CN111381685B - Sentence association method and sentence association device - Google Patents
Sentence association method and sentence association device Download PDFInfo
- Publication number
- CN111381685B CN111381685B CN201811639917.0A CN201811639917A CN111381685B CN 111381685 B CN111381685 B CN 111381685B CN 201811639917 A CN201811639917 A CN 201811639917A CN 111381685 B CN111381685 B CN 111381685B
- Authority
- CN
- China
- Prior art keywords
- content
- sentence
- context
- determining
- search result
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 55
- 230000011218 segmentation Effects 0.000 claims description 20
- 230000006870 function Effects 0.000 abstract description 17
- 230000000875 corresponding effect Effects 0.000 description 75
- 238000012545 processing Methods 0.000 description 20
- 238000004891 communication Methods 0.000 description 10
- 241000219095 Vitis Species 0.000 description 8
- 235000009754 Vitis X bourquina Nutrition 0.000 description 8
- 235000012333 Vitis X labruscana Nutrition 0.000 description 8
- 235000014787 Vitis vinifera Nutrition 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 230000003287 optical effect Effects 0.000 description 5
- 230000003993 interaction Effects 0.000 description 4
- 230000005236 sound signal Effects 0.000 description 4
- 238000012549 training Methods 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000002596 correlated effect Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000001960 triggered effect Effects 0.000 description 2
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
- G06F3/02—Input arrangements using manually operated switches, e.g. using keyboards or dials
- G06F3/023—Arrangements for converting discrete items of information into a coded form, e.g. arrangements for interpreting keyboard generated codes as alphanumeric codes, operand codes or instruction codes
- G06F3/0233—Character input methods
- G06F3/0237—Character input methods using prediction or retrieval techniques
Abstract
The embodiment of the application discloses a sentence association method and a related device, which can be used as keywords to perform network search according to the on-screen content to obtain corresponding search results, and because the corpus related to the on-screen content is far more than a model word stock of a local high-frequency sentence model, the method and the related device are more likely to determine the lower sentences corresponding to the on-screen content according to the search results, and generate sentence association candidates corresponding to the on-screen content according to the determined lower sentences, thereby not only playing the role of providing convenience for users to input, but also realizing sentence association functions in a search mode and improving the application range of the sentence association functions.
Description
Technical Field
The application relates to the field of input methods, in particular to a sentence association method and device.
Background
The input method system can associate the following candidate items based on the content on the screen of the user in the user input process, so that the user can directly screen the screen by selecting the candidate items, and convenience is provided for the user input. The input method system can take the associated text sentence as sentence association candidate items through a sentence association function, wherein the sentence association candidate items comprise sentences which can be word groups or long sentences formed by a plurality of words. If the user selects one sentence association candidate item, more contents can be displayed at one time, and the input efficiency is improved.
At present, the function of sentence association is realized by mainly adopting an N-Gram model commonly used in large-vocabulary continuous speech recognition. The input method system trains the model by counting high-frequency sentence combinations in the historical corpus of the user, for example, if one user inputs the sentence combination of ' eating grape without spitting grape skin ' frequently before, the user does not eat grape with spitting grape skin ', when the user inputs the sentence combination of ' eating grape without spitting grape skin ' again, the input method system can be matched through the model, and sentence association candidates of ' eating no grape with spitting grape skin ' are associatively obtained.
However, the current model training is based on the user history corpus, and meanwhile, the offline model word stock occupies strict storage requirements, so that the high-frequency sentence training high-frequency sentence model can only be extracted from the on-screen context, and the sentence association function is realized by the high-frequency sentence model, so that whether the current context has sentence association candidates of the following sentences or not can be determined, and the application range of the sentence association function is not large.
Disclosure of Invention
In order to solve the technical problems, the application provides the sentence associating method and device, which not only play a role in providing convenience for users to input, but also realize the sentence associating function in a searching mode, and improve the application range of the sentence associating function.
The embodiment of the application discloses the following technical scheme:
in a first aspect, an embodiment of the present application provides a sentence associating method, including:
acquiring the on-screen content;
searching the above content as a keyword to obtain a search result;
determining a lower sentence corresponding to the upper content according to the search result;
and generating sentence association candidates corresponding to the above content according to the context sentence.
Optionally, the determining, according to the search result, a text sentence corresponding to the text content includes:
dividing the search result into a plurality of divided results according to separators in the search result;
determining a first matching relation between the segmentation result and the above content, wherein the first matching relation is used for identifying the probability that the segmentation result appears behind the above content;
and determining the segmentation result of the first matching relation meeting a first preset condition as a text sentence corresponding to the text content.
Optionally, the determining, according to the search result, a text sentence corresponding to the text content includes:
judging whether the type of the above content is a question type;
if yes, obtaining abstract content of the search result according to the above content;
And taking the abstract content as a corresponding following sentence of the above content.
Optionally, the determining, according to the search result, a text sentence corresponding to the text content includes:
determining text content with similarity meeting a second preset condition from the search result;
and taking the text content in the search result as the corresponding text sentence of the text content.
Optionally, if a plurality of context sentences corresponding to the context are determined according to the search result, generating sentence association candidates corresponding to the context according to the context sentences includes:
determining the sorting values corresponding to the plurality of context sentences respectively;
generating sentence association candidates corresponding to the above content according to the context sentences of which the ranking values meet the threshold values;
the ranking values respectively corresponding to the plurality of following sentences are determined according to any one or a combination of the following determination modes:
first determination mode:
determining a second matching relationship between the plurality of context sentences and the context content respectively, wherein the second matching relationship is used for identifying the probability that the context sentences appear behind the context content;
Determining the sorting values respectively corresponding to the plurality of context sentences according to the second matching relation;
the second determination mode:
and determining the ranking values corresponding to the plurality of the context sentences according to the association degree of the search results of the plurality of the context sentences and the context content.
Optionally, the searching with the above content as the keyword to obtain a search result includes:
and taking the above content as a keyword, and searching in search resources related to the type of the above content to obtain the search result.
Optionally, before the searching is performed by using the above content as a keyword to obtain a search result, the method further includes:
judging whether the type of the content is a preset type or belongs to a preset field;
if yes, executing the step of searching by taking the above content as a keyword to obtain a search result.
Optionally, if the display object of the sentence association candidate item is a first user, the above content is on-screen for the first user, or the above content is on-screen for a second user.
