CN111930904A - Information response method, device, equipment and storage medium - Google Patents
Information response method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN111930904A CN111930904A CN202010651380.0A CN202010651380A CN111930904A CN 111930904 A CN111930904 A CN 111930904A CN 202010651380 A CN202010651380 A CN 202010651380A CN 111930904 A CN111930904 A CN 111930904A
- Authority
- CN
- China
- Prior art keywords
- information
- subset
- output information
- client
- generating
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/338—Presentation of query results
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
- G06F40/35—Discourse or dialogue representation
Abstract
The embodiment of the application provides an information response method, an information response device, information response equipment and a storage medium, wherein the method comprises the following steps: generating an element set at least comprising a first element subset and a second element subset according to the acquired input information; generating first output information according to the first element subset; the first output information is sent to a client side, so that the client side feeds back the first output information; generating second output information at least according to a first feedback result fed back by the client and the second element subset in the element set; sending the second output information to the client to enable the client to feed back the second output information; outputting target output information according to a second feedback result fed back by the client; wherein the target output information matches target intent information determined from the second feedback result.
Description
Technical Field
The present application relates to information processing technologies, and in particular, to an information response method, apparatus, device, and storage medium.
Background
In an intelligent customer service system, it is very difficult to directly and accurately obtain the user intention according to the user input, so that a plurality of answer options can be pushed at one time in most intelligent customer service systems to enable the user to select the answer options by himself, but the pushing of the answer options at one time causes the user to be troubled, and the user can also select wrong answers.
Disclosure of Invention
In view of this, embodiments of the present application provide an information response method, apparatus, device, and storage medium.
The technical scheme of the application is realized as follows:
the embodiment of the application provides an information response method, which comprises the following steps:
generating an element set at least comprising a first element subset and a second element subset according to the acquired input information;
generating first output information according to the first element subset;
sending the first output information to a client so that the client feeds back the first output information;
generating second output information at least according to a first feedback result fed back by the client and the second element subset in the element set;
sending the second output information to the client to enable the client to feed back the second output information;
outputting target output information according to a second feedback result fed back by the client; wherein the target output information matches target intent information determined from the second feedback result.
An embodiment of the present application provides an information response device, the information response device includes:
the first generation module is used for generating an element set at least comprising a first element subset and a second element subset according to the acquired input information;
the second generation module is used for generating first output information according to the first element subset;
the first sending module is used for sending the first output information to a client so that the client feeds back the first output information;
a third generating module, configured to generate second output information according to at least a first feedback result fed back by the client and the second subset of elements in the element set;
the second sending module is used for sending the second output information to the client so that the client feeds back the second output information;
the output module is used for outputting target output information according to a second feedback result fed back by the client; wherein the target output information matches target intent information determined from the second feedback result.
An embodiment of the present application provides an electronic device, which at least includes: a controller and a storage medium configured to store executable instructions, wherein:
the controller is configured to execute stored executable instructions configured to perform the message answering method described above.
Correspondingly, the embodiment of the application provides a computer-readable storage medium, in which computer-executable instructions are stored, and the computer-executable instructions are configured to execute the information response method.
According to the information response method, the device, the equipment and the storage medium, an element set at least comprising two element subsets is determined according to acquired input information, first output information is generated according to a first element subset in the element set and is sent to a client, second output information is generated and is sent to the client at least in combination with elements in a second element subset according to a first feedback result of the client, and then target output information is generated and output according to a second feedback result fed back by the client, so that the element information determined according to the acquired input information can be sequentially sent to a user through a split screen form or a classification form, the intention corresponding to the input information of the user is accurately determined according to the feedback result of the user in sequence, and natural interaction between the information response device and the user is realized, the user's intent is refined step by step with minimal confusion to the user.
Drawings
In the drawings, which are not necessarily drawn to scale, like reference numerals may describe similar components in different views. Like reference numerals having different letter suffixes may represent different examples of similar components. The drawings illustrate generally, by way of example, but not by way of limitation, various embodiments discussed herein.
Fig. 1 is a schematic flow chart illustrating an implementation of an information response method according to an embodiment of the present application;
fig. 2 is a schematic flow chart of another implementation of the information response method according to the embodiment of the present application;
fig. 3 is a schematic flowchart illustrating an implementation of an information response method according to an embodiment of the present application;
fig. 4 is a schematic flowchart of another implementation of the information response method according to the embodiment of the present application;
fig. 5 is a schematic structural diagram of an information response device according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of the electronic device according to the embodiment of the present application.
Detailed Description
In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be described in further detail with reference to the attached drawings, the described embodiments should not be considered as limiting the present application, and all other embodiments obtained by a person of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.
