obtain the sorting results of the association candidates of different semantic categories corresponding to the above information.

[0035] Optionally, the candidate acquisition submodule is specifically used to acquire associative candidates whose semantic similarity to the preceding information exceeds a preset value; and / or, acquire associative candidates that conform to the continuation intention corresponding to the preceding information.

[0036] Optionally, the result display module includes:

[0037] The target determination submodule is used to determine the top N target association candidates for each semantic category from the association candidate ranking results of the different semantic categories, where N is determined according to the preset number of displays;

[0038] The target display submodule is used to expand and display the top N target association candidates of the association candidate ranking results for each semantic category.

[0039] Optionally, the device further includes:

[0040] The style conversion module is used to respond to a first trigger operation on the target association candidate, and to perform style conversion on the target association candidate according to the target language style selected by the first trigger operation, so as to obtain the converted association candidate corresponding to the target language style; wherein, the target language style includes any one of court style, melancholic style, martial arts style, colloquial style, written style, dialect style, and witty style.

[0041] A conversion display module is used to display the conversion association candidates.

[0042] Optionally, the device further includes:

[0043] The first selection module is used to display the conversion association candidates in response to the selection operation of the conversion association candidates; or

[0044] The editing module is used to edit the conversion association candidate in response to a second triggering operation on the conversion association candidate to obtain a custom association candidate;

[0045] The second selection module is used to display the custom association candidate in response to the selection operation of the custom association candidate.

[0046] Optionally, the above information includes any one of the following: content already displayed on the screen, content sent in the interactive scenario, and content received in the interactive scenario.

[0047] In another aspect, embodiments of the present invention disclose an apparatus for input association, the apparatus including a memory and one or more programs, wherein one or more programs are stored in the memory and configured to be executed by one or more processors, the one or more programs including instructions for performing input association methods as described above.

[0048] In another aspect, embodiments of the present invention disclose a machine-readable medium having instructions stored thereon that, when executed by one or more processors of a device, cause the device to perform one or more of the aforementioned input association methods.

[0049] The embodiments of the present invention have the following advantages:

[0050] In this embodiment of the invention, during the user's input process using an input method, the preceding text information can be obtained, and the ranking results of the associated candidate suggestions corresponding to different semantic categories of the preceding text information can be determined to obtain more diverse associated candidate suggestions, avoiding the situation where the displayed associated candidate suggestions are semantically singular and repetitive. Furthermore, based on providing more associated candidate suggestions of different semantic categories, this embodiment of the invention expands the display of the target associated candidate suggestion in the ranked results, and collapses the display of other associated candidate suggestions besides the target suggested candidate. Thus, even with a limited number of displayed suggestions, more selectable associated candidate suggestions can be provided to the user to meet the input needs of different users in different scenarios, improving the user's input efficiency. Attached Figure Description

[0051] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a flowchart illustrating the steps of an embodiment of the input association method of the present invention;

[0053] Figure 2 This is a structural block diagram of an embodiment of the input association device of the present invention;

[0054] Figure 3 This is a block diagram of an input association device 800 according to the present invention;

[0055] Figure 4 This is a schematic diagram of the server structure in some embodiments of the present invention. Detailed Implementation

[0056] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] Method Implementation Examples

[0058] Reference Figure 1 The diagram illustrates a flowchart of an embodiment of the input association method of the present invention, which specifically includes the following steps:

[0059] Step 101: Obtain the information above;

[0060] Step 102: Determine the candidate association ranking results for different semantic categories corresponding to the above information;

[0061] Step 103: When the number of association candidates in the association candidate sorting result is greater than the preset number of displays, expand to display the target association candidate in the association candidate sorting result, and collapse to display other association candidates other than the target association candidate.

[0062] The input association method provided in this invention can be applied to various terminals, including but not limited to: servers, smartphones, tablets, e-book readers, MP3 (Moving Picture Experts Group Audio Layer III) players, MP4 (Moving Picture Experts Group Audio Layer IV) players, laptops, in-vehicle computers, desktop computers, set-top boxes, smart TVs, wearable devices, etc.

[0063] The terminal supports operating systems including Windows, Android, iOS, and Windows Phone. It allows input via external input devices, such as a keyboard, and can also run applications that use virtual keyboards, such as input method programs. For example, on a computer, users can input strings by pressing physical keys on a keyboard, while on a mobile terminal with a touchscreen, users can input strings by clicking virtual keys on a virtual keyboard, thus enabling text input.

[0064] The input method mentioned in this embodiment of the invention supports multiple input languages. The input method can be an encoding method used to input various symbols into electronic devices such as computers and mobile phones. Users can conveniently input the required characters or strings into the electronic device using the input method application. It should be noted that, in addition to supporting common Chinese input methods (such as Pinyin, Wubi, Zhuyin, voice input, and handwriting input), the input method in this embodiment can also support input methods for other languages ​​(such as Japanese Hiragana and Korean). No limitations are made here regarding the input method or the language of the input method.

[0065] The embodiments of the present invention can obtain the above information in the input interface. The input interface can be the input interface of any application in the terminal. The application can include, but is not limited to, instant messaging applications, office applications, game applications, social platforms, etc.