In a second aspect, an embodiment of the present application provides a sentence associating apparatus, including an obtaining unit, a searching unit, a determining unit, and a generating unit:
The acquisition unit is used for acquiring the on-screen content;
the searching unit is used for searching the above content as a keyword to obtain a searching result;
the determining unit is used for determining a text sentence corresponding to the text content according to the search result;
the generating unit is used for generating sentence association candidates corresponding to the context according to the context sentence.
Optionally, the determining unit is specifically configured to:
dividing the search result into a plurality of divided results according to separators in the search result;
determining a first matching relation between the segmentation result and the above content, wherein the first matching relation is used for identifying the probability that the segmentation result appears behind the above content;
and determining the segmentation result of the first matching relation meeting a first preset condition as a text sentence corresponding to the text content.
Optionally, the determining unit is specifically configured to:
judging whether the type of the above content is a question type;
if yes, obtaining abstract content of the search result according to the above content;
and taking the abstract content as a corresponding following sentence of the above content.
Optionally, the determining unit is specifically configured to:
determining text content with similarity meeting a second preset condition from the search result;
and taking the text content in the search result as the corresponding text sentence of the text content.
Optionally, if a plurality of context sentences corresponding to the above content are determined according to the search result, the generating unit is specifically configured to:
determining the sorting values corresponding to the plurality of context sentences respectively;
generating sentence association candidates corresponding to the above content according to the context sentences of which the ranking values meet the threshold values;
the ranking values respectively corresponding to the plurality of following sentences are determined according to any one or a combination of the following determination modes:
first determination mode:
determining a second matching relationship between the plurality of context sentences and the context content respectively, wherein the second matching relationship is used for identifying the probability that the context sentences appear behind the context content;
determining the sorting values respectively corresponding to the plurality of context sentences according to the second matching relation;
the second determination mode:
and determining the ranking values corresponding to the plurality of the context sentences according to the association degree of the search results of the plurality of the context sentences and the context content.
Optionally, the search unit is specifically configured to:
and taking the above content as a keyword, and searching in search resources related to the type of the above content to obtain the search result.
Optionally, the device further includes a judging unit:
the judging unit is used for judging whether the type of the content is a preset type or belongs to a preset field;
and if the judging unit judges that the type of the above content is a preset type or belongs to a preset field, triggering the searching unit to execute the step of searching by taking the above content as a keyword to obtain a search result.
Optionally, if the display object of the sentence association candidate item is a first user, the above content is on-screen for the first user, and/or the above content is on-screen for a second user.
In a third aspect, embodiments of the present application provide an apparatus for sentence association, comprising a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by one or more processors, the one or more programs comprising instructions for:
Acquiring the on-screen content;
searching the above content as a keyword to obtain a search result;
determining a lower sentence corresponding to the upper content according to the search result;
and generating sentence association candidates corresponding to the above content according to the context sentence.
In a fourth aspect, embodiments of the present application provide a machine-readable medium having instructions stored thereon, which when executed by one or more processors, cause an apparatus to perform a sentence association method as described in one or more of the first aspects.
According to the technical scheme, according to the above content which is already displayed, the above content can be used as a keyword to perform network search to obtain a corresponding search result, and because the corpus related to the above content on the network is far more than a model word stock of a local high-frequency sentence model, the text sentences corresponding to the above content are more likely to be determined according to the search result, sentence association candidates corresponding to the above content are generated according to the determined text sentences, so that the function of providing convenience for the user to input is achieved, the sentence association function is realized in a searching mode, and the application range of the sentence association function is improved.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive faculty for a person skilled in the art.
Fig. 1 is an exemplary diagram of an application scenario of a sentence associating method according to an embodiment of the present application;
FIG. 2 is a flowchart of a sentence association method according to an embodiment of the present disclosure;
FIG. 3 is a block diagram of a sentence associating device according to an embodiment of the present application;
FIG. 4 is a block diagram of an apparatus for sentence association according to an embodiment of the present application;
fig. 5 is a block diagram of a server according to an embodiment of the present application.
Detailed Description
Embodiments of the present application are described below with reference to the accompanying drawings.
At present, the model used for determining the sentence association candidate is a high-frequency sentence model obtained by training based on high-frequency sentence combinations in the historical corpus of the user, so that the sentence association candidate of the lower sentence corresponding to the upper content can be determined only for the sentence combinations which are frequently input, and once the upper content does not appear in the high-frequency sentence combinations, the sentence association candidate of the lower sentence corresponding to the upper content is difficult to determine by utilizing the high-frequency sentence model, so that the application range of the sentence association function is not large.
In order to solve the above problems, the embodiments of the present application provide a sentence associating method, which does not rely on a model word library of a local high-frequency sentence model any more, but performs network search with the on-screen context as a keyword to obtain a corresponding search result, because the requirement of occupying and storing on the network is far less strict than the requirement of occupying and storing the local model word library, the corpus on the network related to the context is far more than the model word library of the local high-frequency sentence model, and not only includes the corpus corresponding to the high-frequency sentence combination, so that it is more likely to determine, according to the search result, the context sentences corresponding to various contexts (for example, the context in the high-frequency sentence combination and the context in the non-high-frequency sentence combination), and further generate sentence associating candidates corresponding to the context according to the determined context sentences, thereby improving the application range of the sentence associating function.
Next, application scenarios of the embodiments of the present application will be described with reference to the accompanying drawings. Referring to fig. 1, the application scenario may include a terminal device 101, where the terminal device 101 executes the sentence association method provided in the embodiment of the present application. The terminal device 101 may be a smart phone, a computer, a personal digital assistant (Personal Digital Assistant, PDA), a tablet computer, etc.
Of course, in some cases, in addition to the terminal device 101, a server may be included in the application scenario, and the server may obtain the context from the terminal device 101 that has been on-screen, so that the server determines sentence association candidates, and displays the sentence association candidates on the terminal device 101. The server may be an independent server or a cluster server.
For convenience of description, description will be made later on with the terminal device 101 executing the sentence associating method provided in the embodiment of the present application.
The terminal device 101 may obtain the already-on-screen content in an input interface, where the content may refer to the content to be determined that may continue to screen the content. The content is the last content on the screen in the input interface, and the embodiment of the application is not limited to possible compositions of the content, for example, the content can be sentences, characters or words. The input interface may refer to an interface on the terminal device for inputting and presenting the on-screen content and the following sentences.
