Input method, device and device for input

Abstract

The embodiments of the present invention provide an input method, an apparatus and an apparatus for input. The method specifically comprises: determining an input intention according to a context; determining a keyword matching the input intention from the context; and determining an association candidate corresponding to the context according to the input intention and the keyword. The embodiments of the present invention can improve the accuracy of the association candidate.

Description

Technical Field

[0001] The present invention relates to the field of input technology, and in particular to an input method, an input device and a device for input. Background Art

[0002] As a bridge for computer systems to communicate with users or other devices, devices are one of the main means for information exchange between users and computer systems. They can facilitate users to input information in a variety of scenarios. For example, users can enter keywords in search engines to search web pages, enter text in instant messaging apps (applications) to communicate with other users, and enter text in document apps to edit documents, etc.

[0003] The association function of the input method is an extended function of the input method program. Its appearance reduces the number of times the user actively inputs and the number of keystrokes, and increases the intelligence of the input method. One implementation process of the association function includes: obtaining the user's most recent screen content, and querying multiple relationships based on the most recent screen content to obtain association candidates. For example, if the most recent screen content is "Sogou", the corresponding association candidates may include: "input method", "search", "browser", etc.

[0004] The inventors found in the process of implementing the embodiments of the present invention that when the user expresses through a longer content, the most recent content on the screen may not accurately reflect the input intention, and the association candidates obtained in this case will not accurately reflect the input intention. For example, the user inputs "a movie was recently released, its name is", in this case, based on the most recent content on the screen "the name is", the association candidates such as "what" and "what" will be obtained. Obviously, the above association candidates cannot accurately reflect the input intention. Summary of the invention

[0005] The embodiments of the present invention provide an input method, an apparatus and an input device, which can improve the accuracy of association candidates.

[0006] In order to solve the above problem, an embodiment of the present invention discloses an input method, including:

[0007] Determine input intent based on context;

[0008] Determining keywords matching the input intent from the context;

[0009] According to the input intention and the keyword, an association candidate corresponding to the context is determined.

[0010] On the other hand, an embodiment of the present invention discloses an input device, comprising:

[0011] The intention determination module is used to determine the input intention based on the context;

[0012] A keyword determination module, configured to determine keywords matching the input intent from the context; and

[0013] The association candidate determination module is used to determine the association candidates corresponding to the context according to the input intention and the keyword.

[0014] In another aspect, an embodiment of the present invention discloses a device for input, 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, and the one or more programs include instructions for performing the following operations:

[0015] Determine input intent based on context;

[0016] Determining keywords matching the input intent from the context;

[0017] According to the input intention and the keyword, an association candidate corresponding to the context is determined.

[0018] On the other hand, an embodiment of the present invention discloses a machine-readable medium having instructions stored thereon, which, when executed by one or more processors, causes a device to perform one or more of the input methods described above.

[0019] The embodiments of the present invention include the following advantages:

[0020] The embodiment of the present invention determines input intent based on context. The input intent may represent information about the content to be input. For example, the information about the content to be input may include information such as category and entity type.

[0021] Furthermore, the embodiment of the present invention determines keywords matching the above input intent from the context. The keywords can enrich and supplement the input intent, and are used together with the input intent to determine association candidates, thereby improving the accuracy of the association candidates. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0023] Figure 1 is a schematic diagram of an application environment of an input method according to an embodiment of the present invention;

[0024] Figure 2 is a flow chart of steps of an input method embodiment 1 of the present invention;

[0025] Figure 3 is a flow chart of steps of Embodiment 2 of an input method of the present invention;

[0026] Figure 4 is a structural block diagram of an input device embodiment of the present invention;

[0027] Figure 5 is a block diagram of a device 800 for inputting according to the present invention; and

[0028] Figure 6 It is a schematic diagram of the structure of the server in some embodiments of the present invention. DETAILED DESCRIPTION

[0029] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] An embodiment of the present invention provides an input scheme, which can determine the input intention based on the context; determine the keywords matching the input intention from the context; and determine the association candidates corresponding to the context based on the input intention and the keywords.

[0031] In the embodiment of the present invention, the context may include: the previous context corresponding to the input cursor, and / or the following context corresponding to the input cursor. Optionally, the previous context is usually the part before the input cursor, and the following context is usually the part after the input cursor.

[0032] According to one embodiment, the above context may include: the most recent or most recent on-screen content. According to another embodiment, the above context may include: the communication content sent by the communication peer in a communication scenario. According to yet another embodiment, the above context may include: the communication content sent to the communication peer in a communication scenario. It is understood that the embodiments of the present invention are not limited to the specific context.

[0033] The embodiment of the present invention determines the input intention according to the context, and the input intention may represent the information of the content to be input, for example, the information of the content to be input may include: category, entity type, etc. The content to be input may represent the content to be input, such as the word to be input.

[0034] The embodiment of the present invention determines keywords that match the above-mentioned input intent from the context. The keywords can be used to enrich and supplement the input intent. They are used together with the input intent to determine the association candidates, which can improve the accuracy of the association candidates. In other words, the context may include longer content, and the longer content may include irrelevant content that does not match the input intent. Using these irrelevant contents to determine the association candidates will cause a certain degree of interference and easily affect the accuracy of the association candidates. The embodiment of the present invention determines keywords that match the above-mentioned input intent from the context, and uses the above-mentioned keywords and the above-mentioned input intent to determine the association candidates, which can improve the accuracy of the association candidates.