It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
In the following description, suffixes such as "module", "component", or "unit" used to denote elements are used only for the convenience of description of the present application, and have no specific meaning by themselves. Thus, "module", "component" or "unit" may be used mixedly.
The electronic device may be implemented in various forms. For example, the electronic devices described in the present application may include devices such as a mobile phone, a tablet computer, a notebook computer, a palm top computer, a Personal Digital Assistant (PDA), a Portable Media Player (PMP), a desktop computer, a server, and the like.
While the following description will be made taking a mobile electronic device as an example, those skilled in the art will appreciate that the configuration according to the embodiment of the present application can be applied to a fixed type electronic device in addition to elements particularly used for moving purposes.
An information response method is provided in an embodiment of the present application, and fig. 1 is a schematic flow chart illustrating an implementation of the information response method provided in the embodiment of the present application, and as shown in fig. 1, the information response method includes the following steps:
step S101: the information response device generates an element set at least comprising a first element subset and a second element subset according to the acquired input information.
Here, the information answering device may be a computer or a mobile terminal, etc., such as a robot or an intelligent customer service system. The input information may be text information or voice information or a sentence in video input by the user at the client terminal or the information answering device. After the information response device acquires the input information, the semantics of the input information are analyzed based on the natural language understanding model to obtain at least one intention information corresponding to the input information, and then the intention information is subjected to granularity splitting to obtain an element set consisting of at least one element. Intent may refer to a purpose that a user wants to achieve when communicating with an intelligent customer service system. The terminal can split the intention information by adopting a plurality of splitting strategies, such as splitting into binary groups, triple groups or more complex structures, carrying out granularity splitting, determining the constituent elements of each intention information, and obtaining an element set.
In some implementations, when there are multiple elements in the element set, the element set may be divided into multiple element subsets. In some embodiments, the elements may be classified according to part-of-speech categories, such as: nouns, verbs, adverbs, etc.; or dividing the element set according to the semantic attributes of the elements, and dividing the element set into at least a first element subset and a second element subset. The parts of speech or semantic attributes of the elements in each subset of elements are the same. For example, after the element set is divided according to the part-of-speech of the elements, the elements in the first element subset are all nouns, the elements in the second element subset are all verbs, and the elements in the third element subset are all adverbs.
Step S102: and the information response device generates first output information according to the first element subset.
Here, the first output information is generated using a preset template or a natural language generating model according to the element information included in the first element subset. The first output information refers to that element information in the first element subset is combined through reasonable field organizing sequence, proper connection words are added in the determined content, and all related words and determined phrases are combined into natural language text information or natural language voice information with proper structures.
In some embodiments, the first output information is generated directly when there is only one element in the set of elements. When there are a plurality of elements in the element set, first output information including elements in the first element subset is generated according to element information in the element subset classified by the element set. For example, when the elements in the first element subset are short messages, the generated first output message is: do you ask a short message question? When the elements in the first element subset are short messages and address lists, the generated first output information may be: what is you asking is a short message question or an address book question?
Step S103: and the information response device sends the first output information to a client so that the client feeds back the first output information.
Here, the obtained first output information is sent to the client of the user in a text message or voice message manner, and the user can select, reply or respond to the first output information according to the content displayed by the client or the content prompted by the client. And sends the selected or replied or responded information to the information answering device. For example, with the above embodiment, is the first output message a question of a short message or a question of an address book asked? If the information response device sends the first output information in a voice form, the user of the client can feed back in a voice response mode: address book problems; if the information response device sends the first output information in the form of selection information, the user of the client can select one option in the first output information for feedback: address book problem.
In some realizable embodiments, the information response device is an intelligent customer service robot, and after the first output information is generated, the first output information is output in a voice mode or a screen display mode, so that a user interacting with the intelligent customer service robot directly obtains the first output information on the information response device and feeds back the first output information.
Step S104: and the information response device generates second output information at least according to a first feedback result fed back by the client and the second element subset in the element set.
Here, the second subset of elements in the element set is a second subset of elements obtained by dividing the element set. The elements in the second element subset are mostly verbs, or predicates, or actions in semantic attributes. The second output information includes information related to a first feedback result corresponding to the first output information. In some embodiments, when generating the second output information, it is also necessary to combine information of other element subsets in the element set, such as information in a third element subset, where most of elements included in the third element subset are information such as adverbs, predicates, or industries with semantic attributes of elements. In connection with the above embodiment, when the first feedback result is a computer problem, and the elements in the second subset of elements are: and opening and upgrading, wherein element information in the third element subset is as follows: if not, generating a template or a natural language generation model according to a preset statement, and generating second output information as follows: do you not open the address book or upgrade the address book?
Step S105: and the information response device sends the second output information to the client so that the client feeds back the second output information.
