CN111222322B - Information processing method and electronic device - Google Patents
Information processing method and electronic device Download PDFInfo
- Publication number
- CN111222322B CN111222322B CN201911425001.XA CN201911425001A CN111222322B CN 111222322 B CN111222322 B CN 111222322B CN 201911425001 A CN201911425001 A CN 201911425001A CN 111222322 B CN111222322 B CN 111222322B
- Authority
- CN
- China
- Prior art keywords
- information
- user
- phrase
- current stage
- stage
- 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
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
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
- G06Q30/015—Providing customer assistance, e.g. assisting a customer within a business location or via helpdesk
- G06Q30/016—After-sales
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
- H04L51/02—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Human Computer Interaction (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Machine Translation (AREA)
Abstract
The embodiment of the disclosure provides an information processing method, which includes: acquiring first information, wherein the first information is information input by a user at the current stage of man-machine conversation; predicting second information, wherein the second information is information input by a user in the current stage and/or the adjacent stage of the current stage of the man-machine conversation; the first information is corrected based on the second information. The embodiment of the disclosure also provides an electronic device.
Description
Technical Field
The present disclosure relates to an information processing method and an electronic device.
Background
In an intelligent customer service conversation (namely dialogue) system, a conversation mode of a robot is determined according to user input information, and the robot conducts conversation guidance according to a preset conversation process so as to ensure the conversation to be conducted correctly and solve the problem of a user. However, in the process of conversation, the user may wrongly write the key words when inputting information, and if the key words are not corrected at this time, the conversation may not be performed according to the correct flow.
In implementing the disclosed concept, the inventors found that there are at least the following technical problems in the related art: in the related technology, error correction of the user input information is performed according to the local context of the user input information and the syntactic semantics of the user input information, but the user input information may be very brief, and also the syntactic semantics of the user input information may be reasonably smooth, but wrong words or other words (such as word groups which cannot accurately express semantics) substantially exist in the user input information, and it is difficult to recognize and correct the wrong words and other words in the user input information according to the local context and the syntactic semantics of the user input information.
Disclosure of Invention
One aspect of the present disclosure provides an information processing method, including: the method comprises the steps of obtaining first information, predicting second information, and correcting the first information based on the second information, wherein the first information is information input by a user at the current stage of man-machine conversation, and the second information is information expected to be input by the user at the current stage and/or the adjacent stage of the current stage of the man-machine conversation.
Optionally, the modifying the first information based on the second information includes: the first information is explained based on the second information.
Optionally, the modifying the first information based on the second information includes: and correcting the first information based on the second information in response to the data amount included in the first information being equal to or greater than a data amount threshold.
Optionally, the modifying the first information based on the second information further includes: and responding to the fact that the data volume contained in the first information is smaller than the data volume threshold value, acquiring at least one keyword according to the second information, and correcting the first information based on the at least one keyword.
Optionally, the modifying the first information based on the at least one keyword includes: and splitting the first information into at least one phrase, screening out target phrases which are contained in the at least one phrase and do not appear in the at least one keyword, and correcting the target phrases based on the at least one keyword.
Optionally, the modifying the first information based on the second information includes: and splitting the first information into at least one phrase, screening out a target phrase which is contained in the at least one phrase but not present in the second information, and correcting the target phrase based on the second information.
Optionally, the method further includes: before predicting the second information, the current stage is determined according to at least one of the following information, the syntactic characteristic and semantic information of the first information, and the information input by the user before inputting the first information.
Optionally, the determining the current stage according to the syntactic characteristic and the semantic information of the first information includes: and extracting at least one phrase from the first information according to the grammatical features of the first information, and comparing the keywords corresponding to each stage of the man-machine conversation with the at least one phrase to determine the current stage.
Optionally, the method further includes: determining whether a repetition rate of the first information and the dialog information input again is higher than a repetition rate threshold in response to a user inputting the dialog information again within a predetermined time before the second information is predicted, and not performing a correction operation on the first information and the dialog information input again in response to the repetition rate of the first information and the dialog information input again being higher than the repetition rate threshold.
Another aspect of the present disclosure provides an electronic device including: one or more processors, a storage device, for storing executable instructions, which when executed by the processors, implement the method as described above.
Another aspect of the present disclosure provides a computer-readable storage medium storing computer-executable instructions for implementing the method as described above when executed.