[0035] In an application example of an embodiment of the present invention, assuming that the above text A is "A movie was recently released, and its name is", the embodiment of the present invention can determine the input intention as "movie name" for the above text A, and can determine the following keyword A from the above text A: "recently", etc. Further, based on the above text A and the keyword A, the name of the recently released movie can be obtained, such as "Ready Player One", "Avengers 3" and other association candidates.

[0036] The embodiment of the present invention can be applied to input method programs of various input methods such as keyboard symbols and handwriting, that is, users can input text through coded character strings, and the input string can refer to the coded character string input by the user. In the field of input methods, for input method programs such as Chinese, Japanese, Korean, or other languages, the input string input by the user can usually be converted into candidate items of the corresponding language. The following is mainly explained by taking Chinese as an example, and other languages ​​such as Japanese and Korean can be referenced to each other. It can be understood that the Chinese input method can include but is not limited to full spelling, simplified spelling, strokes, Wubi, etc., and the embodiment of the present invention does not limit the specific input method program corresponding to a certain language.

[0037] Taking Chinese input as an example, the types of encoded character strings may include: Pinyin string, character string (such as Wubi string, etc.) Taking English input as an example, the types of encoded character strings may include: letter string, etc.

[0038] The embodiment of the present invention can be applied to an input method program of a voice input mode, and the voice input mode can provide corresponding candidates according to the voice input by the user. Specifically, in the embodiment of the present invention, when the input mode of the input method program is a voice input mode, the embodiment of the present invention can provide corresponding association candidates according to the context.

[0039] For example, in a communication scenario, user A is going to recommend a movie to a friend, but does not remember the name of the movie, but only remembers the actors in the movie. The context of user A's voice input may include: "The movie that was recently released, actor A played the role, it's pretty good", then the embodiment of the present invention can determine the corresponding input intention as "movie" based on the context, and can provide association candidates including "movie name". The user can click on the association candidate to put it on the screen, or the user can continue to input voice.

[0040] The input method provided by the embodiment of the present invention can be applied to Figure 1 In the application environment shown, Figure 1 As shown, the client 100 and the server 200 are located in a wired or wireless network, and the client 100 and the server 200 perform data exchange through the wired or wireless network.

[0041] Optionally, the client 100 may be run on a terminal, which specifically includes but is not limited to: a smart phone, a tablet computer, an e-book reader, an MP3 (Moving Picture Experts Group Audio Layer III) player, an MP4 (Moving Picture Experts Group Audio Layer IV) player, a laptop computer, a car computer, a desktop computer, a set-top box, a smart TV, a wearable device, etc. The client 100 may correspond to a website or an APP (Application).

[0042] In actual applications, for the input method of keyboard symbols, the user can input the above input string through a physical keyboard or a virtual keyboard. For example, for a terminal with a touch screen, a virtual keyboard can be set in the input interface to input the input string by triggering the virtual keys included in the above virtual keyboard. Optionally, examples of the above virtual keyboard may include: a 9-key keyboard and a 26-key keyboard. In addition, it can be understood that in addition to the virtual keys corresponding to letters, the above input interface can also be provided with symbol keys, number keys, such as function keys for switching between Chinese and English, or toolbar keys, etc. It can be understood that the embodiment of the present invention does not limit the specific keys included in the input interface.

[0043] According to some embodiments, the input string may include, but is not limited to, a key symbol or a combination of multiple key symbols input by a user through a key. The key symbol may specifically include: pinyin, strokes, kana, and the like.

[0044] In an embodiment of the present invention, a candidate may be used to represent one or more characters provided by an input method program to be selected by a user. A candidate may correspond to a context, or a candidate may correspond to an input string, or a candidate may correspond to an input string and a context. A candidate may be a character in a language such as Chinese characters, English characters, Japanese characters, or a combination of symbols in the form of emoticons, pictures, and the like. The emoticons include, but are not limited to, pictures composed of lines, symbols, and words. For example, examples of the emoticons may include: “:P”, “:-o”, “:-)”, and the like.

[0045] Method Example 1

[0046] Reference Figure 2 , shows a flow chart of the steps of an input method embodiment 1 of the present invention, which may specifically include the following steps:

[0047] Step 201: Determine input intention based on context;

[0048] Step 202: determining keywords matching the input intention from the context;

[0049] Step 203: Determine association candidates corresponding to the context according to the input intention and the keywords.

[0050] Figure 2 The method embodiment shown may be executed by a client and / or a server. It is to be understood that the embodiment of the present invention does not limit the specific execution subject of the method embodiment.

[0051] In step 201, the context may include: previous context, and / or following context. According to the context, the input intention is determined. The input intention may represent the information of the content to be input, for example, the information of the content to be input may include: category, entity type and other information.

[0052] The embodiment of the present invention may provide the following determination method for determining input intention:

[0053] Determination method 1: Determine the input intent based on the context and the mapping relationship between the context and the input intent; or

[0054] Determination method 2: Analyze the syntactic structure of the context and obtain the input intention based on the result of the syntactic structure analysis.

[0055] For determination method 1, the mapping relationship may represent the regularity of the relationship between the context and the input intention. The above mapping relationship may be established in advance.

[0056] Optionally, the above mapping relationship can be obtained based on the corpus and the annotated input intent corresponding to the corpus. The corpus can include context, and the input intent corresponding to the context can be obtained through annotation.

[0057] The input intention corresponding to the context may specifically include:

[0058] In the above text, "Jay Chou sings", the input intent is "song";

[0059] In the above text “What I ate in the morning”, the input intent is “breakfast”;

[0060] In the above sentence "Go to the cinema to watch movies in the evening", the input intent is "movie";

[0061] In the above text, “President of South Korea”, the input intent is “name”;

[0062] In the above text, “Yao Ming’s height,” the input intent is “character attributes.”

[0063] In an embodiment of the present invention, optionally, a data analyzer can be used to characterize the mapping relationship between context and input intent. Correspondingly, the method may further include: training training data to obtain a data analyzer; the data analyzer may be used to characterize the mapping relationship between context and input intent; the training data may include: corpus, and annotated input intent corresponding to the corpus.

[0064] In an optional embodiment of the present invention, the mathematical model may be trained based on the training data to obtain a data analyzer, which may characterize the mapping relationship between the input data (context) and the output data (input intent).

[0065] Mathematical model is a scientific or engineering model constructed by mathematical logic methods and mathematical language. Mathematical model is a mathematical structure that is expressed in general or approximate terms using mathematical language with reference to the characteristics or quantitative dependencies of a certain system of things. This mathematical structure is a relational structure depicted with the help of mathematical symbols. Mathematical model can be one or a group of algebraic equations, differential equations, difference equations, integral equations or statistical equations and their combinations, which quantitatively or qualitatively describe the mutual relationship or causal relationship between the variables of the system. In addition to mathematical models described by equations, there are also models described by other mathematical tools, such as algebra, geometry, topology, mathematical logic, etc. Among them, mathematical models describe the behavior and characteristics of the system rather than the actual structure of the system. Among them, machine learning, deep learning methods and other methods can be used to train mathematical models. Machine learning methods may include: linear regression, decision tree, random forest, etc. Deep learning methods may include: convolutional neural networks (CNN), long short-term memory networks (LSTM), gated recurrent units (GRU), etc.

[0066] For determination method 2, the context may be subjected to syntactic structure analysis, and the input intent may be obtained based on the result of the syntactic structure analysis.

[0067] Syntactic structure analysis can be used to determine the grammatical structure of a sentence or the dependency relationship between words in a sentence. Optionally, the topic information of the context can be obtained based on the result of the syntactic structure analysis as the input intent.

[0068] In an optional embodiment of the present invention, the above-mentioned determining the input intention specifically includes: determining the input intention according to the context and input environment characteristics.

[0069] The input environment feature can be used to characterize the environment information of the terminal when the user inputs. The input environment feature can reflect the user's input intention to a certain extent. Therefore, the embodiment of the present invention uses the input environment feature in the process of determining the input intention to improve the accuracy of the input intention.

[0070] In practical applications, the input environment features may include various types of features. Optionally, the input environment features may include at least one of time environment features, location environment features, climate environment features, application environment features, and page environment features.

[0071] Even for the same terminal, the environment information is likely to change, and the time environment feature is a typical example. Therefore, the input environment feature of the embodiment of the present invention can be real-time, and the input environment feature corresponding to the input string can be obtained in real time during the input process.

[0072] For an input string, its receiving time can be used as the corresponding time environment feature.

[0073] The location information obtained according to its IP (Internet Protocol) address, the GPS (Global Positioning System) of the terminal or the mobile communication network can be used as the corresponding location environment feature.

[0074] As a host program, the input method program can be hosted in any host program and can be called by the host program to implement input in the host program. For example, a user can type an input string in the host program and select a candidate item corresponding to the input string. In the embodiment of the present invention, the application environment feature corresponding to the input string can be information of the host program corresponding to the input method program.

[0075] Optionally, the application environment characteristics corresponding to the input string can be determined based on the identification characteristics of the current object being served by the input method program. For example, when the input method program is running, it calls GetModuleFilename to find the program path name "C:ProgramFilesMicrosoftOfficeOFFICE11WINWORD.EXE", and the corresponding application environment characteristics can be determined to be "WinWord.exe", that is, the input string is entered in "word". Of course, the embodiment of the present invention does not limit the specific method for obtaining the application environment characteristics corresponding to the input string.

[0076] In an embodiment of the present invention, the application environment feature may include: an application identifier and / or an application category. For example, "word" is an application identifier, and the application category corresponding to "word" is an office category, etc. It can be understood that those skilled in the art can classify applications into corresponding application categories according to actual application requirements. For example, examples of application categories may include but are not limited to: instant messaging category, document category, search category, web page category, shopping category, travel category, etc.

[0077] Page environment features can be used to characterize the page environment provided by an application or website. Optionally, the above-mentioned page environment may include but is not limited to: instant messaging page environment, document page environment, email page environment, password input page environment, game page environment, search page environment, travel page environment, shopping page environment, social page environment, film and television page environment, reading page environment, etc.

[0078] Of course, in addition to time environment features, location environment features, application environment features and page environment features, the input environment features of the embodiment of the present invention may also include other environment features, for example, physical environment features such as air pressure, altitude, temperature, humidity, etc. It can be understood that the embodiment of the present invention does not limit the specific input environment features.

[0079] For example, when the context is "today's weather", the input intention can be determined based on the context and the location environment feature. Assuming the location environment feature is "A region", the input intention is "A region weather" to obtain more accurate association candidates.

[0080] For example, when the context is "I don't know what to eat", the input intention can be determined based on the context and page environment features. Assuming the page environment feature is "takeout", the input intention is "takeout recommendation" to obtain more accurate association candidates.

[0081] For another example, when the context is "what movies are released recently", the input intention can be determined based on the context and page environment features. Assuming the page environment feature is "movie tickets", the input intention is "movie recommendation" to obtain more accurate association candidates.

[0082] Optionally, a mapping relationship between context, input environment features and input intent may be pre-established, so that the mapping relationship between context, input environment features and input intent may be searched based on the context and input environment features to obtain the input intent.

[0083] The process of determining input intent is described in detail above through determination method 1 and determination method 2. It can be understood that those skilled in the art can adopt any one or a combination of determination method 1 and determination method 2 according to actual application requirements, or other determination methods can be adopted. For example, semantic analysis can be performed on the context to obtain input intent, etc. The embodiment of the present invention does not limit the specific process of determining input intent.

[0084] In step 202, keywords matching the input intention may be determined from the context.

[0085] In an optional embodiment of the present invention, the above-mentioned determining the keywords matching the above-mentioned input intent from the above-mentioned context specifically includes: segmenting the context, determining the matching degree between the segmented words and the input intent respectively, and taking the segmented words whose matching degree exceeds the matching degree threshold as keywords. Optionally, the word vectors corresponding to the segmented words and the input intent can be determined respectively, and the matching degree between the segmented words and the input intent can be determined by using the distance between the word vectors.

[0086] In another optional embodiment of the present invention, the above-mentioned determining the keywords matching the above-mentioned input intention from the above-mentioned context specifically includes: identifying entity words in the above-mentioned context, respectively determining the matching degree between the entity words and the input intention, and taking the entity words whose matching degree exceeds the matching degree threshold as keywords.

[0087] In the embodiment of the present invention, an entity is a specific thing or concept. Entities are generally divided into types, such as person entities, movie entities, animal entities, historical entities, etc. The same entity can correspond to multiple entity instances, and an entity instance can be a descriptive page (content) of an entity on the Internet (or other media), such as an encyclopedia page that contains an entity instance corresponding to the entity.

[0088] Optionally, entities may include: named entities, which may refer to names of people, institutions, places, and all other entities identified by names. More extensive named entities may also include: book titles, song titles, film and television drama titles, product names, brand names, numbers, dates, currencies, addresses, and so on.

[0089] Optionally, the entity category may include at least one of the following categories: people, places, fruits, vegetables, animals, plants, buildings, clothing, food, medicines, vehicles, furniture, musical instruments, electrical appliances, and natural phenomena.

[0090] In one embodiment of the present invention, a NER (Named Entity Recognition) method may be used to determine entities in the text content.

[0091] According to one embodiment, the NER method may include: a thesaurus-based method. The thesaurus-based method can construct an entity library for high-frequency words based on the frequency of occurrence of phrases, and directly identify the words that can be retrieved in the entity library as entities. Among them, a phrase can refer to a combination of two or more words. In practical applications, entity-related data can be captured from the Internet, and the entity-related data can be analyzed to obtain corresponding entity words, and the entity words can be stored in the entity library. The embodiment of the present invention does not limit the specific entity words and their acquisition methods.

[0092] According to another embodiment, the NER method may include: a rule-based method. The rule-based method may mark the phrases in the request that meet the corresponding rules as entities according to the composition rules of the phrases.

[0093] According to another embodiment, the NER method may include: a method based on statistical learning. The method based on statistical learning regards named entity recognition as a classification problem, and adopts classification methods such as SVM (Support Vector Machine) and Bayesian; or, regards named entity recognition as a sequence labeling problem, and adopts sequence labeling models such as HMM (Hidden Markov Model), Maximum Entropy Model, CRF (Conditional Random Field, Conditional Random Field Algorithm), LSTM (Long Short-Term Memory Network, Long Short-Term Memory) model.

[0094] In an application example of the present invention, assuming that the input intention is "weather", keywords related to "weather", such as place names, time, etc., can be determined from the context.

[0095] In step 203, association candidates corresponding to the context may be determined according to the input intention and the keywords.

[0096] In an optional embodiment of the present invention, a search may be performed in the mapping relationship between the input intent, the keywords and the elements according to the input intent and the keywords to obtain the association candidates corresponding to the above context.

[0097] In the embodiment of the present invention, optionally, the mapping relationship between the input intention, keywords and elements can be represented by multi-relation data.

[0098] Multivariate relational data may include binary and more than binary relational data. A binary relation, also known as a 2-gram, is used to represent the probability of two elements appearing one after another. In the field of input methods, the elements here may include at least one of: vocabulary, phrases, short sentences, letters, numbers, and symbols. In an embodiment of the present invention, the binary may mainly include binary relations of vocabulary, and other types of binary relations may refer to each other. Relations of more than two elements are used to represent the probability of more than two elements appearing one after another.

[0099] The binary relationship data can be used to reflect the probability of two words being used adjacently. On the one hand, the vocabulary can include language characters composed of letters, which can be words, phrases, and short phrases composed of letters printed on a keyboard, and can be specifically applicable to English, French, German, etc. On the other hand, the vocabulary can also include character sequences corresponding to text languages ​​composed of pinyin and / or strokes. Here, the character sequences corresponding to text languages ​​composed of pinyin and / or strokes can include words corresponding to pinyin, words corresponding to strokes, etc., and can be specifically applicable to Chinese, Japanese, Korean, etc.

[0100] In an optional embodiment of the present invention, multivariate relational data may be represented by a data model. The types of data models may include, but are not limited to: language models, neural network models, and the like.

[0101] The above multivariate relational data can be used to obtain P(any element|input intent, keyword, ...), which is the probability of any element under the conditions of mathematical models such as input intent and keywords. Based on this probability, the association candidates corresponding to the input intent and keywords can be determined. Among them, the corpus used in the data model may include: Internet corpus, user dialogue corpus, user input corpus, etc.; dialogue corpus may refer to the corpus corresponding to the dialogue established by the user in the communication scenario.

[0102] Optionally, the conditions of the multivariate relationship data may further include: input environment characteristics. In this case, the data model may provide P (any element|input intent, keyword, input environment characteristics, ...).

[0103] In another optional embodiment of the present invention, a query can be performed in a database corresponding to the input intent based on the input intent and keywords to obtain association candidates. By obtaining association candidates based on database query, real-time data can be obtained.

[0104] Optionally, a database corresponding to the input intent may be provided by a service provider. For example, the types of databases may include: weather database, music database, etc.

[0105] In an application example of an embodiment of the present invention, it is assumed that the context is "today's weather", and it is assumed that based on the context and geographic location features, the input intention is determined to be "weather in area A", and the keyword "today" corresponding to the input intention can be determined from the context; then, based on "weather in area A" and "today", the weather database is queried to find that the weather in area A today is sunny, and therefore association candidates such as "sunny", "clear sky", and "not bad" can be given.

[0106] In another application example of the present invention, assuming the context is "a science fiction movie will be released soon", the input intention "movie name" can be obtained based on the context, and the following keywords matching the input intention can be obtained from the context: "recently", "science fiction"; and then based on the input intention and keywords, the recently released science fiction movies are queried, including "Ready Player One", "Avengers 3" and other association candidates.

[0107] In summary, the input method of the embodiment of the present invention determines the input intention based on the context. The input intention can represent the information of the content to be input. For example, the information of the content to be input can include: category, entity type and other information.

[0108] Furthermore, the embodiment of the present invention determines keywords matching the above input intent from the context. The keywords can enrich and supplement the input intent, and are used together with the input intent to determine association candidates, thereby improving the accuracy of the association candidates.

[0109] Method Example 2

[0110] Reference Figure 3 , shows a flow chart of the steps of Embodiment 2 of an input method of the present invention, which may specifically include the following steps:

[0111] Step 301: Determine input intention based on context;

[0112] Step 302: determining keywords matching the input intention from the context;

[0113] Step 303: Determine association candidates corresponding to the context according to the input intention and the keywords.

[0114] Relative to Figure 2 The method of the first embodiment shown in the figure may further include:

[0115] Step 304: determining quality parameters of the association candidates according to the parameters of the input intention and / or the parameters of the keyword and / or the parameters of the association candidates;

[0116] Step 305: output the above-mentioned association candidates according to the above-mentioned quality parameters.

[0117] The embodiment of the present invention can also determine the quality parameter of the association candidate, and the quality parameter can characterize the quality of the association candidate. According to the quality parameter, the association candidate is outputted, and several association candidates with higher quality can be provided to the user.

[0118] In step 304, the factors determining the quality parameter may include at least one of the following factors: parameters of the input intention, parameters of the keyword, and parameters of the association candidate.

[0119] Among them, the parameters of the input intent can characterize the quality or accuracy of the input intent. Optionally, the parameters of the input intent can be determined based on the click information of the association candidates corresponding to the input intent. Optionally, the higher the click rate of the association candidates corresponding to the input intent, the higher the quality of the input intent. Conversely, the lower the click rate of the association candidates corresponding to the input intent, the lower the quality of the input intent. The above-mentioned click rate can be determined based on the number of displays of the association candidates corresponding to the input intent and the number of clicks on the association candidates corresponding to the input intent.

[0120] The parameters of the keyword can be used to characterize the importance of the keyword. According to one embodiment, the parameters of the keyword can be obtained based on the distance parameter between the keyword and the association candidate. Generally, the closer the distance represented by the distance parameter, the higher the importance of the keyword, and conversely, the farther the distance represented by the distance parameter, the lower the importance of the keyword.

[0121] According to another embodiment, the attention mechanism of a neural network may be used to determine the importance of a keyword in a context.