[0066] The preceding information may include information related to the most recently displayed or sent content. Optionally, the preceding information may include any one of the following: displayed content, sent content in the interactive scenario, or received content in the interactive scenario.

[0067] In one example, the aforementioned information may include instant message content sent by user A to user B in a chat window of a social software; or, the aforementioned information may include instant message content sent by user B received by user A in a chat window of a social software, etc.

[0068] It should be noted that the embodiments of the present invention do not limit the type of content included in the above-mentioned information. For example, the above-mentioned information may include text information, image information, voice information, etc. When the above-mentioned information includes text information, the associative candidate ranking results for different semantic categories corresponding to the text information can be obtained. Further, when the above-mentioned information includes image information, text recognition can be performed on the image information to identify the text information contained in the image information, and then the associative candidate ranking results for different semantic categories corresponding to the text information can be obtained. When the above-mentioned information includes voice information, speech recognition can be performed on the voice information to convert it into text information, and then the associative candidate ranking results for different semantic categories corresponding to the text information can be obtained. For ease of description, the embodiments of the present invention will be described using text information as an example.

[0069] The input method program can obtain the preceding information in the input interface, and then suggest related candidate words based on that preceding information. This allows users to directly select the content they want to display from the suggested candidates, thereby improving input efficiency. These suggested candidates may include, but are not limited to, characters, words, phrases, long sentences, emoticons, and symbols.

[0070] In this embodiment of the invention, after obtaining the preceding information in the input interface, the ranking results of the association candidates corresponding to different semantic categories of the preceding information are determined to avoid the situation where only the association candidates with higher semantic similarity scores are obtained, resulting in semantic duplication of the displayed association candidates.

[0071] In this embodiment of the invention, the associative candidates of different semantic categories are semantically related to the preceding information, but different associative candidates can have different semantic expressions. For example, for the preceding information "What happened?", the associative candidate "What happened?" is semantically related to the preceding information, and the associative candidate "Who bullied you?" is also semantically related to the preceding information. However, the associative candidates "What happened?" and "Who bullied you?" have different semantic expressions, that is, the associative candidates "What happened?" and "Who bullied you?" correspond to different semantic categories.

[0072] In an optional embodiment of the present invention, determining the associative candidate ranking results for different semantic categories corresponding to the preceding information may include:

[0073] Step S11: Obtain the association candidates corresponding to the above information;

[0074] Step S12: Divide the association candidates corresponding to the above information according to semantic categories to obtain association candidates of different semantic categories;

[0075] Step S13: Sort the association candidates of different semantic categories respectively to obtain the sorted results of the association candidates of different semantic categories corresponding to the above-mentioned information.

[0076] In an example, assume that the obtained context information includes the text information "What's wrong". The input method first obtains all the association candidates corresponding to this context information, such as: "What happened", "What's going on", "What's the matter", "Who bullied you", "Who bullied you", etc.

[0077] Then, divide all the association candidates corresponding to the above-mentioned information according to semantic categories to obtain association candidates of different semantic categories. For example, in this example, the above-mentioned association candidates can be divided into two different semantic categories. Among them, the association candidates "What happened", "What's going on", and "What's the matter" belong to one semantic category; the association candidates "Who bullied you" and "Who bullied you" belong to another semantic category. The association candidates of these two semantic categories are semantically relevant to the above-mentioned information "What's wrong", but belong to different semantic categories.

[0078] Finally, sort the obtained association candidates of different semantic categories respectively to obtain the sorted results of the association candidates of different semantic categories corresponding to the above-mentioned information.

[0079] Through the embodiments of the present invention, after obtaining the input above-mentioned information, association candidates of different semantic categories that are semantically relevant to the above-mentioned information can be obtained and displayed, and then more diversified association candidates can be provided to the user to meet the input needs of different users in different scenarios.

[0080] It should be noted that the embodiments of the present invention do not limit the specific manner of determining the sorted results of the association candidates of different semantic categories corresponding to the above-mentioned information.

[0081] Optionally, determining the sorted results of the association candidates of different semantic categories corresponding to the above-mentioned information may include: determining the sorted results of the association candidates of different semantic categories corresponding to the above-mentioned information according to the weight information corresponding to the association candidates.

[0082] Furthermore, the weight information may be determined according to the first reference information and / or the second reference information. The first reference information is related to user preferences, and the second reference information is related to the input environment.

[0083] The first reference information may include personalized input information from the user, reflecting the user's input preferences. For example, the first reference information may include the user's preference information for different semantic categories, the user's historical screen information for historical association candidates, and the number of times the user has historically entered historical association candidates. By obtaining first reference information related to user preferences, the weight information of association candidates that match the user's preferences can be improved, thereby making the ranking results more in line with the user's requirements.

[0084] The second reference information may include input environment information. This input environment information may include at least one of the following: the application running the input method, and the model, configuration, and operating system information of the terminal on which the input method is located. By obtaining the second reference information related to the input environment, the user's current usage scenario, device status, etc., can be determined, and this information can be used as reference information for ranking Lenovo candidates.

[0085] Optionally, a first model can be pre-trained, where the input of the first model is the preceding information and the output is the associative candidate ranking results of different semantic categories corresponding to the preceding information.