Another aspect of the present disclosure provides a computer program comprising computer executable instructions for implementing the method as described above when executed.
Drawings
For a more complete understanding of the present disclosure and the advantages thereof, reference is now made to the following descriptions taken in conjunction with the accompanying drawings, in which:
fig. 1 schematically shows an application scenario of an information processing method according to an embodiment of the present disclosure;
FIG. 2 schematically shows a flow chart of an information processing method according to an embodiment of the present disclosure;
3-5 schematically illustrate flow charts of modifying first information according to embodiments of the present disclosure;
FIG. 6 schematically shows a flow chart of an information processing method according to another embodiment of the present disclosure;
FIG. 7 schematically illustrates a flow chart for determining a current phase according to an embodiment of the present disclosure;
FIG. 8 schematically shows a flow chart of an information processing method according to another embodiment of the present disclosure; and
fig. 9 schematically shows a block diagram of an electronic device according to an embodiment of the disclosure.
Detailed Description
Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. It should be understood that the description is illustrative only and is not intended to limit the scope of the present disclosure. In the following detailed description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the disclosure. It may be evident, however, that one or more embodiments may be practiced without these specific details. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present disclosure.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The terms "comprises," "comprising," and the like, as used herein, specify the presence of stated features, steps, operations, and/or components, but do not preclude the presence or addition of one or more other features, steps, operations, or components.
All terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs, unless otherwise defined. It is noted that the terms used herein should be interpreted as having a meaning that is consistent with the context of this specification and should not be interpreted in an idealized or overly formal sense.
Where a convention analogous to "A, B and at least one of C, etc." is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., "a system having at least one of A, B and C" would include, but not be limited to, systems that have a alone, B alone, C alone, a and B together, a and C together, B and C together, and/or A, B, C together, etc.). Where a convention analogous to "A, B or at least one of C, etc." is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., "a system having at least one of A, B or C" would include, but not be limited to, systems that have a alone, B alone, C alone, a and B together, a and C together, B and C together, and/or A, B, C together, etc.).
Some block diagrams and/or flowcharts are shown in the figures. It will be understood that some blocks of the block diagrams and/or flowchart illustrations, or combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the instructions, which execute via the processor, create means for implementing the functions/acts specified in the block diagrams and/or flowchart block or blocks. The techniques of this disclosure may be implemented in hardware and/or software (including firmware, microcode, etc.). In addition, the techniques of this disclosure may take the form of a computer program product on a computer-readable storage medium having instructions stored thereon for use by or in connection with an instruction execution system.
Embodiments of the present disclosure provide an information processing method and an electronic device capable of applying the method, which may include the following operations, for example. First information is obtained, wherein the first information is information input by a user at the current stage of man-machine conversation. Second, second information is predicted, wherein the second information is information expected to be input by a user at the current stage and/or the adjacent stage of the man-machine conversation. The first information is corrected again based on the second information.
Fig. 1 schematically shows an application scenario of an information processing method according to an embodiment of the present disclosure. It should be noted that fig. 1 is only an example of a scenario in which the embodiments of the present disclosure may be applied to help those skilled in the art understand the technical content of the present disclosure, but does not mean that the embodiments of the present disclosure may not be applied to other devices, systems, environments or scenarios.
As shown in fig. 1, in the intelligent customer service conversation system, a terminal device 101 is connected with a telephone robot 103 through a network 102, a user inputs conversation information through the terminal device 101, the conversation robot 103 answers the conversation information input by the user in a targeted manner, but at the beginning of a conversation or during the conversation, the user may write wrong key words in the input conversation information, or the conversation information input by the user only contains very short and easily ambiguous words, and if the wrong key words in the input conversation information are not corrected, or the exact meaning of the short words in the conversation information input by the user is not understood, the conversation information is output directly based on the input information of the user, the intention of the user may not be recognized correctly, and the satisfactory conversation information of the user cannot be output, so that the user experience is poor.
By the processing method for the information input by the user, the information input by the user can be predicted based on the expectation of the user, and the predicted information is used for correcting the information input by the user, so that even if the conversation information input by the user contains wrongly written key words or only short words which are easy to generate ambiguity, wrong key words in the input information can be corrected better or the exact meaning of the short words can be determined, the real intention of the user can be grasped better, the conversation information which is satisfactory for the user can be output, and the user experience can be improved.