It should be noted that, in this embodiment, the screen-on manner of the above content may include various manners. The following description will take a common screen-on manner as an example.
The first screen-up mode may be a screen-up by a user using the terminal device 101.
In the present on-screen manner, the input interface may be any possible input interface, for example, may be a local document, or may be a session window for interaction with other users.
For example, a scene in which the user may continue to input content is determined for the user input content using the terminal device 101. If the user using the terminal device 101 is the first user, the first user wants to input a sentence of lyrics such as "out of the long pavilion, ancient road side, and arrears" to express moods, the first user first inputs the sentence of lyrics such as "out of the long pavilion, ancient road side" through the input method, and then, the terminal device 101 is expected to display the content which the first user may continue to input to the first user, so that the first user can select the sentence of lyrics such as "arrears" from the sentence of lyrics to directly input the sentence of the arrears, and convenience is provided for the input of the first user. At this point, the content is on-screen for the first user.
The second approach to screen may be that of a user in an interaction scenario, which interacts with the user using the terminal device 101.
For example, the first user chat with the second user, and the input interface includes the content on the screen of the session related user, and determines, for the content on the screen last time, the content that the user using the terminal device 101 may continue to input. If the user using the terminal device 101 is the first user, the second user wants to ask the first user who is somehow about, "then the second user inputs who is somehow about? And is displayed on the terminal device 101. The first user is on "who is somehow? When the user answers, after the first user starts the input method, the terminal device 101 can display the content which the first user may continue to input to the first user, so that the first user can select some related content from the content, for example, "some content is a singer and a musician of the taiwan running music, and the first user can directly screen the content, thereby providing convenience for the first user to input. At this point, the content is on screen for the second user.
The third approach to screen may be user-screen using terminal device 101 and user-screen interacting with a user using terminal device 101.
For example, the first user chat with the second user, and the input interface includes the content on the screen of the session related user, and determines, for the content on the screen last time, the content that the user using the terminal device 101 may continue to input. If the user using the terminal device 101 is the first user who says "me likes about somewhere" and the second user asks the first user who he is, then "me likes about somewhere" entered by the first user and "who he is? "on screen on terminal device 101. The first user is on "who is he? When the user answers, after the first user starts the input method, the terminal device 101 can display the content which the first user may continue to input to the first user, so that the first user can select some related content from the content, for example, "some content is a singer and a musician of the taiwan running music, and the first user can directly screen the content, thereby providing convenience for the first user to input. Due to "who is he? "does not have complete semantics, based only on" who is he? "search is difficult to obtain search results, so it is also necessary to determine that the content with complete semantics is actually" who is something around? "to perform a search. At this point, the content is on-screen for the first user and the second user.
The terminal device 101 searches for the above content as a keyword to obtain a search result, where the search result refers to a corpus related to the above content acquired by the terminal device 101 from the network 102. Since the corpus related to the above content is far more than the model word stock of the local high-frequency sentence model on the network 102, the following sentence corresponding to the above content, which is a sentence indicating that the following sentence is present and adjacent to the above content, can be determined according to the search result.
The terminal device 101 may generate sentence association candidates corresponding to the above content according to the above sentence, where the sentence association candidates may be the determined content that may continue to be displayed after the above content. In this way, the terminal device 101 can display sentence association candidates to the user using the terminal device 101, and the user using the terminal device 101 can directly screen a text sentence corresponding to the context by selecting the sentence association candidates, thereby providing convenience for the user to input.
Next, a sentence associating method will be described for embodiments of the present application with reference to the accompanying drawings. Referring to fig. 2, the method includes:
s201, acquiring the on-screen content.
In this embodiment, the first user is introduced by the user using the terminal device. Under different application scenes, the conditions for triggering the terminal equipment to acquire the content are different.
For example, in the first screen-on manner described above, after the first user finishes the screen-on of the content, the terminal device may be triggered to acquire the content.
In another example, in the second screen mode and the third screen mode described above, when the first user clicks the input box to call the keyboard, the terminal device may be triggered to obtain the above content.
The manner in which the terminal device determines the above content may include a plurality of ways, and in one possible implementation, the above content may be content that the terminal device intelligently determines to contain complete semantics.
S202, searching by taking the above content as a keyword to obtain a search result.
Since the corpus related to the above content is far more than the model word stock of the local high-frequency sentence model on the network, in order to ensure that the following sentences corresponding to the above content are more likely to be determined, the search result can be obtained from the network.
It should be noted that many search resources may be included in the network, where different search resources have different degrees of correlation with the above content, some have a large degree of correlation with the above content, and some have a small degree of correlation with the above content. When the degree of correlation of the search resource with the above content is greater than a certain threshold, the search resource may be considered to be correlated with the above content, otherwise the search resource is not correlated with the above content. Thus, in one possible implementation, the above content may be used as a keyword to search in a search resource associated with the above content to obtain search results. Because the search resource is related to the above content, the obtained search result is a search result with a relatively high degree of relevance to the above content, so that the subsequently determined text sentences and the sentence association candidates determined according to the text sentences are also relatively high in relevance to the above content, the possibility that the sentence association candidates comprise the text sentences continuously input by the first user is relatively high, and the first user is more likely to select the text sentences continuously input from the sentence association candidates. In addition, as the search results are obtained by searching only from the search resources related to the content, the search results do not need to be obtained by searching from the search resources not related to the content, the number of the search results is greatly reduced, the processing load of terminal equipment for processing the search results is reduced, and the processing efficiency is improved.
In one possible implementation, the search resources related to the above content may be determined by the type of the above content, for example, the above content is "out of a kiosk, a ancient way edge", the type of the above content is a song class, and then the search resources belonging to the song class are related to the above content to a greater extent, possibly greater than a first threshold, and the search resources belonging to the song class may be used as the search resources related to the above content. Thus, when searching the above content, the search can be performed only in the search resources of the song class to obtain the search result.
S203, determining the following sentences corresponding to the above content according to the search result.
The search result is a corpus related to the context, and may include a context sentence corresponding to the context, so that the context sentence corresponding to the context may be determined according to the search result.
S204, generating sentence association candidates corresponding to the above content according to the context sentence.