Here, the information response device sends the obtained second output information to the client of the user, and the user can select, reply or respond to the second output information according to the content displayed by the client. And sending the selected or replied or responded feedback information to the information answering device. In accordance with the above embodiment, the user of the client side, according to the information displayed by the client side or the voice information output by the client side: do you not open the address book or upgrade the address book? The user responds by voice or selects on the screen of the client, and the feedback result of the second information is obtained as follows: and the address list cannot be upgraded, and the client sends the second feedback result to the information response device.
In some embodiments, after the output information is generated according to the element subset, the next output information is generated according to a feedback result fed back by the user to the output information, which is a process of loop iteration until all the element subsets in the element set are used for generating the output information.
Step S106: and the information response device outputs target output information according to a second feedback result fed back by the client.
Here, the target output information is matched with target intention information determined according to the second feedback result.
The information response device obtains a second feedback result sent by the client or directly fed back by the user, determines the target intention of the information input by the user, and then determines the target output information corresponding to the target intention of the user according to the target intention. In connection with the above embodiment, when the second feedback result indicates that the address book cannot be upgraded, it may be determined that the target intention of the user is to upgrade the address book, and the information response device may generate a method for upgrading the address book, that is, target output information, according to the upgrade address book, and output the target output information.
In the embodiment of the application, according to the acquired input information, an element set comprising at least two element subsets is determined, generating first output information according to a first element subset in the element set, sending the first output information to the client, generating second output information and sending the second output information to the client by at least combining elements in the second element subset according to the first feedback result of the client, and then generates and outputs target output information according to a second feedback result fed back by the client, so that the element information determined according to the acquired input information, the corresponding output information is sequentially sent to the user in a split screen form or a classification form, the intention corresponding to the input information of the user is accurately determined according to the feedback result of the user in sequence, and furthermore, the natural interaction between the information response device and the user is realized, and the intention of the user is gradually and accurately determined under the condition of causing the minimum trouble to the user.
In some embodiments, step S101 may also be implemented by:
step S1011: and the information response device determines the element set corresponding to the input information.
Here, after the information response device acquires the input information, the information response device analyzes the semantics of the input information based on the natural language understanding model, and specifies an element set corresponding to the input information from a keyword in the semantics.
Step S1012: and if the element information of the element set meets a preset condition, dividing the element set into at least a first element subset and a second element subset.
Here, the preset condition may be that the number of elements in the element set is greater than 2. When the number of elements in the element set is more than 2, determining the part of speech or semantic attribute of each element, and classifying the elements belonging to the same part of speech or the same attribute into a class to form an element subset.
Therefore, the element set corresponding to the input information can be obtained according to the acquired input information, and when the elements in the element set meet the preset conditions, the element set is classified, so that the elements are output according to the categories of the elements in the subsequent output operation.
In some embodiments, the element set may be divided according to the complexity of the element type, that is, step S1012 may also be implemented by:
the method comprises the following steps: determining a complexity of an element type in the set of elements.
Here, the complexity of the element type may be determined according to the part of speech of the element, such as low complexity of noun and high complexity of adverb.
Step two: and according to the complexity, dividing the element set into at least a first element subset and a second element subset.
Here, the set of elements is divided into at least a first subset of elements and a second subset of elements according to the element complexity, i.e., the part-of-speech type.
In some embodiments, the second output information may be generated by determining a target element in the first output information and combining information in the second subset of elements, i.e. step S104 may also be implemented by:
step S1041: and determining a target element included in the first output information according to a first feedback result fed back by the client.
Here, according to the first feedback result fed back by the client, an element in the first element subset included in the first feedback result may be determined, and then the element may be a target element of the first output information. For example, in connection with the above embodiment, when the user feeds back the feedback result according to the first output information: and if the address book is a problem, determining that the target element in the first output information is the address book according to the feedback result.
Step S1042: and generating second output information at least according to the target element and the second element subset in the element set.
Here, the target element and the second subset of elements in the element set are input into a preset template or a natural language generation model, and second output information is generated.
Therefore, the target element in the first output information can be determined according to the first feedback result fed back by the client, and then the second output information is generated by combining the information in other element subsets in the element set, so that the intention information of the user can be accurately determined.
On the basis of the embodiment shown in fig. 1, the present application further provides an information response method. Fig. 2 is a schematic flow chart of another implementation of the information response method provided in the embodiment of the present application, and will be described with reference to the steps shown in fig. 2:
step S201: and carrying out semantic analysis on the input information, and determining at least one intention information corresponding to the input information.
Here, the intention information is an object that the user wants to achieve when interacting with the information response device. After the information response device acquires the input information, the information response device analyzes the semantics of the input information based on the natural language understanding model to obtain a plurality of intention information corresponding to the input information.