[0122] The parameters of the association candidate can be used to characterize the importance or quality of the association candidate. The parameters of the association candidate can include: the frequency of the word corresponding to the association candidate, and / or whether the frequency of the word corresponding to the association candidate is from the user vocabulary, and / or the matching degree between the association candidate and the input intention, etc. Whether the corresponding word is from the user vocabulary, etc. Generally, if the frequency of the word corresponding to the association candidate is from the user vocabulary, the quality of the association candidate is higher.

[0123] In an optional embodiment of the present invention, the association candidates can be sorted in order from high to low according to the quality represented by the quality parameter to obtain a corresponding sorting result. Optionally, the N association candidates with the highest quality can be selected from the sorting result and displayed to the user. Alternatively, M association candidates whose quality parameters exceed the parameter threshold can be selected from the sorting result and displayed to the user. Wherein, N and M can be natural numbers.

[0124] In step 305, the server may output the association candidates to the client, or the client may display the association candidates to the user.

[0125] It should be noted that, for the method embodiments, for the sake of simplicity of description, they are all described as a series of motion combinations, but those skilled in the art should know that the embodiments of the present invention are not limited by the described sequence of actions, because according to the embodiments of the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the motions involved are not necessarily required by the embodiments of the present invention.

[0126] Device Embodiment

[0127] Reference Figure 4 , shows a structural block diagram of an input device embodiment of the present invention, which may specifically include:

[0128] Intention determination module 401, used to determine input intention based on context;

[0129] A keyword determination module 402, configured to determine keywords matching the input intent from the context; and

[0130] The association candidate determination module 403 is used to determine the association candidates corresponding to the context according to the input intention and the keyword.

[0131] Optionally, the intention determination module 401 may include:

[0132] A first intention determination module is used to determine the input intention according to the context and the mapping relationship between the context and the input intention; or

[0133] The second intention determination module is used to perform syntactic structure analysis on the context and obtain the input intention based on the syntactic structure analysis result.

[0134] Optionally, the above mapping relationship is obtained based on the corpus and the annotation input intention corresponding to the corpus.

[0135] Optionally, the intention determination module 401 may include:

[0136] The third intention determination module is used to determine the input intention based on the context and input environment characteristics.

[0137] Optionally, the above input environment characteristics may include at least one of the following characteristics:

[0138] Time environment characteristics, location environment characteristics, climate environment characteristics, application environment characteristics and page environment characteristics.

[0139] Optionally, the above device may further include:

[0140] A quality parameter determination module, used to determine the quality parameters of the association candidates according to the parameters of the input intention and / or the parameters of the keyword and / or the parameters of the association candidates;

[0141] The association candidate output module is used to output the above association candidates according to the above quality parameters.

[0142] Optionally, the parameter of the keyword is obtained based on a distance parameter between the keyword and the association candidate.

[0143] 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.

[0144] 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.

[0145] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0146] An embodiment of the present invention provides a device for input, 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. The one or more programs include instructions for performing the following operations: determining input intent based on a context; determining keywords matching the input intent from the context; and determining association candidates corresponding to the context based on the input intent and the keywords.

[0147] Figure 5 8 is a block diagram of an input device 800 according to an exemplary embodiment. For example, the device 800 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.

[0148] Reference Figure 5 , the device 800 may include one or more of the following components: a processing component 802 , a memory 804 , a power 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 .

[0149] The processing component 802 generally controls the overall operation of the device 800, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the above-mentioned method. In addition, the processing component 802 may include one or more modules to facilitate the interaction between the processing component 802 and other components. For example, the processing component 802 may include a multimedia module to facilitate the interaction between the multimedia component 808 and the processing component 802.

[0150] The memory 804 is configured to store various types of data to support operations on the device 800. Examples of such data include instructions for any application or method operating on the device 800, contact data, phone book data, messages, pictures, videos, etc. The 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 memory, flash memory, magnetic disk or optical disk.

[0151] The power supply component 806 provides power to the various components of the device 800. The power supply component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 800.

[0152] The multimedia component 808 includes a screen that provides an output interface between the above-mentioned 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 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 above-mentioned touch sensor can not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the above-mentioned touch or slide operation. In some embodiments, the multimedia component 808 includes a front camera and / or a rear camera. When the device 800 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera can receive external multimedia data. Each front camera and rear camera can be a fixed optical lens system or have a focal length and optical zoom capability.

[0153] The audio component 810 is configured to output and / or input audio signals. For example, the audio component 810 includes a microphone (MIC), and when the device 800 is in an operating mode, such as a call mode, a recording mode, and a voice input mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 804 or sent via the communication component 816. In some embodiments, the audio component 810 also includes a speaker for outputting audio signals.

[0154] 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 button, volume button, start button, and lock button.

[0155] The sensor assembly 814 includes one or more sensors for providing various aspects of status assessment for the device 800. For example, the sensor assembly 814 can detect the open / closed state of the device 800, the relative positioning of the components, such as the display and keypad of the device 800, the sensor assembly 814 can also detect the position change of the device 800 or a component of the device 800, the presence or absence of user contact with the device 800, the orientation or acceleration / deceleration of the device 800 and the temperature change of the device 800. The sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 814 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 814 may also include an accelerometer, a gyroscope sensor, a magnetic sensor, a pressure sensor or a temperature sensor.

[0156] The communication component 816 is configured to facilitate wired or wireless communication between the device 800 and other devices. The device 800 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 816 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 816 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, Radio Frequency Identification) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0157] 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 above method.

[0158] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 804 including instructions, and the instructions can be executed by the processor 820 of the device 800 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.

[0159] Figure 61 is a schematic diagram of the structure of the server in some embodiments of the present invention. The server 1900 may have relatively large differences due to different configurations or performances, and may include one or more central processing units (CPU) 1922 (for example, one or more processors) and memory 1932, and one or more storage media 1930 (for example, one or more mass storage devices) storing application programs 1942 or data 1944. Among them, the memory 1932 and the storage medium 1930 can be short-term storage or permanent storage. The program stored in the storage medium 1930 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 1922 can be configured to communicate with the storage medium 1930 and execute a series of instruction operations in the storage medium 1930 on the server 1900.

[0160] The server 1900 may also include one or more power supplies 1926, one or more wired or wireless network interfaces 1950, one or more input and output interfaces 1958, one or more keyboards 1956, and / or, one or more operating systems 1941, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.

[0161] 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 Figure 2 or Figure 3 Input method shown.

[0162] A non-temporary computer-readable storage medium. When the instructions in the storage medium are executed by a processor of a device (server or terminal), the device is able to execute an input method. The method includes: determining an input intention based on a context; determining a keyword matching the input intention from the context; and determining an association candidate corresponding to the context based on the input intention and the keyword.

[0163] The embodiment of the present invention discloses A1, an input method, the method comprising:

[0164] Determine input intent based on context;

[0165] Determining keywords matching the input intent from the context;

[0166] According to the input intention and the keyword, an association candidate corresponding to the context is determined.

[0167] A2. According to the method described in A1, the determining the input intention comprises:

[0168] Determine the input intent based on the context and the mapping relationship between the context and the input intent; or

[0169] Perform syntactic structure analysis on the context and obtain input intent based on the results of syntactic structure analysis.

[0170] A3. According to the method described in A2, the mapping relationship is obtained based on the corpus and the annotation input intention corresponding to the corpus.

[0171] A4. According to the method described in A1, the determining the input intention comprises:

[0172] Determine input intent based on context and input environment characteristics.

[0173] A5. According to the method described in A4, the input environment feature includes at least one of the following features:

[0174] Time environment characteristics, location environment characteristics, climate environment characteristics, application environment characteristics and page environment characteristics.

[0175] A6. The method according to any one of A1 to A5, further comprising:

[0176] Determining quality parameters of the association candidates according to the parameters of the input intention and / or the parameters of the keyword and / or the parameters of the association candidates;

[0177] The association candidates are output according to the quality parameter.

[0178] A7. According to the method described in A6, the parameter of the keyword is obtained based on the distance parameter between the keyword and the association candidate.

[0179] The embodiment of the present invention discloses B8, an input device, comprising:

[0180] The intention determination module is used to determine the input intention based on the context;

[0181] A keyword determination module, configured to determine keywords matching the input intent from the context; and

[0182] The association candidate determination module is used to determine the association candidates corresponding to the context according to the input intention and the keyword.

[0183] B9. According to the device described in B8, the intention determination module includes:

[0184] A first intention determination module is used to determine the input intention according to the context and the mapping relationship between the context and the input intention; or

[0185] The second intention determination module is used to perform syntactic structure analysis on the context and obtain the input intention based on the syntactic structure analysis result.

[0186] B10. According to the device described in B9, the mapping relationship is obtained based on the corpus and the annotation input intention corresponding to the corpus.

[0187] B11. According to the device described in B8, the intention determination module includes:

[0188] The third intention determination module is used to determine the input intention based on the context and input environment characteristics.

[0189] B12. According to the device described in B11, the input environment feature includes at least one of the following features:

[0190] Time environment characteristics, location environment characteristics, climate environment characteristics, application environment characteristics and page environment characteristics.

[0191] B13. The device according to any one of B8 to B12, further comprising:

[0192] A quality parameter determination module, used to determine the quality parameters of the association candidates according to the parameters of the input intention and / or the parameters of the keyword and / or the parameters of the association candidates;

[0193] The association candidate output module is used to output the association candidates according to the quality parameters.

[0194] B14. According to the device described in B13, the parameter of the keyword is obtained based on the distance parameter between the keyword and the association candidate.

[0195] The embodiment of the present invention discloses C15, a device for input, 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:

[0196] Determine input intent based on context;

[0197] Determining keywords matching the input intent from the context;

[0198] According to the input intention and the keyword, an association candidate corresponding to the context is determined.

[0199] C16. According to the device of C15, the determining the input intention comprises:

[0200] Determine the input intent based on the context and the mapping relationship between the context and the input intent; or

[0201] Perform syntactic structure analysis on the context and obtain input intent based on the results of syntactic structure analysis.

[0202] C17. According to the device described in C16, the mapping relationship is obtained based on the corpus and the annotation input intention corresponding to the corpus.

[0203] C18. According to the device of C15, the determining the input intention comprises:

[0204] Determine input intent based on context and input environment characteristics.

