Movatterモバイル変換


[0]ホーム

URL:


CN113887210B - Time information processing method, device, electronic device and storage medium - Google Patents

Time information processing method, device, electronic device and storage medium
Download PDF

Info

Publication number
CN113887210B
CN113887210BCN202111217421.6ACN202111217421ACN113887210BCN 113887210 BCN113887210 BCN 113887210BCN 202111217421 ACN202111217421 ACN 202111217421ACN 113887210 BCN113887210 BCN 113887210B
Authority
CN
China
Prior art keywords
message
time
target
candidate
target operation
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
Application number
CN202111217421.6A
Other languages
Chinese (zh)
Other versions
CN113887210A (en
Inventor
林江华
刘丹
宋双永
邹波
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jingdong Technology Information Technology Co Ltd
Original Assignee
Jingdong Technology Information Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Jingdong Technology Information Technology Co LtdfiledCriticalJingdong Technology Information Technology Co Ltd
Priority to CN202111217421.6ApriorityCriticalpatent/CN113887210B/en
Publication of CN113887210ApublicationCriticalpatent/CN113887210A/en
Application grantedgrantedCritical
Publication of CN113887210BpublicationCriticalpatent/CN113887210B/en
Activelegal-statusCriticalCurrent
Anticipated expirationlegal-statusCritical

Links

Classifications

Landscapes

Abstract

Translated fromChinese

本公开提出一种时间信息的处理方法、装置、电子设备及存储介质,其中,方法包括:获取对话中的多条消息;对各条消息进行识别,以确定各条消息中是否包含执行目标操作的候选时间,以及各条消息是否属于否定目标操作的目标意图;从各条消息中确定目标消息;根据目标消息所包含的执行目标操作的时间信息,确定执行目标操作的目标时间。该方法通过各条消息是否包含执行目标操作的候选时间以及是否属于目标操作的目标意图,从各条消息中确定目标消息,进而,根据目标消息所包含的目标操作的时间信息,确定执行目标操作的执行时间,由此,可准确地确定执行目标操作的执行时间。

The present disclosure proposes a method, device, electronic device and storage medium for processing time information, wherein the method comprises: obtaining multiple messages in a conversation; identifying each message to determine whether each message contains a candidate time for executing a target operation and whether each message belongs to a target intention of negating the target operation; determining a target message from each message; and determining a target time for executing the target operation according to the time information for executing the target operation contained in the target message. The method determines the target message from each message by determining whether each message contains a candidate time for executing the target operation and whether it belongs to the target intention of the target operation, and then determines the execution time of executing the target operation according to the time information of the target operation contained in the target message, thereby accurately determining the execution time of executing the target operation.