The determined context sentence is a sentence that the first user may continue to input after the content above, and the generated sentence association candidate thereof may be presented to the first user so that the first user may screen the context sentence by selecting the sentence association candidate.
According to the technical scheme, according to the above content which is already displayed, the above content can be used as a keyword to perform network search to obtain a corresponding search result, and because the corpus related to the above content on the network is far more than a model word stock of a local high-frequency sentence model, the text sentences corresponding to the above content are more likely to be determined according to the search result, sentence association candidates corresponding to the above content are generated according to the determined text sentences, so that the function of providing convenience for the user to input is achieved, the sentence association function is realized in a searching mode, and the application range of the sentence association function is improved.
It should be noted that, some triggering conditions may exist in the sentence association method provided in the embodiments of the present application. In some cases, since the context may be in a high frequency sentence combination in the history data of the first user, it is also possible to be in a non-high frequency sentence combination. For the above content in the high-frequency sentence combination, sentence association candidates can be accurately generated by the conventional sentence association method, and the subsequent processing of S202-S204 is not required to be performed. Therefore, in order to avoid executing the subsequent processing procedures of S202-S204 on the above content in the high-frequency sentence combination, the trigger of S202-S204 is reduced, the processing burden of the terminal device is reduced, the generating efficiency of the sentence association candidate item is improved, before executing S202, the local high-frequency sentence model can be utilized to determine whether the above content has a corresponding following sentence, if so, the local high-frequency sentence model is utilized to directly determine the following sentence corresponding to the above content, the processing burden of the terminal device is reduced, and the generating efficiency of the sentence association candidate item is improved. If not, the step of S202 is performed, ensuring that sentence association candidates can be generated for the above content that is not applicable in the conventional manner.
In other cases, since the types of the above contents may include many kinds, some types of the above contents are suitable for determining sentence association candidates by the sentence association method provided by the embodiment of the present application, and some types of the above contents may not be suitable for determining sentence association candidates by the sentence association method provided by the embodiment of the present application. Therefore, in order to ensure that the subsequent processing procedures of S202-S204 are performed only on the above content applicable to the sentence association method, unnecessary triggers of S202-S204 are reduced, processing load of the terminal device is reduced, it may be first determined whether the type of the above content is a preset type or belongs to a preset domain before performing S202, and if the type of the above content is a preset type or belongs to a preset domain, for example, the type of the above content belongs to a movie domain, a television play domain, a sports domain, or the type of the above content is a star character class, a poetry class, a stop-and-go class, etc., it is explained that the above content is applicable to the sentence association method provided in the embodiment of the present application, the step of S202 is performed. Note that, in executing S203, the manner of determining the following sentence may include various ways. Next, a description will be given of a determination manner of the following sentence.
In one possible implementation, the determining of the context sentence may be determining the context sentence according to a matching relationship, which may be a probability for identifying that a content appears after the content.
In some cases, the context and the context sentence may constitute a fixed form of context, such as poetry, lyrics, a adam, colloquial, etc., at which time it may be determined which part of the search result may be the context sentence based on the matching relationship of the context and the various parts of the search result.
Since the resulting search results may include the above content, the following sentences, and other content, separators may exist between the above content, the following sentences, and the other content, and the separators may include punctuations, spaces, and the like, which are symbols for separating sentences.
Dividing the search result into a plurality of parts according to the separator in the search result, wherein each part is used as a division result, and each division result can be a sentence (can comprise a phrase or a long sentence) because the separator is used as a separation basis; a first matching relationship between the segmentation result and the content above is determined, wherein the first matching relationship is used to identify a probability that the segmentation result occurs after the content above. If the first matching relationship satisfies the first preset condition, the probability that the segmentation result appears after the content above is considered to be relatively large, so that the segmentation result that the first matching relationship satisfies the first preset condition can be determined as a text sentence corresponding to the content above.
In one possible implementation, the determination of the context sentence may be based on text content similar to the context content.
In such an implementation, the content may be irregular in expression, too spoken, etc., so that it may be difficult to determine the following sentence using the first matching relationship obtained by matching the content itself with the segmentation result in the search result. And text content with a certain semantic similarity with the above content exists in the search result, and the higher the similarity is, the more likely the text content is to be the text content.
In this case, the text content whose similarity to the above content satisfies the second preset condition may be first determined from the search result, and the similarity to the above content may be considered as being higher, and the likelihood that the following sentence of the text content is the following sentence of the above content may be higher, and therefore, the following sentence of the text content in the search result may be regarded as the following sentence to which the above content corresponds.
Therefore, the method can ensure that the lower Wen Gouzi is determined for the text content with nonstandard expression and excessive colloquial expression by determining the similar text content for the text content, so that the corresponding sentence association candidate items are determined, and the application range of the sentence association function is improved.
In one possible implementation, the following sentences may be determined from the summary content of the search results.
The types of the above contents may include many, in some cases, the types of the above contents may be question types, and accordingly, the search result may be an answer to the above contents, the search result includes a related introduction of the object mentioned in the above contents, and the search result includes a large amount of information of the object, however, the following sentence is an input interface for generating sentence association candidates so that the first user selects a satisfactory following sentence on-screen, and generally, the following sentence is not excessively long and only needs to include key information for the object in the search result. Since the summary content in the search result can reflect the key information for the object, if the type of the above content is judged to be the question type, the summary content of the search result can be obtained according to the above content, and the summary content is used as the following sentence corresponding to the above content.
For example, the second user inputs "who is somehow sometime around? "and screen on the terminal device corresponding to the first user," who is somehow? "as the above content, something is an object included in the above content. The search result obtained by the terminal device executing S202 may include a lot of information in the search encyclopedia, and it is difficult to use all the information in the search encyclopedia as the following sentences, so that the summary content of the search result may be determined according to the above content, for example, "the week is a singer and a musician of the taiwan popular music men".
It can be seen that by determining the summary content to be a lower sentence, the determined lower sentence can be made more brief, so that the first user is more likely to select a satisfactory lower sentence from sentence association candidates.
After S203 is executed, the determined following sentences may include one or more sentences. The number of sentence association candidates presented to the first user by the terminal device is limited due to the size limitation of the input interface and due to the efficiency of selecting sentence association candidates. Therefore, when a plurality of the context sentences are determined, a context sentence more likely to be the first user to continue inputting the content can be selected therefrom as the sentence association candidate. The possibility that the following sentence is used as the first user to continue inputting the content can be embodied by the sorting value corresponding to the following sentence, and the larger the sorting value is, the greater the possibility that the following sentence is used as the first user to continue inputting the content is.