[0205] C19. According to the device of C18, the input environment feature includes at least one of the following features:

[0206] Time environment characteristics, location environment characteristics, climate environment characteristics, application environment characteristics and page environment characteristics.

[0207] C20. The device according to any one of C15 to C19, wherein the device is further configured to execute, by one or more processors, the one or more programs including instructions for performing the following operations:

[0208] Determining quality parameters of the association candidates according to the parameters of the input intention and / or the parameters of the keyword and / or the parameters of the association candidates;

[0209] The association candidates are output according to the quality parameter.

[0210] C21. According to the device described in C20, the parameter of the keyword is obtained based on the distance parameter between the keyword and the association candidate.

[0211] An embodiment of the present invention discloses D22, a machine-readable medium having instructions stored thereon, which, when executed by one or more processors, enables a device to execute an input method as described in one or more of A1 to A7.

[0212] Those skilled in the art will readily appreciate other embodiments of the invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art that are not disclosed in this disclosure. The description and examples are to be considered exemplary only, and the true scope and spirit of the present invention are indicated by the following claims.

[0213] It should be understood that the present invention is not limited to the exact construction that has been described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

[0214] The above are only preferred embodiments of the present invention and are 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 in the protection scope of the present invention.

[0215] The above is a detailed introduction to an input method, an input device and a device for input 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: The method comprises: According to the context and input environment features, searching for a mapping relationship between the context, the input environment features and the input intent, and determining the input intent; Building an entity library for high-frequency words according to the frequency of occurrence of phrases, and determining the words in the context retrieved from the entity library as entity words; Determine the matching degree between the entity words and the input intent respectively, and use the entity words whose matching degree exceeds the matching degree threshold as keywords corresponding to the input intent; Determine multivariate relationship data for representing the mapping relationship between the input intention, the keywords and elements; wherein the elements include: at least one of words, phrases, short sentences, letters, numbers and symbols; Taking the input intention, the keyword and the input environment feature as conditions of the multivariate relationship data, and obtaining the probability of the element; According to the probability of the element, an association candidate corresponding to the context is determined.

2. The method according to claim 1, characterized in that: The determining the input intention further comprises: Determine the input intent based on the context and the mapping relationship between the context and the input intent; or Perform syntactic structure analysis on the context and obtain input intent based on the results of syntactic structure analysis.

3. The method according to claim 2, characterized in that The mapping relationship is obtained based on the corpus and the annotation input intention corresponding to the corpus.

4. The method according to claim 1, characterized in that: The input environment feature includes at least one of the following features: Time environment characteristics, location environment characteristics, climate environment characteristics, application environment characteristics and page environment characteristics.

5. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: Determining quality parameters of the association candidates according to the parameters of the input intention and / or the parameters of the keyword and / or the parameters of the association candidates; The association candidates are output according to the quality parameter.

6. The method according to claim 5, characterized in that The parameter of the keyword is obtained according to the distance parameter between the keyword and the association candidate.

7. An input device, characterized in that: include: An intention determination module is used to find a mapping relationship between the context, the input environment characteristics and the input intention according to the context and the input environment characteristics, and determine the input intention; A keyword determination module is used to construct an entity library for high-frequency words according to the frequency of occurrence of the phrase, and determine the words in the context retrieved from the entity library as entity words; respectively determine the matching degree between the entity words and the input intent, and use the entity words whose matching degree exceeds the matching degree threshold as keywords corresponding to the input intent; as well as An association candidate determination module is used to determine multivariate relational data used to characterize the mapping relationship between the input intent, the keywords and the elements; wherein the elements include: at least one of words, phrases, short sentences, letters, numbers and symbols; taking the input intent, the keywords and the input environment characteristics as conditions of the multivariate relational data to obtain the probability of the elements; and determining the association candidates corresponding to the context based on the probability of the elements.

8. The device according to claim 7, characterized in that The intention determination module includes: A first intention determination module is used to determine the input intention according to the context and the mapping relationship between the context and the input intention; or The second intention determination module is used to perform syntactic structure analysis on the context and obtain the input intention based on the syntactic structure analysis result.

9. The device according to claim 8, characterized in that The mapping relationship is obtained based on the corpus and the annotation input intention corresponding to the corpus.

10. The device according to claim 7, characterized in that The input environment feature includes at least one of the following features: Time environment characteristics, location environment characteristics, climate environment characteristics, application environment characteristics and page environment characteristics.

11. The device according to any one of claims 7 to 10, characterized in that The device also includes: A quality parameter determination module, used to determine the quality parameters of the association candidates according to the parameters of the input intention and / or the parameters of the keyword and / or the parameters of the association candidates; The association candidate output module is used to output the association candidates according to the quality parameters.

12. The device according to claim 11, characterized in that The parameter of the keyword is obtained according to the distance parameter between the keyword and the association candidate.

13. A device for input, characterized in that The invention comprises a memory and one or more programs, wherein the one or more programs are stored in the memory, and when the programs are executed by one or more processors, the steps of the method described in any one of claims 1 to 6 are implemented.

14. A computer-readable medium having instructions stored thereon, which when executed by one or more processors can implement the input method according to any one of claims 1 to 6.

Images