Description

Time information processing method and device, electronic equipment and storage medium
Technical Field
The disclosure relates to the technical field of information processing, and in particular relates to a method and a device for processing time information, electronic equipment and a storage medium.
Background
In the field of conversations, for example, in the field of customer service conversations, time information in the conversations needs to be identified and extracted so as to realize target operations such as event reminding, quality inspection and auxiliary answering according to the time information, but only according to the time information, the time for executing the target operations such as event reminding, quality inspection and auxiliary answering cannot be accurately determined, for example, "twelve points in open day are inconvenient for receiving calls", and if a call is made to a customer according to the identified "twelve points", the call can be made to the customer at an incorrect time.
Disclosure of Invention
The object of the present disclosure is to solve, at least to some extent, one of the problems in the art described above.
The disclosure provides a method, a device, an electronic device and a storage medium for processing time information, so as to determine a target message from each message according to whether each message contains a candidate time for executing a target operation and whether each message belongs to a target intention of the target operation, and further determine execution time for executing the target operation according to the time information of the target operation contained in the target message, thereby accurately determining the execution time for executing the target operation.
An embodiment of a first aspect of the present disclosure provides a method for processing time information, including obtaining a plurality of messages in a conversation, identifying each message to determine whether each message includes a candidate time for executing a target operation and whether each message belongs to a target intention for negating the target operation, determining a target message from each message, wherein the target message includes the candidate time for executing the target operation and each message in the conversation after the target message does not belong to the target intention, and determining a target time for executing the target operation according to the time information included in the target message for executing the target operation.
According to the technical scheme, the target message is determined from each message according to whether each message contains the candidate time for executing the target operation and whether the target intention belongs to the target operation, and further, the execution time for executing the target operation is determined according to the time information of the target operation contained in the target message, so that the execution time for executing the target operation can be accurately determined.
Optionally, determining the target message from the messages comprises determining the message containing the candidate time in the dialogue as a candidate message, marking the candidate time contained in each candidate message before one candidate message as invalid for any one candidate message under the condition that the target intention is determined, and taking the candidate message without the invalid mark as the target message.
Optionally, the step of taking the candidate message without the invalid tag as the target message comprises the step of taking the candidate message without the invalid tag, which is ordered later in the dialogue, as the target message when the candidate message without the invalid tag is a plurality of candidate messages.
Optionally, the identifying each message to determine whether each message contains a candidate time for executing a target operation and whether each message belongs to a target intention for negating the target operation comprises identifying time texts for each message, generating a time sequence according to the identified time texts, wherein the time sequence comprises time texts contained in each message, each time text is ordered according to the sequence in the dialogue, identifying whether each time text in the time sequence belongs to the candidate time for executing the target operation, and identifying whether each message belongs to the target intention.
Optionally, the identifying whether each time text in the time sequence belongs to a candidate time for executing the target operation includes inputting a message to which the time text belongs and at least one message before the message into a first identification model for any time text in the time sequence to identify whether the time text belongs to a candidate time for executing the target operation.
Optionally, the identifying whether each message belongs to the target intention includes inputting the one message and at least one message before the one message into a second identification model for any one message in the dialogue to identify whether the one message belongs to the target intention.
An embodiment of the second aspect of the present application provides a processing device for time information, which includes an acquisition module configured to acquire a plurality of messages in a conversation, an identification module configured to identify each of the messages to determine whether each of the messages includes a candidate time for executing a target operation and whether each of the messages belongs to a target intention for negating the target operation, a first determination module configured to determine a target message from each of the messages, wherein the target message includes the candidate time for executing the target operation and each of the messages in the conversation after the target message does not belong to the target intention, and a second determination module configured to determine a target time for executing the target operation according to time information included in the target message for executing the target operation.
An embodiment of the third aspect of the present application provides an electronic device, including at least one processor, and a memory communicatively connected to the at least one processor, where the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can perform the method for processing time information set forth in the embodiment of the first aspect of the present application.
An embodiment of a fourth aspect of the present application proposes a non-transitory computer-readable storage medium storing computer instructions for causing the computer to execute the processing method of time information proposed by the embodiment of the first aspect of the present application.
An embodiment of a fifth aspect of the present application proposes a computer program product, which when executed by an instruction processor in the computer program product, performs the method for processing time information proposed by the embodiment of the first aspect of the present application.
Additional aspects and advantages of the disclosure will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the disclosure.
Drawings
The foregoing and/or additional aspects and advantages of the present disclosure will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:
fig. 1 is a flowchart illustrating a method for processing time information according to an embodiment of the disclosure;
Fig. 2 is a flowchart of a method for processing time information according to a second embodiment of the disclosure;
fig. 3 is a flowchart of a method for processing time information according to a third embodiment of the present disclosure;
FIG. 4 is a flow chart of a method for processing time information according to an embodiment of the disclosure;
FIG. 5 is a flowchart of a method for processing time information according to another embodiment of the present disclosure;
Fig. 6 is a schematic structural diagram of a processing device for time information according to an embodiment of the present disclosure;
fig. 7 is a block diagram of an electronic device showing a processing method for time information according to an exemplary embodiment.
Detailed Description
Embodiments of the present disclosure are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and intended for the purpose of explaining the present disclosure and are not to be construed as limiting the present disclosure.