If a user enters a product after-sale platform (namely a session robot) of an e-commerce through a smart phone, the product after-sale platform is obtained by training a neural network, the product after-sale platform is an unattended platform, and the user inputs' how do the apple is bad? "apple" is well known as both fruit and electronic, i.e. the key word "apple" belongs to an ambiguous word. However, the product after-sale platform expects that the information input by the user is related to services such as after-sale maintenance, after-sale refuge, after-sale complaints and the like of the product, and because fruits belong to the fresh class, and fruit rot within a predetermined period limit belongs to a natural phenomenon, the user is asked how to deal with the damaged fruits, which generally does not belong to the service categories of after-sale maintenance, after-sale refuge and after-sale complaints, and then can determine what do is about the damaged apples input by the user? The "apple in" is an electronic product, not a fruit.
The present disclosure will be described in detail below with reference to the accompanying fig. 2 to 9 in conjunction with the specific embodiment shown in fig. 1.
Fig. 2 schematically shows a flow chart of an information processing method according to an embodiment of the present disclosure.
As an alternative embodiment, as shown in fig. 2, the information processing method may include the following operations S201 to S203.
In operation S201, first information is acquired, where the first information is information input by a user at a current stage of a man-machine conversation.
In an embodiment of the present disclosure, the first information is information input by a user, and the first information may include letters, numbers, symbols, and the like. The first information may contain wrong key words, and the first information may also be very brief. Furthermore, one dialog process of the human-machine dialog may include at least one of the stages of opening a scene, intent confirmation, filling a slot, checking, pushing, providing more help, ending, and the like. Wherein a dialog process can answer at least one question of a user.
In operation S202, second information, which is information expected to be input by a user at a current stage and/or a stage adjacent to the current stage of the man-machine conversation, is predicted.
In an embodiment of the present disclosure, the session robot may train the neural network with historical human-machine session records of a plurality of users to obtain a training model associated with a session stage based on information expected to be predicted by the user input, specifically, for each session stage, and then obtain second information from the database based on the current stage and/or a stage adjacent to the current stage using the training model. For example, the current session stage is an intent confirmation stage, and the training model may be utilized to obtain the second information from the database based on the intent confirmation stage. The second information may be predicted based on a desire for user input at the current stage and/or a stage adjacent to the current stage. Wherein the second information may comprise characters, numbers, symbols, and the like. For example, assume that the first information input by the user is "by", if the current phase of the man-machine conversation is "open white phase", the second information input by the user is expected to be "buy", and if the current phase of the man-machine conversation is "end phase", the second information input by the user is expected to be "bye". In addition, the current stage of the man-machine conversation can be determined according to the information input by the user before the first information is input or/and the grammar and the semantics of the first information.
In operation S203, the first information is corrected based on the second information.
In the embodiment of the present disclosure, whether to extract the keyword from the second information may be determined according to the data amount of the first information input by the user, if the data amount of the first information is greater than or equal to the data amount threshold, the keyword may not be extracted from the second information and the first information may be directly corrected using the second information, and if the data amount of the first information is less than the data amount threshold, the keyword may be extracted from the second information and the first information may be corrected using the extracted keyword. In addition, if the data amount of the first information is larger than that of most of the second information, the first information can be divided into a plurality of phrases, and then the individual phrases are corrected respectively, so that the correction accuracy is improved.
In the embodiment of the present disclosure, the operation S203 corrects the first information based on the second information, and may specifically include the following operations: the first information is interpreted based on the second information.
In the embodiment of the present disclosure, if a user enters an after-sale platform of a product of an e-commerce through a smart phone, and inputs "how do apple is bad on a screen of the smart phone? "since there is no improper or error in the user input, the second message may only explain and annotate that the apple is an electronic product, not a fruit, and does not correct the first message" how is the apple bad? ".
According to the embodiment of the disclosure, the second information is obtained by predicting the expectation of the user on the current stage and/or the adjacent stage of the current stage, and the first information input by the user is corrected by using the second information, so that the dialog intention of the user can be accurately identified, the reply information satisfied by the user is output, and the user experience is improved.
The method shown in fig. 2 is further described with reference to fig. 3-8 in conjunction with specific embodiments.
Fig. 3-5 schematically show flowcharts for modifying first information according to an embodiment of the disclosure.
In an alternative embodiment of the present disclosure, as shown in fig. 3, the operation S203 corrects the first information based on the second information, and may include the following operations: s3031 to S3034.