Step S202: and splitting each intention information according to a preset splitting strategy to generate an element set at least comprising a first element subset and a second element subset.
Here, the preset splitting policy may be a policy that splits the intention information into two-tuple, three-tuple, or a more complex structure, which is generated according to the part of speech, semantic, and the like of the intention element in the intention information. In one example, based on the information input by the user, it is determined that the user's intention information is at least: the address list cannot be opened, the address list cannot be upgraded, the contact person cannot be opened and the contact person cannot be upgraded, and the obtained intention elements comprise: the method comprises the following steps of incapability, opening, upgrading, address book and contact person, and then according to the part of speech of elements: the difference in the types of nouns, verbs and adverbs splits the intent information into a first subset of elements: address book and contacts, second subset of elements: open and upgrade, third element subset: cannot be done.
In some implementations, the preset splitting strategy can also be splitting according to the concept of semantic units, wherein the semantic unit set is some meaningful word or phrase sequence defined by corpora or human. In some embodiments, as in the 3C domain, the set of semantic units may include: short message, address list, upgrade, open, send and fail etc. The splitting method is to split the intention according to an artificial setting mode, for example, the intention is split in the form of a triple < adverb, predicate and noun >, and a corresponding semantic unit is selected from the semantic units according to the intention to describe the intention, for example, the intention of 'unable to open a short message', and the description form of the corresponding triple intention is as follows: < unable (adverb), open (predicate), short message (noun) >.
Therefore, according to the preset splitting strategy, the intention information corresponding to the user input information can be split, elements with finer granularity can be obtained, and more accurate output information can be obtained.
In some embodiments, step S202 can also be implemented in the following two ways.
The first realization mode is as follows:
the method comprises the following steps: and determining a semantic composition unit of each intention information.
Here, the semantic component unit is a keyword included in the intention information, that is, an intention element. For example, the intention information is: the short message cannot be opened, and the included key words, namely semantic composition units, are as follows: short message, open and fail.
Step two: and splitting each intention information according to the part-of-speech type of the semantic composition unit to generate an element set at least comprising a first element subset and a second element subset.
Here, the part-of-speech type refers to types such as a verb, a noun, and an adverb. After determining a semantic composition unit, namely an intention element, of each intention information, splitting each intention information according to the part of speech type of the semantic composition unit to generate an element set at least comprising a first element subset and a second element subset. In one example, three element subsets corresponding to nouns, verbs and adverbs are generated according to short messages, opened and failed part-of-speech types, wherein elements in the first element subset are short messages, elements in the second element subset are opened, and elements in the third element subset are failed.
The second implementation mode is as follows:
the method comprises the following steps: determining a type of semantic attribute of each of the intention information.
Here, semantic attributes refer to a series of attributes defined for an intention, by which the intention can be described. The semantic attributes are four attributes of industry, employment, subject and action, and then an intention is described by the four attributes. The industry refers to the industry to which the intention belongs, and the action refers to the main body of action, namely, the person or thing who sends out the action or changes; a victim is an object of action, i.e., a person or thing subject to action; the action refers to the action related to the action of the affairs and the affairs, and the verb description is adopted.
Step two: and splitting each intention information according to the semantic attributes to generate an element set at least comprising a first element subset and a second element subset.
Here, each intention information is split according to four attributes of industry, employment, subject and action in the semantic attributes, and an element set including at least a first element subset and a second element subset is generated. In one example, the intent information is: can a child think about a trip? The business of semantic attributes can be analyzed as travel, the event as user and child, and the action as buying tickets. Thus, at least three subsets of elements are available, the elements in the first subset of elements being travel, the elements in the second subset of elements being buy tickets, and the elements in the third subset of elements being users and children.
In the embodiment of the application, the intention information corresponding to the input information can be obtained by performing semantic analysis on the input information, and then the intention information is split according to a preset splitting strategy to obtain at least one element subset, so that the elements with finer granularity corresponding to the intention information can be obtained, and further more accurate output information can be obtained.
Fig. 3 is a schematic flowchart illustrating another implementation of the information response method provided in the embodiment of the present application, and will be described with reference to the steps shown in fig. 3:
step S301: and determining fields included in the first element set to obtain a field set.
Here, a field refers to an element included in the first element subset.
Step S302: generating third output information for each field in the set of fields.
Here, according to each field, that is, each element, the output information corresponding to each field, that is, the third output information is generated according to the natural language generation model or the preset pre-acceptance generation template.
Step S303: and determining the first output information according to the third output information.
In this way, first output information including output information for each field may be generated from the fields included in the first subset of elements.