The following describes a method, an apparatus, an electronic device, and a storage medium for processing time information according to an embodiment of the present disclosure with reference to the accompanying drawings.
Fig. 1 is a flowchart of a time information processing method according to an embodiment of the present disclosure, and it should be noted that the time information processing method may be applied to a time information processing apparatus according to an embodiment of the present disclosure, and the apparatus may be configured in an electronic device. The electronic device may be a mobile terminal, such as a mobile phone, a tablet computer, a personal digital assistant, or other hardware devices with various operating systems.
As shown in fig. 1, the method for processing the time information may include the following steps:
Step 101, a plurality of messages in a conversation are acquired.
In the embodiment of the disclosure, multiple utterances of a speaker in a conversation may be acquired, and the multiple utterances of the speaker in the conversation may be corresponding to the multiple utterances as multiple messages in the conversation.
Step 102, identifying each message to determine whether each message contains a candidate time for executing the target operation and whether each message belongs to a target intention for negating the target operation.
Further, content recognition is performed on each of the plurality of messages in the conversation, and whether the candidate time for executing the target operation is included in each of the messages or not and whether each of the messages belongs to the target intention for negating the target operation are determined according to the recognition content. For example, the message is "give me call before 12 points", the target is call back ", the content recognition is performed on the message, it can be determined that the message does not include the candidate time for executing the target operation and the message belongs to the target intention for negating the target operation, for example, the message is" inconvenient, the message is given me call back bar after 12 points ", the target is call back, the content recognition is performed on the message, the message includes the candidate time for executing the target operation and the message belongs to the target intention for negating the target operation, for example," can, something else after 12 points "the target is call back, the content recognition is performed on the message, it can be determined that the message does not include the candidate time for executing the target operation and the message does not belong to the target intention for negating the target operation, for example, the message is" call back can be given me before 12 am days ", the content recognition is performed on the message, it can be determined that the message includes the candidate time for executing the target operation and the message does not belong to the target intention for negating the target operation.
And step 103, determining a target message from the messages, wherein the target message contains candidate time for executing the target operation, and the messages in the dialogue after the target message do not belong to the target intention.
In the embodiment of the disclosure, according to whether each message contains a candidate time for executing the target operation and whether each message belongs to the target intention of the negative target operation, the target message can be determined from each message, wherein the target message contains the candidate time for executing the target operation, and each message after the target message in the dialogue does not belong to the target intention of the negative target operation.
For example, using the target operation as a power back, the messages in the conversation are shown in table 1,
Table 1 opposite messages in dialogue
Identifying 6 messages in table 1 can determine that messages containing candidate time of return are message 2 to message 5, message belonging to target intention of negative return is message 3, messages not belonging to target intention of negative return are messages 2,1 and 5, message 5 is rewriting of message 4, and message 5 is target message.
Step 104, determining the target time for executing the target operation according to the time information for executing the target operation contained in the target message.
In the embodiment of the present disclosure, time information for executing a target operation, which is included in a target message, is taken as a target time for executing the target operation. For example, the target message in table 1 is "can be before 10 am", and the final power back time can be determined to be before 10 am on tomorrow. It should be noted that, in order to improve the feasibility of the embodiment of the present disclosure, for the target time for performing the target operation to be "before x point" or "after x point", correspondingly, the target time for performing the target operation may be set to be N minutes forward of the target time or N minutes backward of the target time, for example, the target time for performing the target operation may be "before 12 points", the target time for performing the target operation may be set to be "57 points", for example, the target time for performing the target operation may be "after 12 points", and the target time for performing the target operation may be set to be "03 points".
In summary, by determining whether each message includes a candidate time for executing the target operation and whether the message belongs to the target intention of the target operation, the target message is determined from each message, and further, the execution time for executing the target operation is determined according to the time information of the target operation included in the target message, thereby accurately determining the execution time for executing the target operation.
In order to accurately determine the target message from the messages, as shown in fig. 2, fig. 2 is a flow chart of a processing method of time information provided in the second embodiment of the disclosure. In an embodiment of the present disclosure, a message in a conversation including a candidate time for performing a target operation may be determined as a candidate message, and a candidate message having no invalid flag may be determined as a target message, and the embodiment shown in fig. 2 may include the steps of:
In step 201, a plurality of messages in a conversation are acquired.
Step 202, identifying each message to determine whether each message contains a candidate time for performing the target operation and whether each message belongs to a target intention for negating the target operation.
In step 203, the message containing the candidate time in the conversation is determined as a candidate message.
In the embodiment of the disclosure, the message content of each message is identified, whether each message contains the candidate time for executing the target operation or not and whether each message belongs to the target intention for negating the target operation or not can be determined, and the message containing the candidate time for executing the target operation is taken as the candidate message.
Step 204, for any one candidate message, in the case of determining that the target intention is included, marking the candidate time contained in each candidate message before the one candidate message as invalid.
In step 205, the candidate message without the invalid flag is taken as the target message.
Further, for any candidate message, it may be determined whether the candidate message belongs to the target intention of the negative target operation, if the message belongs to the target intention of the negative target operation, the candidate time stamp contained in each candidate message before the message may be invalid, if the candidate message without the invalid stamp is one, the candidate message without the invalid stamp may be regarded as the target message, and if the candidate message without the invalid stamp is a plurality of candidate messages with the invalid stamp, the candidate message with the invalid stamp ordered later in the conversation may be regarded as the target message. Wherein the target message contains a candidate time for performing the target operation and each message in the conversation that follows the target message does not belong to the target intent.