In operation S3031, it is determined whether the data amount included in the first information is equal to or greater than a data amount threshold.
In the embodiment of the present disclosure, by comparing the data amount of the first information with the data amount threshold, and determining whether to modify the first information directly using the second information or to modify the first information using the keyword extracted from the second information, the accuracy of modification may be improved.
Then, in operation S3032, in response to the data amount included in the first information being equal to or greater than the data amount threshold, the first information is corrected based on the second information.
In the embodiment of the disclosure, if the data amount included in the first information is greater than or equal to the data amount threshold, that is, the data amount of the first information is comparable to the data amount of each second information, the first information may be modified directly by using the second information, that is, the second information closest to the first information is selected from the plurality of second information (for example, the second information including the most keywords of the first information is selected as the second information closest to the first information), and the first information is modified by using the selected second information. For example, in the human-computer session end phase, the first information input by the user is "see yu next tom", and if the data amount threshold is 3, that is, the data amount of the first information is greater than the data amount threshold, and if the predicted second information includes "see yu next time", "good bye", "bye-bye", the first information may be corrected to "see yu next time" by directly using the second information.
Next, in operation S3033, in response to the data amount included in the first information being less than the data amount threshold, at least one keyword is acquired according to the second information.
In an embodiment of the present disclosure, if the data amount included in the first information is smaller than the data amount threshold, that is, the data amount included in the first information is smaller than the data amount included in each of the second information, one or more keywords may be extracted from the second information, and the first information may be corrected using the extracted keywords.
Then, the first information is modified based on the at least one keyword in operation S3034.
In the embodiment of the disclosure, a keyword closest to the first information may be selected from the extracted keywords, and the first information may be corrected. For example, in the end stage of the human-computer session, the first information input by the user is "by", and if the data amount threshold is 3, that is, the data amount of the first information is smaller than the data amount threshold, and if the predicted second information includes "see you next time", "good bye", "bye-bye", the keywords "see", "next", "time", "you", "good", "bye" may be extracted from the second information, and the "by" may be corrected by using the extracted keywords, thereby correcting "by" to "bye".
According to the embodiment of the disclosure, the first information can be determined to be directly corrected by using the second information according to the data size of the first information, but not corrected by using the keyword of the second information, so that the correction accuracy is improved.
In an alternative embodiment of the present disclosure, as shown in fig. 4, operation S3034 in fig. 3 modifies the first information based on at least one keyword, which may include at least the following operations S40341 to S40343.
In operation S40341, the first information is split into at least one phrase.
In an embodiment of the present disclosure, in case that there is a partial error in the first information input by the user, the first information may be split into a plurality of phrases (or words), thereby separating the erroneous phrase from the correct phrase. For example, the first information input by the user is "sm car not retrieved", and the first information may be split into "sm", "car", "not", and "retrieved".
Next, in operation S40342, target phrases that are included in the at least one phrase but do not appear in the at least one keyword are filtered out.
In the embodiment of the present disclosure, by comparing each phrase obtained by splitting with each keyword, an erroneous phrase can be found out. For example, assuming that at least one keyword includes "sim", "card", "phone", "cell", "can", "not", "connected", "ben", "identified", "telephone", "smartphone", "mobilephone", "mobile", "signal", "simcard", and assuming that the phrase for splitting the first information includes "sm", "car", "not", and "connected", the target phrase not appearing in the keyword includes "sm" and "car".
Then, in operation S40343, the target phrase is corrected based on the at least one keyword.
In embodiments of the present disclosure, the target phrase may be modified with the closest keywords to the target phrase. Following the example described above, "sm" may be corrected using "sim" that is most similar to "sm" and "car" may be corrected using "car" that is most similar to "car", the first information obtained after correction being "sim car not retrieved".
Through the embodiment of the disclosure, the first information can be split into a plurality of phrases (or words), so that an incorrect phrase and a correct phrase are separated, then each phrase is respectively compared with each keyword, phrases which do not appear in the keywords are found, and then the phrases which do not appear in the keywords are corrected. The error phrases in the first information can be effectively corrected, and the correction accuracy is improved.
In an alternative embodiment of the present disclosure, as shown in fig. 5, the operation S203 in fig. 2 corrects the first information based on the second information, which may include the following operations: S5031-S5033.
In operation S5031, the first information is split into at least one phrase.