In some embodiments, step S302 may also be implemented by:
and generating third output information of each field by adopting a natural language generation model.
Here, a Natural Language Generation (NLG) model may convert structured data into text to be expressed in a human Language. Namely, a section of high-quality natural language text can be automatically generated through a planning process according to some fields, namely key information and the expression form of the key information in the machine. NLGs can be further divided into three broad categories, text-to-text (text-to-text), such as translations, abstracts, etc., text-to-other (text-to-other), such as text-to-picture, other-to-text (other-to-text), such as video-to-text. The method comprises the steps of carrying out semantic analysis on acquired user input information, determining at least one piece of intention information, splitting the intention information to obtain an intention element set, inputting at least one field included in a first element subset in the element set into an NLG (non-line segment) model, merging contents through reasonably organizing field sequences, adding proper connecting words in the determined contents, and forming sentences with proper structures, namely third output information, by all related words and the determined phrases.
In the embodiment of the present application, the intention information is determined by inputting the input information into the natural language generation model, and the corresponding output information is determined based on the intention information.
After obtaining user input information, a plurality of user intentions possibly related to the real intentions of the user are obtained based on natural language understanding, the intentions are split in a finer granularity, then the fine-granularity elements obtained based on splitting are summarized and summarized, a clarification technique is planned, and the user intentions are gradually clarified under the conditions of minimum turns and minimum user troubles through natural interaction, so that the user problems are solved.
Fig. 4 is a schematic flow chart of another implementation of the information response method provided in the embodiment of the present application, and will be described with reference to the steps shown in fig. 4:
step S401: and acquiring information input by a user.
Here, the user interacts with the information response device, the user inputs information at the client, the client sends the input information to the information response device, and the information response device acquires the information input by the user.
In some embodiments, the user enters information directly or by voice on the information answering device, which captures the information entered by the user.
Step S402: the Natural Language Understanding (NLU) module is adopted for Understanding, and the first N intentions are output.
Here, NLU gives importance to understanding of a computer mechanism such as natural language (human language characters). Specifically, the language, text, etc. are understood and useful information is extracted for use in downstream tasks. It may be a structuring of natural language, such as word segmentation, part-of-speech tagging, syntactic analysis, etc.; or the method can be characterized learning, vector representation (Embedding) of characters, words and sentences, and text classification of text representation is constructed; and the method can also be used for information extraction, such as information retrieval (including personalized search and semantic search, text matching and the like) and information extraction (named entity extraction, relation extraction, event extraction and the like). The method comprises the steps of adopting a natural language understanding module to carry out semantic analysis on information input by a user, determining at least one intention information, and outputting the first N intention information with high similarity, wherein N is a positive integer.
In one example, the information input by the user is input into the natural language understanding module, and after semantic parsing, five intention information in the following table 1 are obtained:
TABLE 1 intention information Table
| Intention information | Intention component element |
| Can not open short message | Fail, open, short message |
| Address book can not be opened | Fail, open, address book |
| Short message can not be upgraded | Failure to upgrade and short message |
| The address book can not be upgraded | Unable, upgraded and address book |
| Can not send short messages | The number of times of failure, transmission,short message |
Step S403: and acquiring the constituent elements of each intention to obtain an element set.
Here, for splitting the intention information, various splitting strategies may be adopted, such as splitting into two-tuple, three-tuple, or more complex structures, such as objectification, objectification of the intention, addition of attributes, and the like. For convenience of description, in the embodiment of the application, the triple < adverb, predicate, noun > is split to obtain an intention element set. As shown in table one, after splitting the determined intention information, five groups of intention composition elements are obtained.
Step S404: and selecting the intention elements from the element set one by one to confirm the intention elements to the user.
Here, when the number of elements is greater than 1, the same type of elements is obtained from the intention element set, and the corresponding clarifying dialog (the same as the first output information in the above embodiments) is generated through a template based on the elements, or a natural language generation model is used to generate the corresponding clarifying dialog, and the corresponding clarifying dialog is pushed to the user for the first clarification, that is, feedback, and then the target element corresponding to the type (the same as the first output information in the above embodiments) is determined according to the user feedback. Similarly, the intention element set is traversed one by one, and the intention elements of the respective types (the same as the output information in the above-described embodiments) are determined in sequence.
In some embodiments, when an element is obtained from an element set, different strategies may be customized according to actual needs, so as to reduce difficulty in user clarification (feedback) and the round of interaction required, for example, in the embodiment of the present application, splitting is performed in the form of a triple < adverb, predicate, noun >, because it is determined that a noun is easier to understand than a user by priority.
Step S405: a user intent is determined.
Here, a target intention of the user is determined according to the feedback of the user and a target element determined according to the feedback result, and then target output information is generated, the target output information matching the target intention of the user.
In the embodiment of the application, information input by a user is understood and analyzed based on natural language to obtain a plurality of user intention information possibly related to the real intention of the user, the intention information is split in a finer granularity, then the fine granularity elements obtained by splitting are summarized and summarized, a clarification technique is planned, and the user intention is gradually determined under the condition of the least turns and the least trouble of the user through natural interaction, so that the user problem is solved.
Fig. 5 is a schematic structural diagram of an information response device provided in an embodiment of the present application, and as shown in fig. 5, the information response device 500 includes: a first generating module 501, a second generating module 502, a first sending module 503, a third generating module 504, a second sending module 505, and an output module 506, wherein:
the first generating module 501 is configured to generate an element set including at least a first element subset and a second element subset according to the obtained input information.
The second generating module 502 is configured to generate first output information according to the first element subset.
The first sending module 503 is configured to send the first output information to a client, so that the client feeds back the first output information.
The third generating module 504 is configured to generate second output information at least according to the first feedback result fed back by the client and the second subset of elements in the element set.
The second sending module 505 is configured to send the second output information to the client, so that the client feeds back the second output information.
The output module 506 is configured to output target output information according to a second feedback result fed back by the client; wherein the target output information matches target intent information determined from the second feedback result.
In the above apparatus, the first generating module 501 includes:
the first determining submodule is used for determining an element set corresponding to the input information;
and the dividing sub-module is used for dividing the element set into at least the first element subset and the second element subset if the element information of the element set meets a preset condition.
In the above apparatus, the first molecular division module includes:
a first determining unit, configured to determine complexity of element types in the element set;
and the first dividing unit is used for dividing the element set into at least a first element subset and a second element subset according to the complexity.
In the above apparatus, the first generating module 501 includes:
the second determining submodule is used for carrying out semantic analysis on the input information and determining at least one intention information corresponding to the input information;
and the splitting submodule is used for splitting each intention information according to a preset splitting strategy to generate an element set at least comprising a first element subset and a second element subset.
In the above apparatus, the first splitting sub-module includes:
a second determination unit configured to determine a semantic component unit of each of the intention information;
and the first splitting unit is used for splitting each intention information according to the part of speech type of the semantic composition unit to generate an element set at least comprising a first element subset and a second element subset.
In the above apparatus, the first splitting sub-module includes:
a third determining unit configured to determine a type of semantic attribute of each of the intention information;
and the second splitting unit is used for splitting each intention information according to the semantic attributes to generate an element set at least comprising a first element subset and a second element subset.
In the above apparatus, when the first element set includes at least two fields, the second generating module 502 includes:
a third determining submodule, configured to determine fields included in the first element set, to obtain a field set;
the first generation submodule is used for generating third output information of each field in the field set;
and the fourth determining submodule is used for determining the first output information according to the third output information.
In the above apparatus, the first generating sub-module is further configured to generate third output information for each of the fields by using a natural language generation model.
In the above apparatus, the third generating module 504 includes:
a fifth determining submodule, configured to determine, according to a first feedback result fed back by the client, a target element included in the first output information;
and the second generation submodule is used for generating second output information according to the target element and the second element subset in the element set.
It should be noted that the description of the apparatus in the embodiment of the present application is similar to the description of the method embodiment, and has similar beneficial effects to the method embodiment, and therefore, the description is not repeated. For technical details not disclosed in the embodiments of the present apparatus, reference is made to the description of the method embodiments of the present application for understanding.
The embodiment of the application further provides an information response device, which comprises modules, sub-modules and units, and can be realized by a processor in electronic equipment; of course, the implementation can also be realized through a specific logic circuit; in implementation, the processor may be a central processing unit, a microprocessor, a digital signal processor, a field programmable gate array, or the like.
Correspondingly, an embodiment of the present application provides an electronic device, fig. 6 is a schematic view of a composition structure of the electronic device provided in the embodiment of the present application, and as shown in fig. 6, the electronic device 600 at least includes: a controller 601 and a storage medium 602 configured to store executable instructions, wherein:
the controller 601 is configured to execute stored executable instructions for implementing the provided information response method.
Here, it should be noted that: the above description of the storage medium and device embodiments is similar to the description of the method embodiments above, with similar advantageous effects as the method embodiments. For technical details not disclosed in the embodiments of the storage medium and apparatus of the present application, reference is made to the description of the embodiments of the method of the present application for understanding.
It should be appreciated that reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present application. Thus, the appearances of the phrases "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. It should be understood that, in the various embodiments of the present application, the sequence numbers of the above-mentioned processes do not mean the execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application. The above-mentioned serial numbers of the embodiments of the present application are merely for description and do not represent the merits of the embodiments.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described device embodiments are merely illustrative, for example, the division of the unit is only a logical functional division, and there may be other division ways in actual implementation, such as: multiple units or components may be combined, or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the coupling, direct coupling or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between the devices or units may be electrical, mechanical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units; can be located in one place or distributed on a plurality of network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, all functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may be separately regarded as one unit, or two or more units may be integrated into one unit; the integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.
Those of ordinary skill in the art will understand that: all or part of the steps for realizing the method embodiments can be completed by hardware related to program instructions, the program can be stored in a computer readable storage medium, and the program executes the steps comprising the method embodiments when executed; and the aforementioned storage medium includes: various media that can store program codes, such as a removable Memory device, a Read Only Memory (ROM), a magnetic disk, or an optical disk.
Alternatively, the integrated units described above in the present application may be stored in a computer-readable storage medium if they are implemented in the form of software functional modules and sold or used as independent products. Based on such understanding, the technical solutions of the embodiments of the present application may be essentially implemented or portions thereof contributing to the related art may be embodied in the form of a software product stored in a storage medium, and including several instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a removable storage device, a ROM, a magnetic or optical disk, or other various media that can store program code.
The above description is only for the embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (12)
1. An information answering method, the method comprising:
generating an element set at least comprising a first element subset and a second element subset according to the acquired input information;
generating first output information according to the first element subset;
sending the first output information to a client so that the client feeds back the first output information;
generating second output information at least according to a first feedback result fed back by the client and the second element subset in the element set;
sending the second output information to the client to enable the client to feed back the second output information;
outputting target output information according to a second feedback result fed back by the client; wherein the target output information matches target intent information determined from the second feedback result.
2. The method of claim 1, wherein generating a set of elements including at least a first subset of elements and a second subset of elements according to the obtained input information comprises:
determining an element set corresponding to the input information;
and if the element information of the element set meets a preset condition, dividing the element set into at least a first element subset and a second element subset.
3. The method according to claim 2, wherein if the element information of the element set satisfies a preset condition, dividing the element set into at least a first element subset and a second element subset, including:
determining a complexity of an element type in the set of elements;
and according to the complexity, dividing the element set into at least a first element subset and a second element subset.
4. The method of claim 1, wherein generating a set of elements including at least a first subset of elements and a second subset of elements according to the obtained input information comprises:
performing semantic analysis on the input information, and determining at least one intention information corresponding to the input information;
and splitting each intention information according to a preset splitting strategy to generate an element set at least comprising a first element subset and a second element subset.
5. The method according to claim 4, splitting each of the intention information according to a preset splitting policy to generate an element set including at least a first element subset and a second element subset, including:
determining a semantic component unit of each intention information;
and splitting each intention information according to the part-of-speech type of the semantic composition unit to generate an element set at least comprising a first element subset and a second element subset.
6. The method according to claim 4, splitting each of the intention information according to a preset splitting policy to generate an element set including at least a first element subset and a second element subset, including:
determining a type of semantic attribute of each of the intention information;
and splitting each intention information according to the semantic attributes to generate an element set at least comprising a first element subset and a second element subset.
7. The method of claim 1, when the first element set includes at least two fields, the generating first output information from the first element subset comprising:
determining fields included in the first element set to obtain a field set;
generating third output information of each field in the field set;
and determining the first output information according to the third output information.
8. The method of claim 7, the generating third output information for each field of the set of fields, comprising:
and generating third output information of each field by adopting a natural language generation model.
9. The method of claim 1, wherein generating second output information according to the first feedback result fed back by the client and the second subset of elements in the element set comprises:
determining a target element included in the first output information according to a first feedback result fed back by the client;
and generating second output information according to the target element and the second element subset in the element set.
10. An information answering device, said device comprising:
the first generation module is used for generating an element set at least comprising a first element subset and a second element subset according to the acquired input information;
the second generation module is used for generating first output information according to the first element subset;
the first sending module is used for sending the first output information to a client so that the client feeds back the first output information;
a third generating module, configured to generate second output information according to at least a first feedback result fed back by the client and the second subset of elements in the element set;
the second sending module is used for sending the second output information to the client so that the client feeds back the second output information;
the output module is used for outputting target output information according to a second feedback result fed back by the client; wherein the target output information matches target intent information determined from the second feedback result.
11. An electronic device, the electronic device comprising at least: a controller and a storage medium configured to store executable instructions, wherein:
the controller is configured to execute stored executable instructions configured to perform the information answering method of any one of the preceding claims 1 to 9.
12. A computer-readable storage medium having computer-executable instructions stored therein, the computer-executable instructions being configured to perform the message answering method as provided in any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010651380.0A CN111930904A (en) | 2020-07-08 | 2020-07-08 | Information response method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010651380.0A CN111930904A (en) | 2020-07-08 | 2020-07-08 | Information response method, device, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111930904A true CN111930904A (en) | 2020-11-13 |
Family
ID=73313554
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010651380.0A Pending CN111930904A (en) | 2020-07-08 | 2020-07-08 | Information response method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111930904A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150347500A1 (en) * | 2014-05-27 | 2015-12-03 | Baidu Online Network Technology (Beijing) Co., Ltd. | Interactive searching method and apparatus |
| US20160259846A1 (en) * | 2015-03-02 | 2016-09-08 | International Business Machines Corporation | Query disambiguation in a question-answering environment |
| CN109670088A (en) * | 2018-12-13 | 2019-04-23 | 平安科技（深圳）有限公司 | Voice response exchange method, device, computer equipment and storage medium |
| CN109710739A (en) * | 2018-12-24 | 2019-05-03 | 联想(北京)有限公司 | A kind of information processing method and device, storage medium |
| US20190147871A1 (en) * | 2017-11-16 | 2019-05-16 | Baidu Online Network Technology (Beijing) Co., Ltd | Intelligent interaction processing method and apparatus, device and computer storage medium |
| CN110929014A (en) * | 2019-12-09 | 2020-03-27 | 联想(北京)有限公司 | Information processing method, information processing device, electronic equipment and storage medium |
-
2020
- 2020-07-08 CN CN202010651380.0A patent/CN111930904A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150347500A1 (en) * | 2014-05-27 | 2015-12-03 | Baidu Online Network Technology (Beijing) Co., Ltd. | Interactive searching method and apparatus |
| US20160259846A1 (en) * | 2015-03-02 | 2016-09-08 | International Business Machines Corporation | Query disambiguation in a question-answering environment |
| US20190147871A1 (en) * | 2017-11-16 | 2019-05-16 | Baidu Online Network Technology (Beijing) Co., Ltd | Intelligent interaction processing method and apparatus, device and computer storage medium |
| CN109670088A (en) * | 2018-12-13 | 2019-04-23 | 平安科技（深圳）有限公司 | Voice response exchange method, device, computer equipment and storage medium |
| CN109710739A (en) * | 2018-12-24 | 2019-05-03 | 联想(北京)有限公司 | A kind of information processing method and device, storage medium |
| CN110929014A (en) * | 2019-12-09 | 2020-03-27 | 联想(北京)有限公司 | Information processing method, information processing device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Poongodi et al. | Chat-bot-based natural language interface for blogs and information networks | |
| Ranoliya et al. | Chatbot for university related FAQs | |
| CN110785763B (en) | Automated assistant-implemented method and related storage medium | |
| KR102580322B1 (en) | Automated assistants with conference capabilities | |
| US11250033B2 (en) | Methods, systems, and computer program product for implementing real-time classification and recommendations | |
| US11086601B2 (en) | Methods, systems, and computer program product for automatic generation of software application code | |
| US10705796B1 (en) | Methods, systems, and computer program product for implementing real-time or near real-time classification of digital data | |
| CN109314660B (en) | Method and device for providing news recommendation in automatic chat | |
| US20190272269A1 (en) | Method and system of classification in a natural language user interface | |
| Athreya et al. | Enhancing Community Interactions with Data-Driven Chatbots--The DBpedia Chatbot | |
| CN110110302B (en) | Identifying tasks in messages | |
| CN110770694B (en) | Obtaining response information from multiple corpora | |
| KR102249437B1 (en) | Automatically augmenting message exchange threads based on message classfication | |
| Nguyen et al. | NEU-chatbot: Chatbot for admission of National Economics University | |
| US10467122B1 (en) | Methods, systems, and computer program product for capturing and classification of real-time data and performing post-classification tasks | |
| CN110765244A (en) | Method and device for acquiring answering, computer equipment and storage medium | |
| US10169490B2 (en) | Query disambiguation in a question-answering environment | |
| US11429834B1 (en) | Neural-based agent assistance interface for providing answers based on a query vector | |
| US20200226216A1 (en) | Context-sensitive summarization | |
| Khin et al. | University chatbot using artificial intelligence markup language | |
| Milhorat et al. | A conversational dialogue manager for the humanoid robot ERICA | |
| US20200272791A1 (en) | Systems and methods for automated conversations with a transactional assistant | |
| McIntyre | From Cave Paintings to Shakespeare and Back Again: What Are Emoji and Should I Be Afraid? | |
| CN111930904A (en) | Information response method, device, equipment and storage medium | |
| Cebrián et al. | Considerations on creating conversational agents for multiple environments and users |
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 |