For example, in table 1, messages 2 to 5 are candidate messages, message 3 belongs to the target intention of the negative target operation, the candidate time contained in message 2 before message 3 may be marked as invalid, messages 4 and 5 have no invalid mark, and the candidate message 5 with the invalid mark after the sequence is the target message.
In step 206, the target time for executing the target operation is determined according to the time information for executing the target operation contained in the target message.
In the embodiment of the present disclosure, steps 201 to 202 and step 206 may be implemented in any manner in each embodiment of the present disclosure, which is not limited to this embodiment, and is not repeated herein.
In summary, a message including candidate time in a conversation is determined as a candidate message, when a target intention is determined to belong to any one of the candidate messages, the candidate time included in each candidate message preceding the one candidate message is marked as invalid, and the candidate message without the invalid mark is taken as the target message. Thus, the target message can be accurately determined from the messages.
In order to accurately determine whether each message contains a candidate time for performing the target operation and whether each message belongs to a target intention for negating the target operation, as shown in fig. 3, fig. 3 is a flowchart of a method for processing time information according to the third embodiment of the present disclosure. In the embodiment of the disclosure, a time sequence may be generated according to the time text in the message, and each time text in the time sequence may be identified to determine whether each message contains a candidate time for performing the target operation and whether each message belongs to a target intention for negating the target operation. The embodiment shown in fig. 3 may include the following steps:
step 301, a plurality of messages in a conversation are acquired.
Step 302, time text is identified for each message.
In the embodiment of the disclosure, time text recognition may be performed on each message, optionally, text content matched with one or more preset regular expressions is extracted from each message according to the preset regular expressions, and the matched text content is parsed to obtain a corresponding time text, where it is to be noted that each preset regular expression corresponds to a parsing mode field, and the parsing mode field may be used to parse the time text to a specific time point. For example, as shown in table 2, text content matched with a preset regular expression can be extracted from each message, and further, as shown in table 3, the matched text content is processed according to an analysis mode based on the current time, so as to obtain a corresponding time text. For example, the matched text content is processed with the date of 2021 being 5 months and 8 days, the acquired corresponding time text is 2021-05-08, for example, the matched text content is processed with the date of 2021-05-01:00:00 before 6 pm, for example, the matched text content is processed with the date of 2021-05-01:00:00, for example, the matched text content is processed within 24 hours, and the acquired corresponding time text is 2021-05-02:14:15.
Table 2 extracts text content matching a preset regular expression
TABLE 3 processing matched text content according to parsing scheme
In order to effectively identify the time text in each message, each message in the dialogue may be text cleaned before identifying the time text in each message, for example, english uppercase lowercase in each message, chinese digit to arabic digit, traditional chinese to simplified chinese, full-angle character to half-angle character, emoji removal, non-text message removal, blank message removal, pure symbol removal, URL (Uniform Resource Locator ) removal link, and the like may be performed.
Step 303, generating a time sequence according to the identified time text, wherein the time sequence comprises time texts contained in each message, and each time text is ordered according to the sequence in the dialogue.
It should be appreciated that for each message in a conversation, the sequence number of each message may be marked in the order of transmission, and the speaker of the message may be marked, along with the speaking time, where the speaker may be used to mark which speaker the message was sent by, and the speaking time may be used to mark the time of transmission of the message, so that the subsequent computation time is referenced to the speaking time.
In the embodiment of the disclosure, after the time texts are identified, the time texts may be sequentially arranged according to the marking order of the messages in the dialogue, and the arrangement result is used as a time sequence. The time sequence may include a time text, an parsing manner corresponding to the time text, a start position of the time text in the message, an end position of the time text in the message, and the like.
Step 304, identifying whether each time text in the time series belongs to a candidate time for performing the target operation, and identifying whether each message belongs to a target intention.
Further, each time text in the time series may be identified to determine whether the time text belongs to a candidate time for performing the target operation.
Optionally, any one time text in the time sequence, a message to which the one time text belongs and at least one message preceding the one time text are input into the first recognition model to recognize whether the one time text belongs to a candidate time for performing the target operation.
For example, as shown in table 4, for example, the time text in the time sequence is "before 2 pm today", the message corresponding to the time text is "bar before 2 pm today" and the previous message is "give you back before 12 pm in me open day," ask you when you are convenient to answer a call. For another example, the time text in the time sequence is "before 12 points", the message "call me before 12 points" corresponding to the time text and the previous message "ask me for an air ticket of 5 months 21 days to Beijing", "ask you when to get things convenient for goods, and" before two afternoon bar today "input a first recognition model, the model can recognize the time text, and it is determined that the message corresponding to the time text" before 12 points "is not a sentence expressing the callback time, and the time text does not belong to the candidate time for executing the target operation. It should be noted that the first recognition model may be a trained neural network model.
Table 4 identifies whether the time text belongs to the return time
In embodiments of the present disclosure, each message in a conversation may be identified while identifying each time text in a time series to determine whether the time text belongs to a candidate time for performing a target operation, to determine whether the message belongs to a target intent for a negative target operation.
Optionally, for any one of the messages in the conversation, one message and at least one message preceding the one message are input into the second recognition model to recognize whether the one message belongs to a target intention of a negative target operation.
For example, as shown in table 5, for example, the message "i am not available in the afternoon" and the previous message "i am give you a return to electricity before 18 pm" are input into the second recognition model, and the model can recognize the message, and determine that the message "i am not available in the afternoon" belongs to a negative return, which indicates that the message belongs to the target intention of the negative target operation. For another example, the message "good" and the previous message "give you a return to electricity before 12 am in my tomorrow" are input into a second recognition model, which can recognize the message, determine that the message "good" belongs to a non-negative return to electricity, and indicate that the message does not belong to the target intention of the negative target operation. It should be noted that the second recognition model may be a trained neural network model.
Table 5 identifies whether the message belongs to a negative return call