[0086] The first model can be obtained by supervised training of an existing neural network using a large number of training samples and machine learning methods. It should be noted that this embodiment of the invention does not limit the model structure or training method of the first model. The first model can integrate multiple neural networks. These neural networks include, but are not limited to, at least one or more combinations, superpositions, or nestings of the following: CNN (Convolutional Neural Network), LSTM (Long Short-Term Memory) network, RNN (Simple Recurrent Neural Network), attention neural network, etc.

[0087] Furthermore, the first model can be located on a cloud server. After the input method program on the terminal obtains the context information from the input interface, it can send the context information to the first model on the cloud server. The first model on the cloud server then obtains the association candidate ranking results for different semantic categories corresponding to the context information, and the cloud server returns them to the input method program on the terminal.

[0088] In practical applications, due to the limited size of the input method's display interface, the number of suggested suggestions that can be displayed is also limited. To allow users to access more suggested suggestions, this embodiment of the invention, when the number of suggested suggestions in the sorted results exceeds a preset display limit, expands and displays the target suggested suggestion in the sorted results on the input interface; that is, expands and displays the target suggested suggestion in the input method's suggested suggestion bar, while collapsing and displaying other suggested suggestions besides the target one.

[0089] "Expand Display" means showing the complete list of suggested suggestions. "Collapse Display" means hiding the suggested suggestions and showing a collapse button. Triggering this collapse button reveals the hidden suggested suggestions, for example, by expanding the list as a dropdown menu. Triggering the collapse button again hides the dropdown menu and its suggested suggestions.

[0090] For example, in the example above, the above information includes the text "What happened?". Assuming the input method's default display count is 2, and the target suggested candidates are determined to include "What happened?" and "Who bullied you?", then the target suggested candidates "What happened?" and "Who bullied you?" can be expanded and displayed in the suggested candidate bar, while other suggested candidates are collapsed.

[0091] Optionally, the method may further include: switching the display of association candidates of different semantic categories in response to a switching instruction.

[0092] In this embodiment of the invention, after expanding and displaying the target associative candidate in the associative candidate ranking result, a switching instruction can be received and responded to to switch the display of associative candidates of different semantic categories. For example, the currently displayed target associative candidate can be switched to another associative candidate with a different semantic category from the target associative candidate.

[0093] In one example, the preceding information includes the text "What happened?". Assuming the input method's default display count is 1, the target suggested suggestion "What happened?" is expanded and displayed, while other suggested suggestions are hidden. Upon receiving a switching command, the expanded target suggested suggestion "What happened?" can be switched to suggested suggestions with a different semantic category than "What happened?", such as "Who bullied you?".

[0094] It is understood that, for ease of description, the preset display count is 1. In specific implementations, the preset display count can be set according to actual needs. Furthermore, this embodiment of the invention does not limit the specific method of switching displays. Optionally, the top-ranked associative candidate in each semantic category can be switched for display, as shown in the example above.

[0095] Optionally, the overall display of the association candidates for each semantic category can also be switched. For example, if the preceding information includes the text "What happened?", the association candidates for the first semantic category obtained based on this preceding information include: "What happened?", "What happened?", and "What's going on?", and the association candidates for the second semantic category include: "Who bullied you?" and "Who bullied you?". Assume that the association candidates for the first semantic category are expanded as the target association candidates (only the top N can be expanded), while the association candidates for the second semantic category are collapsed. Upon receiving a switching instruction, the overall display of the association candidates for these two semantic categories can be switched. That is, the association candidates for the second semantic category are expanded (only the top N can be expanded), and the association candidates for the first semantic category are hidden. Upon receiving another switching instruction, the association candidates for the first semantic category are expanded, and the association candidates for the second semantic category are hidden.

[0096] It should be noted that the embodiments of the present invention do not limit the triggering method of the switching command. For example, buttons, operation gestures, voice commands, etc., can be set to trigger the switching command.

[0097] In an optional embodiment of the present invention, expanding and displaying the target associative candidate in the associative candidate sorting result may include:

[0098] Step S21: In the association candidate ranking results of different semantic categories, determine the top N target association candidates for each semantic category, where N is determined according to the preset number of displays;

[0099] Step S22: Expand and display the top N target association candidates of the association candidate ranking results for each semantic category.

[0100] In this embodiment of the invention, after determining the ranking results of the associative candidates for different semantic categories corresponding to the preceding information, the top N target associative candidates for each semantic category are determined from the ranking results of the associative candidates for different semantic categories. This ensures that associative candidates are available to be displayed in the ranking results of associative candidates for different semantic categories, providing users with diverse associative candidates.

[0101] For example, given the information string1 above, suppose we obtain the associative candidate ranking results for three different semantic categories, denoted as associative candidate ranking result 1 for semantic category 1, associative candidate ranking result 2 for semantic category 2, and associative candidate ranking result 3 for semantic category 3. Assume associative candidate ranking result 1 has n1 associative candidates, associative candidate ranking result 2 has n2 associative candidates, and associative candidate ranking result 3 has n3 associative candidates. Assume that the highest-scoring associative candidate in associative candidate ranking result 3 for semantic category 3 has a relatively low score among the three ranking results. Therefore, if all semantic category associative candidates are uniformly ranked according to semantic similarity, the associative candidate in associative candidate ranking result 3 may not have a chance to be displayed; that is, the top-ranked associative candidates usually have repetitive semantic expressions.