In an embodiment of the present disclosure, if the data amount of the first information is equal to or greater than the data amount of the second information, the first information may be split into a plurality of phrases without processing the second information. For example, the first information input by the user is "I boot a new router, how to change my wife name". And based on the input expectation of the user, the predicted second information includes "I", "buy", "new router", "home to", "change", "wifi name", "wifi", "name", "user", "get", "receive", "wireless", "network", "line", "home spot", "home point", "access spot", "wave to", "home to", "new to", "wireless", "shell", "purchase". It can be seen that the data size of the first information is larger than that of the second information, the first information can be split into a plurality of phrases, auxiliary words such as articles and prepositions in the first information are removed, and the first information can be split into "I", "buy", "new route", "how to", "change" and "life name" after the tense of the verb and the single and plural number of nouns are prototype.
In operation S5032, a target phrase included in the at least one phrase but not appearing in the second information is filtered out.
In an embodiment of the present disclosure, after the first information is divided into a plurality of phrases, in order to find out an incorrect phrase in the phrases, each phrase may be compared with each second information, and if the phrase is not found in the second information, it may be determined that the phrase is an incorrect phrase. The above example of the joint sequentially searches the second information for "I", "buy", "new router", "how to", "change", and "wife name", and finally determines that "wife name" does not appear in the second information, that is, the target phrase is "wife name".
In operation S5033, the target phrase is corrected based on the second information.
In an embodiment of the present disclosure, the target phrase may be modified using the second information that is closest to the target phrase. In the above example, if the second information closest to the target phrase "wifi name" in the second information is "wifi name", the target phrase "wifi name" may be modified by using the "wifi name".
Through the embodiment of the disclosure, when the data volume of the first information is larger than that of the second information, the first information can be considered to be split into a plurality of phrases, then each split phrase is compared with each second information, error phrases in the plurality of phrases are found out, and then the second information is used for correcting the error phrases. For the condition that the first information input by the user is long, the first information can be divided into a plurality of phrases, and then the wrong phrases in the divided phrases are corrected, so that the correction accuracy rate is high.
Fig. 6 schematically shows a flow chart of an information processing method according to another embodiment of the present disclosure.
In an alternative embodiment of the present disclosure, as shown in fig. 6, before operation S202 in fig. 2, the information processing method may further include the following operation S601.
In operation S601, a current stage is determined according to at least one of the following information: syntactic and semantic information of the first information, information that a user inputs before inputting the first information.
In an embodiment of the disclosure, the grammatical feature of the first information may include a sentence pattern of the first information, such as a question, a statement sentence, a question-back sentence, and the like, and the semantic information of the first information includes a meaning to be expressed by the first information, a subject to be expressed by the first information, an idea to be expressed by the first information, and the like, for example, the first information is to purchase a mobile phone, and the performance of the mobile phone is to be queried. If the first message is "buy the mobile phone with headset? "it can be determined that the first information is an interrogative sentence and the subject to be expressed is the information of the consulting product accessories, therefore, it can be determined that the current stage is the check stage, and the conversation robot only needs to answer whether to buy the mobile phone or not to send the mobile phone.
In an embodiment of the present disclosure, if the user inputs a greeting before inputting the first information, it may be determined that the current stage is an intention confirmation stage, i.e. the current stage needs to determine the purpose of the user for this session. For example, the user entered "do you, at? ", the conversation robot replied with" what can help you, in worsted? At this point, when the user inputs the first information, the session enters the intention confirming stage.
Through the embodiment of the disclosure, the current stage of the man-machine conversation can be determined by combining the first information input by the user or the information input by the user before the first information is input. The current stage of the human-computer conversation can be accurately determined, the second information can be accurately predicted according to the expectation of the current stage and/or the adjacent stage of the current stage to the user input, and the first information is accurately corrected, so that the human-computer conversation is better guided, and the user experience is improved.
Fig. 7 schematically shows a flow chart of determining a current phase according to an embodiment of the present disclosure.
Specifically, in an alternative embodiment of the present disclosure, as shown in fig. 7, the operation S601 in fig. 6 may determine the current stage according to the syntax feature and the semantic information of the first information, and may include the following operations S6011 to S6012.
In operation S6011, at least one phrase is extracted from the first information according to a grammatical feature of the first information.
In the embodiment of the present disclosure, if the first information input by the user is "buy this huahua is whether the handset is attached with a headset? "it can be determined that the first information is an interrogative sentence, and therefore, phrases that can be extracted from the first information include" purchase "," huashi mobile "," headset ", and" accompanying ", which are also just semantic information of the first information.