It should be noted that, when the message including the time text is identified by the first identification model and the second identification model, the message including the time text may be classified into a category in which the time text in the message belongs to a candidate time for executing the target operation, the message belongs to a target intention for negating the target operation, the time text in the message does not belong to a candidate time for executing the target operation, the message belongs to a target intention for negating the target operation, the time text in the message belongs to a candidate time for executing the target operation, the message does not belong to a target intention for negating the target operation, and the time text of the message does not belong to a candidate time for executing the target operation.
For example, as shown in table 6, taking the candidate time for executing the target operation as the return time, the target for the negative target operation is intended to be negative return, for example, "before 12 points" in the message "give me call before 12 points" belongs to the non-return time, and belongs to the negative return time, for the message "inconvenient," after 12 points "in the call bar of me call belongs to the return time, and the message belongs to the negative return time, for the message" can, "after 12 points" in the something "after 12 points" belongs to the non-return time, and the message belongs to the non-negative return time, for the message "before 12 points" in the call bar of me call before 12 points "belongs to the return time, and the message belongs to the non-negative return time.
Table 6 message categories including time text
Telephone me before 12 pointsTime of non-return of electricityNegative return electricity
Inconvenient, telephone bar for me after 12 pointsTime of return of electricityNegative return electricity
Can be there is something after I12 pointsTime of non-return of electricityNon-negative return electricity
Telephone for me before 12 pointsTime of return of electricityNon-negative return electricity
And step 305, determining a target message from the messages, wherein the target message contains candidate time for executing the target operation, and the messages in the dialogue after the target message do not belong to the target intention.
Step 306, determining the target time for executing the target operation according to the time information for executing the target operation contained in the target message.
In the embodiment of the present disclosure, steps 301 and steps 306 to 306 may be implemented in any manner in each embodiment of the present disclosure, which is not limited to this embodiment, and will not be described in detail.
In summary, a time sequence is generated by identifying a time text for each message, wherein the time sequence includes time texts contained in each message, each time text is ordered according to a sequence in a conversation, each time text in the time sequence identifies whether each time text belongs to a candidate time for performing a target operation, and each message identifies whether each message belongs to a target intention. Thus, it can be accurately determined whether or not the candidate time for performing the target operation is included in each message, and whether or not each message belongs to the target intention for negating the target operation.
In order to make the present disclosure more apparent to those skilled in the art, examples are now described.
The processing procedure of the time information in the customer dialogue is as shown in fig. 4, and is exemplified by (1) acquiring the customer dialogue, (2) text cleaning, namely sorting cleaned texts in the customer dialogue to acquire a message sequence, (4) time identification, namely determining whether the message contains candidate time for executing target operation or not, and (5) intention identification, namely after time identification, conducting intention identification on the message sequence, (6) time correction, namely marking the message containing the candidate time as invalid when the message contains the candidate time and belongs to target intention of negative target operation, namely acquiring the candidate message sequence without invalid mark in the dialogue, (7) sequence integration, namely, taking the candidate message without invalid mark after the conversation is sequenced in the dialogue as target message, (9) identification result, namely determining the time information for executing the target operation contained in the target message as the target time for executing the target operation.
It should be noted that, as shown in fig. 5, taking the candidate time for executing the target operation as the power-back time, the target intention of the negative target operation is the negative power-back, for example, the process of time correction in fig. 4 may be as follows:
(1) Checking whether the message identifies time in the time identification step, and if not, jumping to the step (3);
(2) If the classification result of the power-back time is 'non-power-back time', deleting all the results of the message from the time sequence;
(3) Checking whether a record with a valid field exists in the time sequence, and ending the flow if no record exists;
(4) As shown in tables 7 and 8, if the negative power-back classification result is "negative power-back", the "valid or not" field in all the results corresponding to the last time-available message preceding the present message in the time sequence is modified to be invalid. It should be noted that, this step does not affect the time identified by the current sentence, so as to better process a plurality of expression scenes contained in one message.
TABLE 7 scene 1
TABLE 8 scene 2
According to the time information processing method, whether each message contains the candidate time for executing the target operation or not and whether the message belongs to the target intention of the target operation or not is used for determining the target message from each message, and further, the execution time for executing the target operation is determined according to the time information of the target operation contained in the target message, so that the execution time for executing the target operation can be accurately determined.
In order to implement the above embodiment, the present disclosure also proposes a processing apparatus of time information.
Fig. 6 is a schematic structural diagram of a processing device for time information according to an embodiment of the present disclosure.
As shown in fig. 6, the processing apparatus of time information includes an acquisition module 610, an identification module 620, a first determination module 630, and a second determination module 640.
The system comprises an acquisition module 610, a recognition module 620 and a second determination module 640, wherein the acquisition module 610 is used for acquiring a plurality of messages in a conversation, the recognition module 620 is used for recognizing each message to determine whether each message contains candidate time for executing target operation and whether each message belongs to target intention for negating target operation, the first determination module 630 is used for determining target message from each message, wherein the target message contains the candidate time for executing target operation and each message in the conversation after the target message does not belong to target intention, and the second determination module 640 is used for determining target time for executing target operation according to the time information for executing target operation contained in the target message.
As one possible implementation manner of the embodiment of the present disclosure, the first determining module 630 is specifically configured to determine a message including a candidate time in a conversation as a candidate message, for any one candidate message, in a case where it is determined that the message belongs to a target intention, mark candidate time included in each candidate message before the one candidate message as invalid, and use a candidate message without an invalid mark as a target message.
As one possible implementation manner of the embodiment of the present disclosure, the first determining module 630 is further configured to, in a case where there are a plurality of candidate messages without an invalid flag, take, as the target message, the candidate messages without an invalid flag that are ordered later in the conversation.
As one possible implementation of the embodiment of the disclosure, the identifying module 620 is specifically configured to identify a time text for each message, generate a time sequence according to the identified time text, where the time sequence includes time texts included in each message, each time text is ordered according to a sequence in a conversation, identify whether each time text in the time sequence belongs to a candidate time for performing a target operation, and identify whether each message belongs to a target intention.