[0102] To avoid the aforementioned situation, embodiments of the present invention sort the association candidates for different semantic categories separately, and select the top N target association candidates from the sorting results of the association candidates for each semantic category. For example, the top N association candidates are selected as target association candidates from candidate sorting results 1, 2, and 3, respectively. In one example, assuming N=1, the association candidate with the highest score is selected as the target association candidate from candidate sorting results 1, 2, and 3, respectively. Therefore, it can be ensured that the final displayed target association candidates can include association candidates from different semantic categories, avoiding the display of semantically singular and repetitive association candidates, and improving the diversity of displayed association candidates.

[0103] Furthermore, after obtaining the sorted results of the associative candidates for different semantic categories corresponding to the preceding information from the cloud server, the client input method can receive the sorted results of the associative candidates for different semantic categories corresponding to the preceding information returned by the cloud server. The input method can then select the top N as target associative candidates from the sorted results of each semantic category for expanded display, and collapse the other associative candidates for display.

[0104] It is understood that the above method of displaying associative candidates is only an application example. In specific implementations, the embodiments of the present invention do not limit the display method or the number of associative candidates (the value of N) for each semantic category.

[0105] In an optional embodiment of the present invention, obtaining the association candidates corresponding to the above information may include:

[0106] Obtain association candidates whose semantic similarity to the above information exceeds a preset value; and / or

[0107] Obtain association candidates that match the continuation writing intent corresponding to the above information.

[0108] When a user inputs data using an input method, the input method can obtain the preceding text information in the input interface and determine suggested candidates based on that information. Optionally, the input method can determine suggested candidates based on the semantic similarity to the preceding text information, such as obtaining suggested candidates whose semantic similarity to the preceding text information exceeds a preset value, thereby obtaining suggested candidates with high semantic similarity to the preceding text information.

[0109] For example, given the preceding text "What happened?", the system can retrieve related candidate phrases with semantic similarity exceeding a preset value, such as "What happened?", "What happened?", "What's going on?", "Who bullied you?", etc. The preset value can be set according to actual needs.

[0110] Optionally, the input method can also determine the suggested candidates based on the continuation intention corresponding to the above information, so as to obtain suggested candidates that match the continuation intention corresponding to the above information.

[0111] Among them, the intention to continue writing refers to the tendency of the preceding information to continue to extend.

[0112] For example, in a chat window of a social media app, User A sends the text message "Let's play a poetry relay game" along with the poem line "White egrets fly before Xisai Mountain." User B's client receives these messages. After User B activates their input method in the chat window, the input method can retrieve the preceding text, including "Let's play a poetry relay game" and "White egrets fly before Xisai Mountain." By semantically analyzing this information, it can determine that the message implies a continuation of the poem and retrieve related suggestions, such as "Flying snow welcomes spring." User B can then directly select the suggestion "Flying snow welcomes spring" and input it, improving their input efficiency.

[0113] This invention does not limit the specific method for identifying the continuation intention corresponding to the preceding information. Optionally, taking text information as an example, this invention can perform semantic analysis on the text information to determine the continuation intention corresponding to the preceding information. Further, in specific implementations, a neural network model for analyzing continuation intentions can be trained based on the user's historical behavior data. After the model training is completed, the preceding information is input into the neural network model to obtain the continuation intention corresponding to the preceding information.

[0114] In an optional embodiment of the present invention, the method may further include:

[0115] Step S31: In response to a first triggering operation on the target association candidate, perform style conversion on the target association candidate according to the target language style selected by the first triggering operation, to obtain a converted association candidate corresponding to the target language style of the target association candidate; wherein, the target language style includes any one of a court style, a melancholy style, a martial arts style, a spoken style, a written style, a dialect style, and a wisecrack style.

[0116] Step S32: Display the converted association candidate.

[0117] In order to provide users with richer and more diversified association candidates, embodiments of the present invention can also preset multiple target language styles to meet the input requirements of different users in different scenarios. The target language style includes any one of a court style, a melancholy style, a martial arts style, a spoken style, a written style, a dialect style, and a wisecrack style.

[0118] The first triggering operation is used to trigger style conversion of the target association candidate according to the selected target language style. It can be understood that embodiments of the present invention do not limit the operation type of the first triggering operation. The first triggering operation can be a preset operation or operation sequence. For example, the first triggering operation can include click (including single click, double click, triple click) operations, long press operations, swipe operations, drag operations, an operation sequence of first selecting, then right clicking, and then clicking, etc.

[0119] In one example, assume that the input method receives a first triggering operation on the target association candidate "What's going on". The target language style selected by the first triggering operation is the court style. In response to this first triggering operation, convert the target association candidate "What's going on" to the court style. The converted association candidates corresponding to the court style of this target association candidate can include: "What变故 has occurred in the palace", "What has happened to the empress", etc.

[0120] In another example, assume that the input method receives a first triggering operation on the target association candidate "Great". The target language style selected by the first triggering operation is the Shaanxi dialect. In response to this first triggering operation, convert the target association candidate "Great" to the Shaanxi dialect, and the converted association candidate corresponding to the Shaanxi dialect of this target association candidate can be "It's wonderful".