In this case, one possible implementation manner of S204 is to determine ranking values corresponding to the plurality of context sentences respectively, and generate sentence association candidates corresponding to the context according to the context sentences whose ranking values satisfy the threshold. Generating sentence association candidates according to the context sentences with larger ranking values can further improve the possibility that the first user determines the context sentences by selecting the sentence association candidates.
Wherein the ranking values respectively corresponding to the plurality of following sentences are determined according to the following manner.
The first determining mode is to determine second matching relations between the plurality of text sentences and the above content respectively, wherein the second matching relations are used for identifying the probability that the text sentences appear behind the above content, and the larger the probability is, the more likely the text sentences are to be sentences which are continuously input by the first user, and the sorting values corresponding to the plurality of text sentences are determined according to the second matching relations.
In addition, when the search result is obtained through S202, the association degrees between different search results and the above content are different, and the greater the association degree is, the more accurate the following sentence determined according to the search result is, and the greater the likelihood that the first user continues to input the following sentence is. Therefore, the second determining manner is to determine ranking values corresponding to the plurality of context sentences according to the association degrees of the search results of the plurality of context sentences and the context content.
Of course, in order to improve accuracy of determining the ranking value, the first determining manner and the second determining manner may be combined to determine ranking values corresponding to the plurality of context sentences together.
Based on the corresponding embodiment of fig. 2, the present embodiment provides a sentence associating apparatus, referring to fig. 3, which includes an obtaining unit 301, a searching unit 302, a determining unit 303, and a generating unit 304:
The acquiring unit 301 is configured to acquire the on-screen content;
the searching unit 302 is configured to search for the above content as a keyword to obtain a search result;
the determining unit 303 is configured to determine a context sentence corresponding to the context content according to the search result;
the generating unit 304 is configured to generate sentence association candidates corresponding to the context according to the context sentence.
Optionally, the determining unit is specifically configured to:
dividing the search result into a plurality of divided results according to separators in the search result;
determining a first matching relation between the segmentation result and the above content, wherein the first matching relation is used for identifying the probability that the segmentation result appears behind the above content;
and determining the segmentation result of the first matching relation meeting a first preset condition as a text sentence corresponding to the text content.
Optionally, the determining unit is specifically configured to:
judging whether the type of the above content is a question type;
if yes, obtaining abstract content of the search result according to the above content;
and taking the abstract content as a corresponding following sentence of the above content.
Optionally, the determining unit is specifically configured to:
determining text content with similarity meeting a second preset condition from the search result;
and taking the text content in the search result as the corresponding text sentence of the text content.
Optionally, if a plurality of context sentences corresponding to the above content are determined according to the search result, the generating unit is specifically configured to:
determining the sorting values corresponding to the plurality of context sentences respectively;
generating sentence association candidates corresponding to the above content according to the context sentences of which the ranking values meet the threshold values;
the ranking values respectively corresponding to the plurality of following sentences are determined according to any one or a combination of the following determination modes:
first determination mode:
determining a second matching relationship between the plurality of context sentences and the context content respectively, wherein the second matching relationship is used for identifying the probability that the context sentences appear behind the context content;
determining the sorting values respectively corresponding to the plurality of context sentences according to the second matching relation;
the second determination mode:
and determining the ranking values corresponding to the plurality of the context sentences according to the association degree of the search results of the plurality of the context sentences and the context content.
Optionally, the search unit is specifically configured to:
and taking the above content as a keyword, and searching in search resources related to the type of the above content to obtain the search result.
Optionally, the device further includes a judging unit:
the judging unit is used for judging whether the type of the content is a preset type or belongs to a preset field;
and if the judging unit judges that the type of the above content is a preset type or belongs to a preset field, triggering the searching unit to execute the step of searching by taking the above content as a keyword to obtain a search result.
Optionally, if the display object of the sentence association candidate item is a first user, the above content is on-screen for the first user, and/or the above content is on-screen for a second user.
The present embodiment also provides an apparatus for sentence association, which may be a terminal apparatus, and fig. 4 is a block diagram of a terminal apparatus 400 according to an exemplary embodiment. For example, the terminal device 400 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, an exercise device, a personal digital assistant, and the like.
Referring to fig. 4, a terminal device 400 may include one or more of the following components: a processing component 402, a memory 404, a power supply component 406, a multimedia component 408, an audio component 410, an input/output (I/O) interface 412, a sensor component 414, and a communication component 416.
The processing component 402 generally controls overall operation of the terminal device 400, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing element 402 may include one or more processors 420 to execute instructions to perform all or part of the steps of the methods described above. Further, the processing component 402 can include one or more modules that facilitate interaction between the processing component 402 and other components. For example, the processing component 402 may include a multimedia module to facilitate interaction between the multimedia component 408 and the processing component 402.
The memory 404 is configured to store various types of data to support operations at the terminal device 400. Examples of such data include instructions for any application or method operating on the apparatus 400, contact data, phonebook data, messages, pictures, videos, and the like. The memory 404 may be implemented by any type or combination of volatile or nonvolatile memory devices 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 or optical disk.
The power supply component 406 provides power to the various components of the terminal device 400. The power supply components 406 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the apparatus 400.
The multimedia component 408 comprises a screen between the terminal device 400 and the user providing an output interface. 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 a user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensor may sense not only the boundary of a touch or slide action, but also the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 408 includes a front camera and/or a rear camera. The front camera and/or the rear camera may receive external multimedia data when the terminal device 400 is in an operation mode, such as a photographing mode or a video mode. Each front camera and rear camera may be a fixed optical lens system or have focal length and optical zoom capabilities.
The audio component 410 is configured to output and/or input audio signals. For example, the audio component 410 includes a Microphone (MIC) configured to receive external audio signals when the apparatus 400 is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may be further stored in the memory 404 or transmitted via the communication component 416. In some embodiments, audio component 410 further includes a speaker for outputting audio signals.
The I/O interface 412 provides an interface between the processing component 402 and peripheral interface modules, which may be a keyboard, click wheel, buttons, etc. These buttons may include, but are not limited to: homepage button, volume button, start button, and lock button.