As one possible implementation manner of the embodiment of the present disclosure, the identifying module 620 is further configured to input, into the first identifying model, a message to which the time text belongs and at least one message preceding the message to which the time text belongs in any time text in the time sequence, so as to identify whether the time text belongs to a candidate time for performing the target operation.
As one possible implementation of an embodiment of the present disclosure, the identifying module 620 is further configured to input, for any one message in the conversation, one message and at least one message preceding the one message into the second identifying model, so as to identify whether the one message belongs to the target intention.
It should be noted that the foregoing explanation of the embodiment of the method for processing time information is also applicable to the apparatus for processing time information of this embodiment, and will not be repeated here.
According to the processing device of the time information, whether each message contains the candidate time for executing the target operation or not and whether the message belongs to the target intention of the target operation or not is used for determining the target message from each message, and further, the execution time for executing the target operation is determined according to the time information of the target operation contained in the target message, so that the execution time for executing the target operation can be accurately determined.
In order to implement the above-described embodiments, the present application also proposes an electronic device, as shown in fig. 7, fig. 7 being a block diagram of an electronic device showing a processing method for time information according to an exemplary embodiment.
As shown in fig. 7, the electronic device 700 includes:
A memory 710 and a processor 720, a bus 730 connecting the different components (including the memory 710 and the processor 720), the memory 710 storing a computer program which when executed by the processor 720 implements the method of processing time information according to the embodiments of the present disclosure.
Bus 730 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures. By way of example, and not limitation, such architectures include Industry Standard Architecture (ISA) bus, micro channel architecture (MAC) bus, enhanced ISA bus, video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect (PCI) bus.
Electronic device 700 typically includes a variety of electronic device readable media. Such media can be any available media that is accessible by electronic device 700 and includes both volatile and nonvolatile media, removable and non-removable media.
Memory 710 may also include computer system readable media in the form of volatile memory, such as Random Access Memory (RAM) 740 and/or cache memory 750. Electronic device 700 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 760 may be used to read from or write to non-removable, non-volatile magnetic media (not shown in FIG. 7, commonly referred to as a "hard disk drive"). Although not shown in fig. 7, a magnetic disk drive for reading from and writing to a removable non-volatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading from or writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In such cases, each drive may be coupled to bus 730 through one or more data medium interfaces. Memory 710 may include at least one program product having a set (e.g., at least one) of program modules configured to carry out the functions of the various embodiments of the disclosure.
A program/utility 780 having a set (at least one) of program modules 770 may be stored in, for example, memory 710, such program modules 770 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment. Program modules 770 typically carry out the functions and/or methods of the embodiments described in this disclosure.
The electronic device 700 may also communicate with one or more external devices 790 (e.g., keyboard, pointing device, display, etc.), one or more devices that enable a user to interact with the electronic device 700, and/or any devices (e.g., network card, modem, etc.) that enable the electronic device 700 to communicate with one or more other computing devices. Such communication may occur through an input/output (I/O) interface 792. Also, the electronic device 700 may communicate with one or more networks such as a Local Area Network (LAN), a Wide Area Network (WAN) and/or a public network, such as the Internet, through a network adapter 793. As shown in fig. 7, the network adapter 793 communicates with other modules of the electronic device 700 over the bus 730. It should be appreciated that although not shown in FIG. 7, other hardware and/or software modules may be used in connection with electronic device 700, including, but not limited to, microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, data backup storage systems, and the like.
The processor 720 executes various functional applications and data processing by running programs stored in the memory 710.
It should be noted that, the implementation process and the technical principle of the electronic device in this embodiment refer to the foregoing explanation of the processing of the time information in the embodiment of the disclosure, and are not repeated herein.
In order to achieve the above embodiments, the present application also proposes a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the processing method of time information described in the above embodiments.
In order to implement the above embodiments, the present disclosure also provides a computer program product which, when executed by an instruction processor in the computer program product, performs the method for processing time information described in the above embodiments.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present disclosure. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present disclosure, the meaning of "a plurality" is at least two, such as two, three, etc., unless explicitly specified otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and additional implementations are included within the scope of the preferred embodiment of the present disclosure in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the embodiments of the present disclosure.
Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include an electrical connection (an electronic device) having one or more wires, a portable computer diskette (a magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer readable medium may even be paper or other suitable medium on which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.
It should be understood that portions of the present disclosure may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented in hardware as in another embodiment, may be implemented using any one or combination of techniques known in the art, discrete logic circuits with logic gates for implementing logic functions on data signals, application specific integrated circuits with appropriate combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), etc.
Those of ordinary skill in the art will appreciate that all or a portion of the steps carried out in the method of the above-described embodiments may be implemented by a program to instruct related hardware, where the program may be stored in a computer readable storage medium, and where the program, when executed, includes one or a combination of the steps of the method embodiments.
Furthermore, each functional unit in the embodiments of the present disclosure may be integrated in one processing module, or each unit may exist alone physically, or two or more units may be integrated in one module. The integrated modules may be implemented in hardware or in software functional modules. The integrated modules may also be stored in a computer readable storage medium if implemented in the form of software functional modules and sold or used as a stand-alone product.
The above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, or the like. Although embodiments of the present disclosure have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the present disclosure, and that variations, modifications, alternatives, and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the present disclosure.