Next, in operation S6012, the keywords corresponding to each stage of the man-machine conversation are compared with at least one phrase to determine a current stage.
In the embodiment of the present disclosure, if each stage of the human-computer conversation corresponds to one keyword group, in the above example, each phrase in the phrases "purchase", "hua mobile phone", "earphone" and "attach" may be compared with the keyword group corresponding to each stage of the human-computer conversation, and the conversation stage corresponding to the keyword group containing the most phrases is determined as the current stage.
Through the embodiment of the disclosure, a plurality of phrases can be extracted from the first information, and compared with the key phrases corresponding to the conversation stage, and the conversation stage corresponding to the key phrase containing the most phrases in the plurality of phrases can be used as the current stage, so that the current stage of the man-machine conversation can be objectively and accurately determined.
Fig. 8 schematically shows a flowchart of an information processing method according to another embodiment of the present disclosure.
In an alternative embodiment of the present disclosure, as shown in fig. 8, before operation S202 in fig. 2, the information processing method may further include the following operations S801 to S802.
In operation S801, in response to a user re-inputting dialog information within a predetermined time, it is determined whether a repetition rate of the first information and the re-input dialog information is higher than a repetition rate threshold.
In the embodiment of the present disclosure, if the user inputs a piece of dialogue information immediately after inputting the first information and thus the dialogue information is close to the first information, it can be assumed that the piece of dialogue information is a revision or a supplementary description of the first information. For example, a user wants to buy a mobile phone with millet, and then inputs a first message "is there millet? "if the user finds that the message is not specific enough, he immediately inputs a dialog message" do you go for millet 7? ", by comparing" does millet have goods left? "and" do millet 7 have good? ", the repetition rate of both can be determined. In addition, the predetermined time may be set to a time taken for the user to input one piece of information, such as 3 seconds, 5 seconds, 10 seconds, 15 seconds, or the like.
Then, in response to the repetition rate of the first information and the dialog information input again being higher than the repetition rate threshold, a correction operation is not performed on the first information and the dialog information input again in operation S802.
In the embodiment of the present disclosure, if the repetition rate of the first information and the dialog information input again is higher than the repetition rate threshold, it may be determined that the dialog information input again is a revision or a supplementary description of the first information, and a correction operation may not be performed on the first information and the dialog information input again. If the repetition rate threshold is 70%, the first information input by the user is "how to change my wife name? ", the dialog information input again by the user within 5 seconds after the first information is input is" how to change my wifi name? ", the first information" how to change my wife name? "with the dialog information" how to change my wifi name again entered? "the word repetition rate is 83.33%, it can be determined that the dialog information" how to change my wifi name? "is for the first information" how to change my wife name? And therefore, the correction operation may not be performed on the first information and the dialog information input again.
According to the embodiment of the disclosure, after the user finds that the information input by the user is wrong or the meaning of the input information is ambiguous, the user can actively correct the wrong input information in time or supplement and explain the input information with ambiguous meaning, and the wrong or ambiguous input information, the revised and supplemented information can not be corrected, so that the resources of the conversation robot are saved, the response efficiency of the conversation robot is improved, and the user experience is further improved.
Fig. 9 schematically shows a block diagram of an electronic device adapted to implement an information processing method according to an embodiment of the present disclosure. The electronic device shown in fig. 9 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present disclosure.
As shown in fig. 9, the electronic device 900 includes a processor 910 and a computer-readable storage medium 920. The electronic device 900 may perform a method according to an embodiment of the disclosure.
In particular, processor 910 may include, for example, a general purpose microprocessor, an instruction set processor and/or related chip set and/or a special purpose microprocessor (e.g., an Application Specific Integrated Circuit (ASIC)), and/or the like. The processor 910 may also include onboard memory for caching purposes. The processor 910 may be a single processing unit or a plurality of processing units for performing the different actions of the method flows according to embodiments of the present disclosure.
Computer-readable storage media 920, such as may be non-volatile computer-readable storage media, specific examples include, but are not limited to: magnetic storage devices, such as magnetic tape or Hard Disk Drives (HDDs); optical storage devices, such as compact disks (CD-ROMs); a memory, such as a Random Access Memory (RAM) or a flash memory; and so on.