The sensor assembly 414 includes one or more sensors for providing status assessment of various aspects of the terminal device 400. For example, the sensor assembly 414 may detect the on/off state of the terminal device 400, the relative positioning of the components, such as the display and keypad of the terminal device 400, the sensor assembly 414 may also detect the change in position of the terminal device 400 or a component of the terminal device 400, the presence or absence of user contact with the terminal device 400, the orientation or acceleration/deceleration of the terminal device 400, and the change in temperature of the terminal device 400. The sensor assembly 414 may include a proximity sensor configured to detect the presence of nearby objects in the absence of any physical contact. The sensor assembly 414 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 414 may also include an acceleration sensor, a gyroscopic sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.
The communication component 416 is configured to facilitate communication between the terminal device 400 and other devices, either wired or wireless. The terminal device 400 may access a wireless network based on a communication standard, such as WiFi,2G or 3G, or a combination thereof. In one exemplary embodiment, the communication part 416 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 416 further includes a Near Field Communication (NFC) module to facilitate short range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, ultra Wideband (UWB) technology, bluetooth (BT) technology, and other technologies.
In an exemplary embodiment, the terminal device 400 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 elements for executing the methods described above.
In an exemplary embodiment, a non-transitory computer readable storage medium is also provided, such as memory 404, including instructions executable by processor 420 of terminal device 400 to perform the above-described method. For example, the non-transitory computer readable storage medium may be ROM, random Access Memory (RAM), CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.
A non-transitory computer readable storage medium, which when executed by a processor of a mobile terminal, causes the mobile terminal to perform a sentence association method, the method comprising:
acquiring the on-screen content;
searching the above content as a keyword to obtain a search result;
determining a lower sentence corresponding to the upper content according to the search result;
and generating sentence association candidates corresponding to the above content according to the context sentence.
The device for sentence association provided in this embodiment may also be a server, and fig. 5 is a schematic structural diagram of the server in this embodiment of the present invention. The server 500 may vary considerably in configuration or performance and may include one or more central processing units (central processing units, CPU) 522 (e.g., one or more processors) and memory 532, one or more storage media 530 (e.g., one or more mass storage devices) storing applications 542 or data 544. Wherein memory 532 and storage medium 530 may be transitory or persistent. The program stored in the storage medium 530 may include one or more modules (not shown), each of which may include a series of instruction operations on a server. Still further, the central processor 522 may be configured to communicate with a storage medium 530 and execute a series of instruction operations in the storage medium 530 on the server 500.
The server 500 may also include one or more power supplies 526, one or more wired or wireless network interfaces 550, one or more input/output interfaces 558, one or more keyboards 556, and/or one or more operating systems 541, such as Windows ServerTM, mac OS XTM, unixTM, linuxTM, freeBSDTM, etc.
Those of ordinary skill in the art will appreciate that: all or part of the steps for implementing the above method embodiments may be implemented by hardware related to program instructions, where the above program may be stored in a computer readable storage medium, and when the program is executed, the program performs steps including the above method embodiments; and the aforementioned storage medium may be at least one of the following media: read-only memory (ROM), RAM, magnetic disk or optical disk, etc., which can store program codes.
It should be noted that, in the present specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment is mainly described in a different point from other embodiments. In particular, for the apparatus and system embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, with reference to the description of the method embodiments in part. The apparatus and system embodiments described above are merely illustrative, in which elements illustrated as separate elements may or may not be physically separate, and elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
The foregoing is merely one specific embodiment of the present application, but the protection scope of the present application is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present application should be covered in the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (18)
1. A sentence association method for use in determining, for user input content, a scenario in which the user is likely to continue to input content, the method comprising:
acquiring the on-screen content;
determining whether the above content has a corresponding below sentence using a local high frequency sentence module;
if yes, determining a lower sentence corresponding to the upper content by using a local high-frequency sentence module;
if not, searching by taking the above content as a keyword to obtain a search result;
determining a text sentence corresponding to the text content according to the search result, wherein the text sentence is a sentence indicating that the text content is followed by and adjacent to the text content, and the determined text sentence is a sentence which can be continuously input after the text content;
And generating sentence association candidates corresponding to the above content according to the context sentence.
2. The method of claim 1, wherein the determining the corresponding context sentence of the context content from the search result comprises:
dividing the search result into a plurality of divided results according to separators in the search result;
determining a first matching relation between the segmentation result and the above content, wherein the first matching relation is used for identifying the probability that the segmentation result appears behind the above content;
and determining the segmentation result of the first matching relation meeting a first preset condition as a text sentence corresponding to the text content.
3. The method of claim 1, wherein the determining the corresponding context sentence of the context content from the search result comprises:
judging whether the type of the above content is a question type;
if yes, obtaining abstract content of the search result according to the above content;
and taking the abstract content as a corresponding following sentence of the above content.
4. The method of claim 1, wherein the determining the corresponding context sentence of the context content from the search result comprises:
Determining text content with similarity meeting a second preset condition from the search result;
and taking the text content in the search result as the corresponding text sentence of the text content.
5. The method according to any one of claims 1-4, wherein if a plurality of context sentences corresponding to the context are determined according to the search result, the generating sentence association candidates corresponding to the context according to the context sentences includes:
determining the sorting values corresponding to the plurality of context sentences respectively;
generating sentence association candidates corresponding to the above content according to the context sentences of which the ranking values meet the threshold values;
the ranking values respectively corresponding to the plurality of following sentences are determined according to any one or a combination of the following determination modes:
first determination mode:
determining a second matching relationship between the plurality of context sentences and the context content respectively, wherein the second matching relationship is used for identifying the probability that the context sentences appear behind the context content;
determining the sorting values respectively corresponding to the plurality of context sentences according to the second matching relation;
the second determination mode:
And determining the ranking values corresponding to the plurality of the context sentences according to the association degree of the search results of the plurality of the context sentences and the context content.
6. The method of claim 1, wherein searching the content as a keyword to obtain a search result comprises:
and taking the above content as a keyword, and searching in search resources related to the type of the above content to obtain the search result.
7. The method of claim 1, wherein prior to searching for the contextual content as a keyword, the method further comprises:
judging whether the type of the content is a preset type or belongs to a preset field;
if yes, executing the step of searching by taking the above content as a keyword to obtain a search result.