Claims (12)

Translated fromChinese
1.一种时间信息的处理方法,其特征在于,包括:1. A method for processing time information, comprising:获取对话中的多条消息;Get multiple messages in a conversation;对各条所述消息进行识别,以确定各条所述消息中是否包含执行目标操作的候选时间,以及各条所述消息是否属于否定所述目标操作的目标意图;Identify each of the messages to determine whether each of the messages contains a candidate time for performing a target operation, and whether each of the messages is a target intention to negate the target operation;从各条消息中确定目标消息;其中,所述目标消息包含执行所述目标操作的候选时间,且所述对话中处于所述目标消息之后的各条消息不属于所述目标意图;Determine a target message from each message; wherein the target message includes a candidate time for performing the target operation, and each message after the target message in the conversation does not belong to the target intention;根据所述目标消息所包含的执行所述目标操作的时间信息,确定执行所述目标操作的目标时间;Determining a target time for executing the target operation according to the time information for executing the target operation contained in the target message;其中,所述从各条消息中确定目标消息,包括:Wherein, determining the target message from each message includes:将所述对话中包含所述候选时间的消息确定为候选消息;Determine the message in the conversation containing the candidate time as a candidate message;针对任意的一条候选消息,在确定属于所述目标意图的情况下,将所述一条候选消息之前的各条候选消息所含的候选时间标记为无效;For any candidate message, when it is determined that the candidate message belongs to the target intention, the candidate time contained in each candidate message before the candidate message is marked as invalid;将未具有无效标记的候选消息作为所述目标消息。The candidate message not having an invalid flag is used as the target message.2.根据权利要求1所述的方法,其特征在于,所述将未具有所述无效标记的候选消息作为所述目标消息,包括:2. The method according to claim 1, characterized in that the step of taking the candidate message without the invalid mark as the target message comprises:在未具有所述无效标记的候选消息为多个的情况下,将所述对话中排序在后的所述未具有所述无效标记的候选消息作为所述目标消息。In the case that there are multiple candidate messages that do not have the invalid mark, the candidate messages that do not have the invalid mark and are ranked later in the conversation are used as the target messages.3.根据权利要求1-2任一项所述的方法,其特征在于,所述对各条所述消息进行识别,以确定各条所述消息中是否包含执行目标操作的候选时间,以及各条所述消息是否属于否定所述目标操作的目标意图,包括:3. The method according to any one of claims 1-2, characterized in that the identifying of each of the messages to determine whether each of the messages contains a candidate time for performing a target operation and whether each of the messages belongs to a target intention of negating the target operation comprises:对各条所述消息识别时间文本;Identifying a time text for each of the messages;根据识别出的时间文本,生成时间序列,其中,所述时间序列包括各条消息中所含的时间文本,各所述时间文本依据所述对话中的顺序排序;Generate a time sequence according to the identified time text, wherein the time sequence includes the time text contained in each message, and each of the time texts is sorted according to the order in the conversation;对所述时间序列中的各时间文本识别是否属于执行所述目标操作的候选时间;以及,对各条所述消息识别是否属于所述目标意图。Identify whether each time text in the time sequence belongs to a candidate time for executing the target operation; and identify whether each message belongs to the target intention.4.根据权利要求3所述的方法,其特征在于,所述对所述时间序列中的各时间文本识别是否属于执行所述目标操作的候选时间,包括:4. The method according to claim 3, characterized in that the step of identifying whether each time text in the time sequence belongs to a candidate time for executing the target operation comprises:对所述时间序列中的任意的一条时间文本,将所述一条时间文本所属的消息及其之前的至少一条消息输入第一识别模型,以识别所述一条时间文本是否属于执行所述目标操作的候选时间。For any time text in the time sequence, the message to which the time text belongs and at least one message before the time text are input into the first recognition model to identify whether the time text belongs to a candidate time for executing the target operation.5.根据权利要求3所述的方法,其特征在于,所述对各条所述消息识别是否属于所述目标意图,包括:5. The method according to claim 3, characterized in that the step of identifying whether each message belongs to the target intention comprises:对所述对话中任意的一条消息,将所述一条消息以及所述一条消息之前的至少一条消息输入第二识别模型,以识别所述一条消息是否属于所述目标意图。For any message in the conversation, the message and at least one message before the message are input into a second recognition model to identify whether the message belongs to the target intent.6.一种时间信息的处理装置,其特征在于,包括:6. A time information processing device, characterized in that it comprises:获取模块,用于获取对话中的多条消息;The acquisition module is used to obtain multiple messages in a conversation;识别模块,用于对各条所述消息进行识别,以确定各条所述消息中是否包含执行目标操作的候选时间,以及各条所述消息是否属于否定所述目标操作的目标意图;an identification module, used to identify each of the messages to determine whether each of the messages contains a candidate time for performing a target operation, and whether each of the messages belongs to a target intention of negating the target operation;第一确定模块,用于从各条消息中确定目标消息;其中,所述目标消息包含执行所述目标操作的候选时间,且所述对话中处于所述目标消息之后的各条消息不属于所述目标意图;A first determination module is used to determine a target message from each message; wherein the target message includes a candidate time for performing the target operation, and each message after the target message in the conversation does not belong to the target intention;第二确定模块,用于根据所述目标消息所包含的执行所述目标操作的时间信息,确定执行所述目标操作的目标时间;A second determining module, configured to determine a target time for executing the target operation according to the time information for executing the target operation contained in the target message;其中,所述第一确定模块,具体用于:Wherein, the first determining module is specifically used to:将所述对话中包含所述候选时间的消息确定为候选消息;Determine the message in the conversation containing the candidate time as a candidate message;针对任意的一条候选消息,在确定属于所述目标意图的情况下,将所述一条候选消息之前的各条候选消息所含的候选时间标记为无效;For any candidate message, when it is determined that the candidate message belongs to the target intention, the candidate time contained in each candidate message before the candidate message is marked as invalid;将未具有无效标记的候选消息作为所述目标消息。The candidate message not having an invalid flag is used as the target message.7.根据权利要求6所述的装置,其特征在于,所述第一确定模块,还用于:7. The device according to claim 6, wherein the first determining module is further used to:在未具有所述无效标记的候选消息为多个的情况下,将所述对话中排序在后的所述未具有所述无效标记的候选消息作为所述目标消息。In the case that there are multiple candidate messages that do not have the invalid mark, the candidate messages that do not have the invalid mark and are ranked later in the conversation are used as the target messages.8.根据权利要求6-7任一项所述的装置,其特征在于,所述识别模块,具体用于:8. The device according to any one of claims 6 to 7, characterized in that the identification module is specifically used to:对各条所述消息识别时间文本;Identifying a time text for each of the messages;根据识别出的时间文本,生成时间序列,其中,所述时间序列包括各条消息中所含的时间文本,各所述时间文本依据所述对话中的顺序排序;Generate a time sequence according to the identified time text, wherein the time sequence includes the time text contained in each message, and each of the time texts is sorted according to the order in the conversation;对所述时间序列中的各时间文本识别是否属于执行所述目标操作的候选时间;以及,对各条所述消息识别是否属于所述目标意图。Identify whether each time text in the time sequence belongs to a candidate time for executing the target operation; and identify whether each message belongs to the target intention.9.根据权利要求8所述的装置,其特征在于,所述识别模块,还用于:9. The device according to claim 8, characterized in that the identification module is further used for:对所述时间序列中的任意的一条时间文本,将所述一条时间文本所属的消息及其之前的至少一条消息输入第一识别模型,以识别所述一条时间文本是否属于执行所述目标操作的候选时间。For any time text in the time sequence, the message to which the time text belongs and at least one message before the time text are input into the first recognition model to identify whether the time text belongs to a candidate time for executing the target operation.10.根据权利要求8所述的装置,其特征在于,所述识别模块,还用于:10. The device according to claim 8, characterized in that the identification module is further used for:对所述对话中任意的一条消息,将所述一条消息以及所述一条消息之前的至少一条消息输入第二识别模型,以识别所述一条消息是否属于所述目标意图。For any message in the conversation, the message and at least one message before the message are input into a second recognition model to identify whether the message belongs to the target intent.11.一种电子设备,其特征在于,包括:11. An electronic device, comprising:至少一个处理器;以及at least one processor; and与所述至少一个处理器通信连接的存储器;其中,a memory communicatively connected to the at least one processor; wherein,所述存储器存储有可被所述至少一个处理器执行的指令,所述指令被所述至少一个处理器执行,以使所述至少一个处理器能够执行权利要求1-5中任一项所述的方法。The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 5.12.一种存储有计算机指令的非瞬时计算机可读存储介质,其特征在于,所述计算机指令用于使所述计算机执行权利要求1-5中任一项所述的方法。12. A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 5.
CN202111217421.6A2021-10-192021-10-19 Time information processing method, device, electronic device and storage mediumActiveCN113887210B (en)