[0121] In embodiments of the present invention, style conversion can be performed on the target association candidate displayed in an expanded manner, or style conversion can be performed on the folded display association candidate after triggering its expanded display.

[0122] This invention not only provides more semantically diverse associative candidates to meet users' basic input needs, but also offers a language style conversion function, providing users with associative candidates in different language styles to meet their diverse input needs, further improving input efficiency and enjoyment.

[0123] It should be noted that the specific method of style transfer for the target association candidate is not limited in the embodiments of the present invention. Optionally, a second model can be pre-trained, wherein the input of the second model is the target association candidate and the target language style, and the output is the transformed association candidate corresponding to the target language style.

[0124] The second model can be obtained by supervised training of an existing neural network using a large number of training samples and machine learning methods. It should be noted that this embodiment of the invention does not limit the model structure or training method of the second model. The second model can integrate multiple neural networks. These neural networks include, but are not limited to, at least one or a combination, superposition, or nesting of at least two of the following: CNN, LSTM network, RNN, attention neural network, etc.

[0125] Furthermore, the second model can be located on a cloud server. After the input method program on the terminal receives the first trigger operation on the target association candidate, it can send the target association candidate selected by the first trigger operation and the target language style to the second model on the cloud server. The second model obtains the conversion association candidate corresponding to the target language style, and the cloud server returns it to the input method program on the terminal.

[0126] Optionally, after displaying the conversion association candidates, the conversion association candidates can be displayed on the screen in response to the selection operation of the conversion association candidates.

[0127] Furthermore, after displaying the conversion association candidates, the method may further include:

[0128] Step S41: In response to the second triggering operation on the conversion association candidate, edit the conversion association candidate to obtain a custom association candidate;

[0129] Step S42: In response to the selection operation of the custom association candidate, display the custom association candidate on the screen.

[0130] Among them, the second trigger operation can be used to trigger the selected conversion association candidate to enter the text editing state, and in the text editing state, it can receive the user's editing operation on the text content. Through the embodiment of the present invention, by providing the editing function of the association candidate to the user, the user can edit based on the provided association candidate to obtain a custom association candidate that more meets the user's input requirements, improving the convenience and flexibility of user input.

[0131] After the embodiment of the present invention performs style conversion on the target association candidate, if the user is not satisfied with the obtained conversion association candidate, the conversion association candidate can also be edited through the second trigger operation to obtain a custom association candidate.

[0132] In one example, it is assumed that the input method receives the first trigger operation on the target association candidate "I'm tired", and the target language style selected by the first trigger operation is the court style. In response to this first trigger operation, the target association candidate "I'm tired" is converted into the court style, and the conversion association candidate corresponding to the court style of the target association candidate is "Your servant is exhausted", and this conversion association candidate is displayed in the input interface, such as being displayed in the association candidate column of the input method. If the user is not satisfied with this conversion association candidate, the conversion association candidate can be edited through the second trigger operation. It is assumed that the input method receives the second trigger operation of the user on this conversion association candidate "Your servant is exhausted", then the conversion association candidate is set to enter the text editing state so that the user can make modifications. It is assumed that the user modifies the conversion association candidate "Your servant is exhausted" into a custom association candidate "Your servant is exhausted and going to rest". In response to the selection operation on this custom association candidate, the custom association candidate "Your servant is exhausted and going to rest" can be input onto the screen.

[0133] Furthermore, after editing the conversion association candidate to obtain a custom association candidate, the input method can save and record the custom association candidate, and the second model can be further optimized by using the recorded custom association candidate.

[0134] It should be noted that in addition to editing the conversion association candidate to obtain a custom association candidate, the embodiment of the present invention can also edit any association candidate in the input interface to obtain a custom association candidate. For example, the input method can receive the second trigger operation on the target association candidate displayed in the expanded form in the association candidate column, and then edit the target association candidate selected by the second trigger operation to obtain a custom association candidate. Or, after triggering the expansion display of the folded association candidate, it can also be edited through the second trigger operation to obtain a custom association candidate.

[0135] In summary, during the user's input process using an input method, this embodiment of the invention can obtain the preceding text information and determine the sorting results of the associated candidate suggestions corresponding to different semantic categories of the preceding text information, thereby obtaining more diverse associated candidate suggestions and avoiding the situation where the displayed associated candidate suggestions are semantically singular and repetitive. Furthermore, based on providing more associated candidate suggestions of different semantic categories, this embodiment of the invention expands the display of the target associated candidate suggestion in the sorting results and collapses the display of other associated candidate suggestions besides the target suggestion. Thus, even with a limited number of displayed suggestions, it can provide users with more selectable associated candidate suggestions to meet the input needs of different users in different scenarios and improve user input efficiency.

[0136] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0137] Device Examples

[0138] Reference Figure 2 The diagram illustrates a structural block diagram of an embodiment of the input association device of the present invention, the device including:

[0139] The above text retrieval module 201 is used to retrieve the above text information;

[0140] The result determination module 202 is used to determine the association candidate ranking results for different semantic categories corresponding to the above information;

[0141] The result display module 203 is used to expand and display the target associated candidate in the associated candidate sorting result when the number of associated candidates in the associated candidate sorting result is greater than the preset number of displayable candidates, and to collapse and display other associated candidates other than the target associated candidate.

[0142] Optionally, the result determination module includes:

[0143] The candidate acquisition submodule is used to acquire the associated candidates corresponding to the above information;

[0144] The semantic segmentation submodule is used to segment the association candidates corresponding to the above information according to semantic categories to obtain association candidates of different semantic categories;

[0145] The result sorting submodule is used to sort the association candidates of different semantic categories respectively, and obtain the sorting results of the association candidates of different semantic categories corresponding to the above information.

[0146] Optionally, the candidate acquisition submodule is specifically used to acquire associative candidates whose semantic similarity to the preceding information exceeds a preset value; and / or, acquire associative candidates that conform to the continuation intention corresponding to the preceding information.

[0147] Optionally, the result display module includes:

[0148] The target determination submodule is used to determine the top N target association candidates for each semantic category from the association candidate ranking results of the different semantic categories, where N is determined according to the preset number of displays;

[0149] The target display submodule is used to expand and display the top N target association candidates of the association candidate ranking results for each semantic category.

[0150] Optionally, the device further includes:

[0151] The style conversion module is used to respond to a first trigger operation on the target association candidate, and to perform style conversion on the target association candidate according to the target language style selected by the first trigger operation, so as to obtain the converted association candidate corresponding to the target language style; wherein, the target language style includes any one of court style, melancholic style, martial arts style, colloquial style, written style, dialect style, and witty style.

[0152] A conversion display module is used to display the conversion association candidates.

[0153] Optionally, the device further includes:

[0154] The first selection module is used to display the conversion association candidates in response to the selection operation of the conversion association candidates; or

[0155] The editing module is used to edit the conversion association candidate in response to a second triggering operation on the conversion association candidate to obtain a custom association candidate;

[0156] The second selection module is used to display the custom association candidate in response to the selection operation of the custom association candidate.

[0157] Optionally, the above information includes any one of the following: content already displayed on the screen, content sent in the interactive scenario, and content received in the interactive scenario.

[0158] Optionally, the result determination module is specifically used to determine the ranking results of the association candidates for different semantic categories corresponding to the above information based on the weight information corresponding to the association candidates.

[0159] Furthermore, the weight information can be determined based on the first reference information and / or the second reference information. The first reference information is related to user preferences, and the second reference information is related to the input environment.

[0160] The first reference information may include personalized input information of the user, used to reflect the user's input preferences. The second reference information may include input environment information. The input environment information may include at least one of the following: the application in which the input method is located, and the model information, configuration information, and operating system information of the terminal in which the input method is located.

[0161] Optionally, the device may further include: a switching display module, used to switch the display of association candidates of different semantic categories in response to a switching command.

[0162] In this embodiment of the invention, after expanding and displaying the target associative candidate in the associative candidate ranking result, a switching instruction can be received and responded to to switch the display of associative candidates of different semantic categories. For example, the currently displayed target associative candidate can be switched to another associative candidate with a different semantic category from the target associative candidate.

[0163] It should be noted that the embodiments of the present invention do not limit the triggering method of the switching command. For example, buttons, operation gestures, voice commands, etc., can be set to trigger the switching command.

[0164] In this embodiment of the invention, during the user's input process using an input method, the preceding text information can be obtained, and the ranking results of the associated candidate suggestions corresponding to different semantic categories of the preceding text information can be determined to obtain more diverse associated candidate suggestions, avoiding the situation where the displayed associated candidate suggestions are semantically singular and repetitive. Furthermore, based on providing more associated candidate suggestions of different semantic categories, this embodiment of the invention expands the display of the target associated candidate suggestion in the ranked results, and collapses the display of other associated candidate suggestions besides the target suggested candidate. Thus, even with a limited number of displayed suggestions, more selectable associated candidate suggestions can be provided to the user to meet the input needs of different users in different scenarios, improving the user's input efficiency.

[0165] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0166] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. Regarding the apparatus in the above embodiments, the specific methods by which each module performs its operations have been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0167] This invention provides an apparatus for input association, the apparatus including a memory and one or more programs, wherein one or more programs are stored in the memory and configured to be executed by one or more processors. The one or more programs include instructions for performing the following operations: obtaining context information; determining the association candidate ranking results for different semantic categories corresponding to the context information; when the number of association candidates in the association candidate ranking results is greater than a preset number of displays, expanding and displaying the target association candidate in the association candidate ranking results, and collapsing and displaying other association candidates other than the target association candidate.

[0168] Optionally, determining the candidate associative ranking results for different semantic categories corresponding to the preceding information includes:

[0169] Obtain the association candidates corresponding to the above information;

[0170] The association candidates corresponding to the above information are divided according to semantic categories to obtain association candidates of different semantic categories;

[0171] The association candidates of different semantic categories are sorted to obtain the sorting results of the association candidates of different semantic categories corresponding to the above information.

[0172] Optionally, obtaining the association candidates corresponding to the above information includes:

[0173] Obtain association candidates whose semantic similarity to the above information exceeds a preset value; and / or

[0174] Obtain association candidates that match the continuation writing intent corresponding to the above information.

[0175] Optionally, expanding and displaying the target association candidates in the association candidate ranking results includes:

[0176] In the ranking results of the association candidates for different semantic categories, the top N target association candidates for each semantic category are determined, where N is determined according to the preset number of displays;

[0177] Expand to display the top N target association candidates in the association candidate ranking results for each semantic category.

[0178] Optionally, the device is further configured to be executed by one or more processors, comprising instructions for performing the following operations:

[0179] In response to a first triggering operation on the target association candidate, the target association candidate is style-transformed according to the target language style selected by the first triggering operation to obtain a transformed association candidate corresponding to the target language style; wherein, the target language style includes any one of court style, melancholic style, martial arts style, colloquial style, written style, dialect style, and witty style;

[0180] Display the conversion association candidates.

[0181] Optionally, the device is further configured to be executed by one or more processors, comprising instructions for performing the following operations:

[0182] In response to the selection operation of the conversion association candidate, the conversion association candidate is displayed on the screen; or

[0183] In response to a second triggering operation on the conversion association candidate, the conversion association candidate is edited to obtain a custom association candidate;

[0184] In response to the selection operation of the custom association candidate, the custom association candidate is displayed on the screen.

[0185] Optionally, the above information includes any one of the following: content already displayed on the screen, content sent in the interactive scenario, and content received in the interactive scenario.

[0186] Figure 3 This is a block diagram illustrating a device 800 for input association according to an exemplary embodiment. For example, device 800 may be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, etc.

[0187] Reference Figure 3 The device 800 may include one or more of the following components: a processing component 802, a memory 804, a power supply component 806, a multimedia component 808, an audio component 810, an input / output (I / O) interface 812, a sensor component 814, and a communication component 816.

[0188] Processing component 802 typically controls the overall operation of device 800, such as operations associated with display, telephone calls, data communication, camera operation, and recording operations. Processing component 802 may include one or more processors 820 to execute instructions to perform all or part of the steps of the methods described above. Furthermore, processing component 802 may include one or more modules to facilitate interaction between processing component 802 and other components. For example, processing component 802 may include a multimedia module to facilitate interaction between multimedia component 808 and processing component 802.

[0189] Memory 804 is configured to store various types of data to support the operation of device 800. Examples of this data include instructions for any application or method operating on device 800, contact data, phonebook data, messages, pictures, videos, etc. Memory 804 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 storage, flash memory, magnetic disk, or optical disk.

[0190] Power supply component 806 provides power to various components of device 800. Power supply component 806 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power to device 800.

[0191] Multimedia component 808 includes a screen that provides an output interface between the device 800 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 touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 808 includes a front-facing camera and / or a rear-facing camera. When the device 800 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0192] Audio component 810 is configured to output and / or input audio signals. For example, audio component 810 includes a microphone (MIC) configured to receive external audio signals when device 800 is in an operating mode, such as call mode, recording mode, and voice information processing mode. The received audio signals may be further stored in memory 804 or transmitted via communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.

[0193] I / O interface 812 provides an interface between processing component 802 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0194] Sensor assembly 814 includes one or more sensors for providing status assessments of various aspects of device 800. For example, sensor assembly 814 can detect the on / off state of device 800, the relative positioning of components such as the display and keypad of device 800, changes in the position of device 800 or a component of device 800, the presence or absence of user contact with device 800, the orientation or acceleration / deceleration of device 800, and temperature changes of device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 814 may also include an accelerometer, a gyroscope, a magnetometer, a pressure sensor, or a temperature sensor.

[0195] Communication component 816 is configured to facilitate wired or wireless communication between device 800 and other devices. Device 800 can access wireless networks based on communication standards, such as WiFi, 2G, or 3G, or combinations thereof. In one exemplary embodiment, communication component 816 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 816 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency information processing (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0196] In an exemplary embodiment, the apparatus 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described above.

[0197] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 804 including instructions, which can be executed by a processor 820 of the device 800 to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0198] Figure 4 This is a schematic diagram of the server structure in some embodiments of the present invention. The server 1900 can vary considerably depending on its configuration or performance, and may include one or more central processing units (CPUs) 1922 (e.g., one or more processors) and memory 1932, and one or more storage media 1930 (e.g., one or more mass storage devices) for storing application programs 1942 or data 1944. The memory 1932 and storage media 1930 can be temporary or persistent storage. The program stored in the storage media 1930 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the server. Furthermore, the CPU 1922 may be configured to communicate with the storage media 1930 and execute the series of instruction operations in the storage media 1930 on the server 1900.

[0199] Server 1900 may also include one or more power supplies 1926, one or more wired or wireless network interfaces 1950, one or more input / output interfaces 1958, one or more keyboards 1956, and / or one or more operating systems 1941, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.

[0200] A non-transitory computer-readable storage medium that, when instructions in the storage medium are executed by a processor of a device (server or terminal), enables the device to perform... Figure 1 The input association method shown.

[0201] A non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by a processor of a device (server or terminal), enables the device to perform an input association method, the method comprising: acquiring preceding information; determining the association candidate ranking results for different semantic categories corresponding to the preceding information; when the number of association candidates in the association candidate ranking results is greater than a preset number of displays, expanding and displaying a target association candidate in the association candidate ranking results, and collapsing and displaying other association candidates other than the target association candidate.

[0202] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0203] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is limited only by the appended claims. The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0204] The above provides a detailed description of an input association method, an input association device, and an input association apparatus provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. An input association method, characterized in that, The method includes: Get the information above; Determining the ranking results of the association candidates for different semantic categories corresponding to the preceding information includes: obtaining the association candidates corresponding to the preceding information; dividing the association candidates corresponding to the preceding information according to semantic categories to obtain association candidates for different semantic categories; and ranking the association candidates for different semantic categories to obtain the ranking results of the association candidates for different semantic categories corresponding to the preceding information. When the number of association candidates in the association candidate sorting result is greater than the preset number of displays, the target association candidate in the association candidate sorting result is expanded to display, and other association candidates other than the target association candidate are collapsed to display. The target association candidate is an association candidate of a different semantic category that matches the preset number of displays.

2. The method according to claim 1, characterized in that, The association candidates include at least one of the following: characters, words, phrases, long sentences, emoticons, and symbols.

3. The method according to claim 2, characterized in that, The step of obtaining the association candidates corresponding to the above information includes: Obtain association candidates whose semantic similarity to the above information exceeds a preset value; and / or Obtain association candidates that match the continuation writing intent corresponding to the above information.

4. The method according to claim 1, characterized in that, The expanded display of the target association candidates in the association candidate sorting results includes: In the ranking results of the association candidates for different semantic categories, the top N target association candidates for each semantic category are determined, where N is determined according to the preset number of displays; Expand to display the top N target association candidates in the association candidate ranking results for each semantic category.

5. The method according to claim 1, characterized in that, The method further includes: In response to a first triggering operation on the target association candidate, the target association candidate is style-transformed according to the target language style selected by the first triggering operation to obtain a transformed association candidate corresponding to the target language style; wherein, the target language style includes any one of court style, melancholic style, martial arts style, colloquial style, written style, dialect style, and witty style; Display the conversion association candidates.

6. The method according to claim 5, characterized in that, The method further includes: In response to the selection operation of the conversion association candidate, the conversion association candidate is displayed on the screen; or In response to a second triggering operation on the conversion association candidate, the conversion association candidate is edited to obtain a custom association candidate; In response to the selection operation of the custom association candidate, the custom association candidate is displayed on the screen.

7. The method according to claim 1, characterized in that, The above information includes any one of the following: content already displayed on the screen, content sent in an interactive scenario, or content received in an interactive scenario.

8. An input association device, characterized in that, The device includes: The above text retrieval module is used to retrieve the above text information; The result determination module is used to determine the ranking results of the association candidates for different semantic categories corresponding to the above information, including: obtaining the association candidates corresponding to the above information; dividing the association candidates corresponding to the above information according to semantic categories to obtain association candidates for different semantic categories; and ranking the association candidates for different semantic categories to obtain the ranking results of the association candidates for different semantic categories corresponding to the above information. The results display module is used to expand and display the target associative candidate in the associative candidate ranking result when the number of associative candidates in the associative candidate ranking result is greater than the preset display number, and to collapse and display other associative candidates other than the target associative candidate. The target associative candidate is an associative candidate of a different semantic category that matches the preset display number.

9. The apparatus according to claim 8, characterized in that, The association candidates include at least one of the following: characters, words, phrases, long sentences, emoticons, and symbols.

10. The apparatus according to claim 9, characterized in that, The result determination module is specifically used to obtain association candidates whose semantic similarity to the above information exceeds a preset value; and / or to obtain association candidates that conform to the continuation intention corresponding to the above information.

11. The apparatus according to claim 8, characterized in that, The results display module includes: The target determination submodule is used to determine the top N target association candidates for each semantic category from the association candidate ranking results of the different semantic categories, where N is determined according to the preset number of displays; The target display submodule is used to expand and display the top N target association candidates of the association candidate ranking results for each semantic category.

12. The apparatus according to claim 8, characterized in that, The device further includes: The style conversion module is used to respond to a first trigger operation on the target association candidate, and to perform style conversion on the target association candidate according to the target language style selected by the first trigger operation, so as to obtain the converted association candidate corresponding to the target language style; wherein, the target language style includes any one of court style, melancholic style, martial arts style, colloquial style, written style, dialect style, and witty style. A conversion display module is used to display the conversion association candidates.

13. The apparatus according to claim 12, characterized in that, The device further includes: The first selection module is used to display the conversion association candidates in response to the selection operation of the conversion association candidates; or The editing module is used to edit the conversion association candidate in response to a second triggering operation on the conversion association candidate to obtain a custom association candidate; The second selection module is used to display the custom association candidate in response to the selection operation of the custom association candidate.

14. A device for inputting associations, characterized in that, The device includes a memory and one or more programs, wherein one or more programs are stored in the memory and configured to be executed by one or more processors, the one or more programs including instructions for performing the input association method as described in any one of claims 1 to 7.

15. A machine-readable medium having instructions stored thereon that, when executed by one or more processors of a device, cause the device to perform the input association method as described in any one of claims 1 to 7.

Images