8. The method of claim 1, wherein if the presentation object of the sentence association candidate is a first user, the context is on-screen for the first user and/or the context is on-screen for a second user.
9. Sentence associating apparatus, applied to a scene in which user input content is determined to be likely to continue to be input by the user, comprising an acquisition unit, a search unit, a determination unit, and a generation unit:
The acquisition unit is used for acquiring the on-screen content;
the searching unit is used for determining whether the above content has a corresponding text sentence or not by utilizing a local high-frequency sentence module; if yes, determining a lower sentence corresponding to the upper content by using a local high-frequency sentence module; if not, searching by taking the above content as a keyword to obtain a search result;
the determining unit is configured to determine a following sentence corresponding to the above content according to the search result, where the following sentence is a sentence indicating that the following sentence is present after the above content and is adjacent to the above content, and the determined following sentence is a sentence that may be input continuously after the above content;
the generating unit is used for generating sentence association candidates corresponding to the context according to the context sentence.
10. The apparatus according to claim 9, wherein the determining unit is specifically configured to:
dividing the search result into a plurality of divided results according to separators in the search result;
determining a first matching relation between the segmentation result and the above content, wherein the first matching relation is used for identifying the probability that the segmentation result appears behind the above content;
And determining the segmentation result of the first matching relation meeting a first preset condition as a text sentence corresponding to the text content.
11. The apparatus according to claim 9, wherein the determining unit is specifically configured to:
judging whether the type of the above content is a question type;
if yes, obtaining abstract content of the search result according to the above content;
and taking the abstract content as a corresponding following sentence of the above content.
12. The apparatus according to claim 9, wherein the determining unit is specifically configured to:
determining text content with similarity meeting a second preset condition from the search result;
and taking the text content in the search result as the corresponding text sentence of the text content.
13. The apparatus according to any one of claims 9-12, wherein if a plurality of context sentences corresponding to the context content are determined according to the search result, the generating unit is specifically configured to:
determining the sorting values corresponding to the plurality of context sentences respectively;
generating sentence association candidates corresponding to the above content according to the context sentences of which the ranking values meet the threshold values;
The ranking values respectively corresponding to the plurality of following sentences are determined according to any one or a combination of the following determination modes:
first determination mode:
determining a second matching relationship between the plurality of context sentences and the context content respectively, wherein the second matching relationship is used for identifying the probability that the context sentences appear behind the context content;
determining the sorting values respectively corresponding to the plurality of context sentences according to the second matching relation;
the second determination mode:
and determining the ranking values corresponding to the plurality of the context sentences according to the association degree of the search results of the plurality of the context sentences and the context content.
14. The apparatus according to claim 9, wherein the search unit is specifically configured to:
and taking the above content as a keyword, and searching in search resources related to the type of the above content to obtain the search result.
15. The apparatus according to claim 9, wherein the apparatus further comprises a judging unit:
the judging unit is used for judging whether the type of the content is a preset type or belongs to a preset field;
and if the judging unit judges that the type of the above content is a preset type or belongs to a preset field, triggering the searching unit to execute the step of searching by taking the above content as a keyword to obtain a search result.
16. The apparatus of claim 9, wherein if the presentation object of the sentence association candidate is a first user, the context is on-screen for the first user and/or the context is on-screen for a second user.
17. An apparatus for sentence association comprising a memory and one or more programs for determining, for user input content, a scenario in which the user is likely to continue to input content, wherein the one or more programs are stored in the memory and configured to be executed by one or more processors, the one or more programs comprising instructions for:
acquiring the on-screen content;
determining whether the above content has a corresponding below sentence using a local high frequency sentence module;
if yes, determining a lower sentence corresponding to the upper content by using a local high-frequency sentence module;
if not, searching by taking the above content as a keyword to obtain a search result;
determining a text sentence corresponding to the text content according to the search result, wherein the text sentence is a sentence indicating that the text content is followed by and adjacent to the text content, and the determined text sentence is a sentence which can be continuously input after the text content;
And generating sentence association candidates corresponding to the above content according to the context sentence.
18. A machine readable medium having instructions stored thereon, which when executed by one or more processors, cause an apparatus to perform the sentence association method of one or more of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811639917.0A CN111381685B (en) | 2018-12-29 | 2018-12-29 | Sentence association method and sentence association device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811639917.0A CN111381685B (en) | 2018-12-29 | 2018-12-29 | Sentence association method and sentence association device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111381685A CN111381685A (en) | 2020-07-07 |
| CN111381685B true CN111381685B (en) | 2024-03-22 |
Family
ID=71216751
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811639917.0A Active CN111381685B (en) | 2018-12-29 | 2018-12-29 | Sentence association method and sentence association device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111381685B (en) |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1734445A (en) * | 2004-07-26 | 2006-02-15 | 索尼株式会社 | Method, apparatus, and program for dialogue, and storage medium including a program stored therein |