Priority Applications (1)

Application NumberPriority DateFiling DateTitle
CN202111217421.6ACN113887210B (en)2021-10-192021-10-19 Time information processing method, device, electronic device and storage medium

Applications Claiming Priority (1)

Application NumberPriority DateFiling DateTitle
CN202111217421.6ACN113887210B (en)2021-10-192021-10-19 Time information processing method, device, electronic device and storage medium

Publications (2)

Publication NumberPublication Date
CN113887210A CN113887210A (en)2022-01-04
CN113887210Btrue CN113887210B (en)2025-02-21

Family

ID=79003675

Family Applications (1)

Application NumberTitlePriority DateFiling Date
CN202111217421.6AActiveCN113887210B (en)2021-10-192021-10-19 Time information processing method, device, electronic device and storage medium

Country Status (1)

CountryLink
CN (1)CN113887210B (en)

Citations (1)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN102792320A (en)*2010-01-182012-11-21苹果公司Intelligent automated assistant

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN104951428B (en)*2014-03-262019-04-16阿里巴巴集团控股有限公司User's intension recognizing method and device
CN109032381B (en)*2018-08-012022-05-17平安科技(深圳)有限公司Input method and device based on context, storage medium and terminal
CN109960801A (en)*2019-03-152019-07-02北京字节跳动网络技术有限公司Data processing method and device
CN111917878B (en)*2020-08-032023-01-10腾讯科技(深圳)有限公司Message processing method, device, equipment and storage medium
CN113469711B (en)*2021-07-202024-11-12阳光保险集团股份有限公司 Intelligent customer service interaction method, device, electronic device and storage medium

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
CN102792320A (en)*2010-01-182012-11-21苹果公司Intelligent automated assistant

Also Published As

Publication numberPublication date
CN113887210A (en)2022-01-04

Similar Documents

PublicationPublication DateTitle
CN108682420B (en)Audio and video call dialect recognition method and terminal equipment
US11947872B1 (en)Natural language processing platform for automated event analysis, translation, and transcription verification
CN113205814B (en)Voice data labeling method and device, electronic equipment and storage medium
CN112560453B (en)Voice information verification method and device, electronic equipment and medium
CN111292751B (en)Semantic analysis method and device, voice interaction method and device, and electronic equipment
CN112527994A (en)Emotion analysis method, emotion analysis device, emotion analysis equipment and readable storage medium
CN112699671B (en)Language labeling method, device, computer equipment and storage medium
CN113158656B (en)Ironic content recognition method, ironic content recognition device, electronic device, and storage medium
CN112201253B (en)Text marking method, text marking device, electronic equipment and computer readable storage medium
CN113807096B (en)Text data processing method, device, computer equipment and storage medium
CN112951233A (en)Voice question and answer method and device, electronic equipment and readable storage medium
CN112288584A (en)Insurance application processing method and device, computer readable medium and electronic equipment
CN112151019A (en)Text processing method and device and computing equipment
CN116453023B (en)Video abstraction system, method, electronic equipment and medium for 5G rich media information
CN114281995A (en)Premium payment hastening call extraction and analysis method, device, equipment and medium
CN118051600A (en) Method, device, storage medium and electronic device for constructing multimodal knowledge based on large model
CN111507114B (en)Reverse translation-based spoken language text enhancement method and system
CN114462364B (en)Method and device for inputting information
CN112149680A (en)Wrong word detection and identification method and device, electronic equipment and storage medium
WO2025130888A1 (en)Audio processing method and apparatus, device and storage medium
CN113887210B (en) Time information processing method, device, electronic device and storage medium
CN117579888A (en)Video subtitle extraction method, device, equipment and storage medium
CN115269805A (en) Method, device and electronic device for constructing question and answer knowledge base
CN116127011A (en) Intent recognition method, device, electronic device and storage medium
CN115186656A (en)Social text word segmentation method and device

Legal Events

DateCodeTitleDescription
PB01Publication
PB01Publication
SE01Entry into force of request for substantive examination
SE01Entry into force of request for substantive examination
GR01Patent grant
GR01Patent grant

[8]ページ先頭

©2009-2025 Movatter.jp