The computer-readable storage medium 920 may include a computer program 921, which computer program 921 may include code/computer-executable instructions that, when executed by the processor 910, cause the processor 910 to perform a method according to an embodiment of the present disclosure, or any variation thereof.
The computer program 921 may be configured with, for example, computer program code comprising computer program modules. For example, in an example embodiment, code in computer program 921 may include one or more program modules, including, for example, 921A, modules 921B, … …. It should be noted that the division and number of the modules are not fixed, and those skilled in the art may use suitable program modules or program module combinations according to actual situations, and when the program modules are executed by the processor 910, the processor 910 may execute the method according to the embodiment of the present disclosure or any variation thereof.
The present disclosure also provides a computer-readable storage medium, which may be contained in the apparatus/device/system described in the above embodiments; or may exist separately and not be assembled into the device/apparatus/system. The computer-readable storage medium carries one or more programs which, when executed, implement the method according to an embodiment of the disclosure.
According to embodiments of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, which may include, for example but is not limited to: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based devices that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
Those skilled in the art will appreciate that various combinations and/or combinations of features recited in the various embodiments and/or claims of the present disclosure can be made, even if such combinations or combinations are not expressly recited in the present disclosure. In particular, various combinations and/or combinations of the features recited in the various embodiments and/or claims of the present disclosure may be made without departing from the spirit or teaching of the present disclosure. All such combinations and/or associations are within the scope of the present disclosure.
While the disclosure has been shown and described with reference to certain exemplary embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the disclosure as defined by the appended claims and their equivalents. Accordingly, the scope of the present disclosure should not be limited to the above-described embodiments, but should be defined not only by the appended claims, but also by equivalents thereof.
Claims (7)
1. An information processing method comprising:
acquiring first information, wherein the first information is information input by a user at the current stage of man-machine conversation;
predicting second information, wherein the second information is information input by a user in the current stage of the man-machine conversation and/or the adjacent stage of the current stage; wherein predicting the second information comprises: obtaining second information from a database based on the current stage and/or a stage adjacent to the current stage, or predicting the second information based on a desire for user input at the current stage and/or the stage adjacent to the current stage;
correcting the first information based on the second information;
wherein the modifying the first information based on the second information comprises:
splitting the first information into at least one phrase; filtering out target phrases that are included in the at least one phrase but do not appear in the second information; and modifying the target phrase based on the second information;
the correcting the first information based on the second information further comprises:
in response to the fact that the data volume contained in the first information is larger than or equal to a data volume threshold value, correcting the first information based on the second information;
responding to the fact that the data volume contained in the first information is smaller than the data volume threshold value, and acquiring at least one keyword according to the second information; and
modifying the first information based on the at least one keyword.
2. The method of claim 1, wherein the modifying the first information based on the second information comprises:
interpreting the first information based on the second information.
3. The method of claim 1, wherein said modifying the first information based on the at least one keyword comprises:
splitting the first information into at least one phrase;
screening out target phrases that are contained in the at least one phrase but do not appear in the at least one keyword; and
revising the target phrase based on the at least one keyword.
4. The method of claim 1, wherein the method further comprises: prior to said predicting second information, determining said current stage based on at least one of:
syntactic and semantic information of the first information;
information entered by a user prior to entering the first information.
5. The method of claim 4, wherein said determining the current stage from the syntactic and semantic information of the first information comprises:
extracting at least one phrase from the first information according to grammatical features of the first information; and
and comparing the keywords corresponding to each stage of the man-machine conversation with the at least one phrase to determine the current stage.
6. The method of claim 1, wherein the method further comprises: prior to the prediction of the second information,
determining whether a repetition rate of the first information and the dialog information input again is higher than a repetition rate threshold value in response to the user inputting the dialog information again within a predetermined time; and
in response to a repetition rate of the first information and the re-entered dialog information being higher than the repetition rate threshold, not performing a correction operation on the first information and the re-entered dialog information.
7. An electronic device, comprising:
one or more of the plurality of processors may,
storage means for storing executable instructions which, when executed by the processor, implement the method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911425001.XA CN111222322B (en) | 2019-12-31 | 2019-12-31 | Information processing method and electronic device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911425001.XA CN111222322B (en) | 2019-12-31 | 2019-12-31 | Information processing method and electronic device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111222322A CN111222322A (en) | 2020-06-02 |
| CN111222322B true CN111222322B (en) | 2022-10-25 |
Family
ID=70829230
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911425001.XA Active CN111222322B (en) | 2019-12-31 | 2019-12-31 | Information processing method and electronic device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111222322B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022199784A1 (en) * | 2021-03-22 | 2022-09-29 | Telefonaktiebolaget Lm Ericsson (Publ) | Electronic communications device, and methods of the electronic communications device for updating a previously sent text message |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2575128A3 (en) * | 2011-09-30 | 2013-08-14 | Apple Inc. | Using context information to facilitate processing of commands in a virtual assistant |
| US20160110327A1 (en) * | 2014-10-20 | 2016-04-21 | Lenovo (Singapore) Pte. Ltd. | Text correction based on context |
| CN104715066B (en) * | 2015-03-31 | 2017-04-12 | 北京奇付通科技有限公司 | Searching optimization method, searching optimization device and searching optimization system |
| JPWO2017195388A1 (en) * | 2016-05-12 | 2019-03-07 | ソニー株式会社 | Information processing apparatus, information processing method, and program |
| CN106777018B (en) * | 2016-12-08 | 2020-05-22 | 竹间智能科技（上海）有限公司 | Method and device for optimizing input sentences in intelligent chat robot |
| CN107728783B (en) * | 2017-09-25 | 2021-05-18 | 联想(北京)有限公司 | Artificial intelligence processing method and system |
| CN108804414A (en) * | 2018-05-04 | 2018-11-13 | 科沃斯商用机器人有限公司 | Text modification method, device, smart machine and readable storage medium storing program for executing |
| CN109002501A (en) * | 2018-06-29 | 2018-12-14 | 北京百度网讯科技有限公司 | For handling method, apparatus, electronic equipment and the computer readable storage medium of natural language dialogue |
| CN109086329B (en) * | 2018-06-29 | 2021-01-05 | 出门问问信息科技有限公司 | Topic keyword guide-based multi-turn conversation method and device |
| CN110211577B (en) * | 2019-07-19 | 2021-06-04 | 宁波方太厨具有限公司 | Terminal equipment and voice interaction method thereof |
-
2019
- 2019-12-31 CN CN201911425001.XA patent/CN111222322B/en active Active
Non-Patent Citations (1)
| Title |
|---|
| 面向导航的智能对话系统;范雄雄;《中国优秀硕士学位论文全文数据库信息科技辑》;20160315(第3期);第I138-7962页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111222322A (en) | 2020-06-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107437416B (en) | Consultation service processing method and device based on voice recognition | |
| CN106796787B (en) | Context interpretation using previous dialog behavior in natural language processing | |
| KR101418163B1 (en) | Speech recognition repair using contextual information | |
| US20170125013A1 (en) | Language model training method and device | |
| CN110970021B (en) | Question-answering control method, device and system | |
| WO2017166650A1 (en) | Voice recognition method and device | |
| KR20200108775A (en) | Training corpus generating method, apparatus, device and storage medium | |
| CN110415679B (en) | Voice error correction method, device, equipment and storage medium | |
| CN106997342B (en) | Intention identification method and device based on multi-round interaction | |
| JP6677419B2 (en) | Voice interaction method and apparatus | |
| CN109783626B (en) | Question generating method, intelligent question-answering system, medium and computer system | |
| CN112286485B (en) | Method and device for controlling application through voice, electronic equipment and storage medium | |
| CN109410923B (en) | Speech recognition method, apparatus, system and storage medium | |
| CN111159364A (en) | Dialogue system, dialogue device, dialogue method, and storage medium | |
| CN111159358A (en) | Multi-intention recognition training and using method and device | |
| CN108831444B (en) | Semantic resource training method and system for voice conversation platform | |
| CN115203394A (en) | Model training method, service execution method and device | |
| CN112084317A (en) | Method and apparatus for pre-training a language model | |
| KR102102287B1 (en) | Method for crowdsourcing data of chat model for chatbot | |
| CN115129878A (en) | Conversation service execution method, device, storage medium and electronic equipment | |
| CN105373527B (en) | Omission recovery method and question-answering system | |
| CN111222322B (en) | Information processing method and electronic device | |
| KR102450816B1 (en) | Real-time auto-interpretation system and method for verifying utterance thereof | |
| US20140156256A1 (en) | Interface device for processing voice of user and method thereof | |
| KR102536944B1 (en) | Method and apparatus for speech signal processing |
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 |