| JP2011248499A (en) * | 2010-05-25 | 2011-12-08 | Nec Corp | Character input device, character input control method, and character input control program |
| CN102629160A (en) * | 2012-03-16 | 2012-08-08 | 华为终端有限公司 | Input method, input device and terminal |
| EP2778861A1 (en) * | 2013-03-15 | 2014-09-17 | BlackBerry Limited | System and method for predictive text input |
| CN104268166A (en) * | 2014-09-09 | 2015-01-07 | 北京搜狗科技发展有限公司 | Input method, device and electronic device |
| CN104281649A (en) * | 2014-09-09 | 2015-01-14 | 北京搜狗科技发展有限公司 | Input method and device and electronic equipment |
| CN104484057A (en) * | 2014-12-04 | 2015-04-01 | 百度在线网络技术（北京）有限公司 | Associative result providing method and device |
| CN104699836A (en) * | 2015-03-31 | 2015-06-10 | 北京奇虎科技有限公司 | Multi-keyword search prompting method and multi-keyword search prompting device |
| CN104899322A (en) * | 2015-06-18 | 2015-09-09 | 百度在线网络技术（北京）有限公司 | Search engine and implementation method thereof |
| CN104917669A (en) * | 2015-05-29 | 2015-09-16 | 北京搜狗科技发展有限公司 | Information processing method and device |
| CN105069037A (en) * | 2015-07-22 | 2015-11-18 | 无锡天脉聚源传媒科技有限公司 | Search content display method and apparatus |
| CN105183848A (en) * | 2015-09-07 | 2015-12-23 | 百度在线网络技术（北京）有限公司 | Human-computer chatting method and device based on artificial intelligence |
| CN105912712A (en) * | 2016-04-29 | 2016-08-31 | 华南师范大学 | Big data-based robot conversation control method and system |
| CN106774970A (en) * | 2015-11-24 | 2017-05-31 | 北京搜狗科技发展有限公司 | The method and apparatus being ranked up to the candidate item of input method |
| CN107621886A (en) * | 2016-07-15 | 2018-01-23 | 北京搜狗科技发展有限公司 | Method, apparatus and electronic equipment are recommended in one kind input |
| CN108932066A (en) * | 2018-06-13 | 2018-12-04 | 北京百度网讯科技有限公司 | Method, apparatus, equipment and the computer storage medium of input method acquisition expression packet |
| CN109074354A (en) * | 2016-07-22 | 2018-12-21 | 华为技术有限公司 | Show the method and terminal device of candidate item |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018101671A1 (en) * | 2016-11-29 | 2018-06-07 | Samsung Electronics Co., Ltd. | Apparatus and method for providing sentence based on user input |
-
2018
- 2018-12-29 CN CN201811639917.0A patent/CN111381685B/en active Active
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1734445A (en) * | 2004-07-26 | 2006-02-15 | 索尼株式会社 | Method, apparatus, and program for dialogue, and storage medium including a program stored therein |
| JP2011248499A (en) * | 2010-05-25 | 2011-12-08 | Nec Corp | Character input device, character input control method, and character input control program |
| CN102629160A (en) * | 2012-03-16 | 2012-08-08 | 华为终端有限公司 | Input method, input device and terminal |
| EP2778861A1 (en) * | 2013-03-15 | 2014-09-17 | BlackBerry Limited | System and method for predictive text input |
| CN104268166A (en) * | 2014-09-09 | 2015-01-07 | 北京搜狗科技发展有限公司 | Input method, device and electronic device |
| CN104281649A (en) * | 2014-09-09 | 2015-01-14 | 北京搜狗科技发展有限公司 | Input method and device and electronic equipment |
| CN104484057A (en) * | 2014-12-04 | 2015-04-01 | 百度在线网络技术（北京）有限公司 | Associative result providing method and device |
| CN104699836A (en) * | 2015-03-31 | 2015-06-10 | 北京奇虎科技有限公司 | Multi-keyword search prompting method and multi-keyword search prompting device |
| CN104917669A (en) * | 2015-05-29 | 2015-09-16 | 北京搜狗科技发展有限公司 | Information processing method and device |
| CN104899322A (en) * | 2015-06-18 | 2015-09-09 | 百度在线网络技术（北京）有限公司 | Search engine and implementation method thereof |
| CN105069037A (en) * | 2015-07-22 | 2015-11-18 | 无锡天脉聚源传媒科技有限公司 | Search content display method and apparatus |
| CN105183848A (en) * | 2015-09-07 | 2015-12-23 | 百度在线网络技术（北京）有限公司 | Human-computer chatting method and device based on artificial intelligence |
| CN106774970A (en) * | 2015-11-24 | 2017-05-31 | 北京搜狗科技发展有限公司 | The method and apparatus being ranked up to the candidate item of input method |
| CN105912712A (en) * | 2016-04-29 | 2016-08-31 | 华南师范大学 | Big data-based robot conversation control method and system |
| CN107621886A (en) * | 2016-07-15 | 2018-01-23 | 北京搜狗科技发展有限公司 | Method, apparatus and electronic equipment are recommended in one kind input |
| CN109074354A (en) * | 2016-07-22 | 2018-12-21 | 华为技术有限公司 | Show the method and terminal device of candidate item |
| CN108932066A (en) * | 2018-06-13 | 2018-12-04 | 北京百度网讯科技有限公司 | Method, apparatus, equipment and the computer storage medium of input method acquisition expression packet |
Non-Patent Citations (2)
| Title |
|---|
| Deep Sentence Embedding Using Long Short-Term Memory Networks: Analysis and Application to Information Retrieval;Hamid Palangi et al;《IEEE/ACM TRANSACTIONS ON AUDIO, SPEECH, AND LANGUAGE PROCESSING》;20160430;第24卷(第4期);694-707 * |
| 面向写作辅助的中文智能输入法系统;代贤俊;《中国优秀硕士学位论文全文数据库信息科技辑》;20150215;第2015年卷(第2期);I138-1512 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111381685A (en) | 2020-07-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107527619B (en) | Method and device for positioning voice control service | |
| CN107621886B (en) | Input recommendation method and device and electronic equipment | |
| CN108073606B (en) | News recommendation method and device for news recommendation | |
| CN108304412B (en) | Cross-language search method and device for cross-language search | |
| CN108073292B (en) | Intelligent word forming method and device for intelligent word forming | |
| RU2733816C1 (en) | Method of processing voice information, apparatus and storage medium | |
| CN111708943A (en) | Search result display method and device and search result display device | |
| CN111831806A (en) | Semantic integrity determination method and device, electronic equipment and storage medium | |
| CN107424612B (en) | Processing method, apparatus and machine-readable medium | |
| CN107291260B (en) | Information input method and device for inputting information | |
| CN112307281A (en) | Entity recommendation method and device | |
| CN109521888B (en) | Input method, device and medium | |
| CN113987128A (en) | Related article searching method and device, electronic equipment and storage medium | |
| CN108073294B (en) | Intelligent word forming method and device for intelligent word forming | |
| CN108108356B (en) | Character translation method, device and equipment | |
| CN111381685B (en) | Sentence association method and sentence association device | |
| CN109917927B (en) | Candidate item determination method and device | |
| CN109426359B (en) | Input method, device and machine readable medium | |
| CN113589949A (en) | Input method and device and electronic equipment | |
| CN112052395A (en) | Data processing method and device | |
| CN108073566B (en) | Word segmentation method and device and word segmentation device | |
| CN112181163A (en) | Input method, input device and input device | |
| CN110929122A (en) | Data processing method and device and data processing device | |
| CN110858100B (en) | Method and device for generating association candidate words | |
| CN110716653B (en) | Method and device for determining association source |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |