Time standardization method and device, equipment, storage medium and program product

Abstract

The invention relates to a time standardization method and device, equipment, a storage medium and a program product. The method comprises the steps that semantic analysis is carried out on a user instruction, time types corresponding to all parts of time content in the user instruction are recognized, all the parts of the time content represent time positioning information of different dimensions, then standardization processing is carried out on all the parts of the time content according to the time types, and standardized time is obtained. According to the scheme provided by the invention, the spoken user instruction can be standardized, and the accuracy of time identification is improved.

Description

Technical Field

[0001] The present application relates to the technical field of time standardization processing, and in particular to a time standardization method, apparatus, device, storage medium, and program product. Background Art

[0002] In related technologies, when voice assistants execute commands such as weather queries, itinerary planning, and smart reminders, they perform information queries or related operations based on the date specified by the user. However, since users often use colloquial time expressions, the accuracy of time recognition is low, and the corresponding queries or related operations cannot be effectively executed.

[0003] Therefore, how to improve the accuracy of time recognition has become a technical problem that needs to be solved urgently by those skilled in the art. Summary of the Invention

[0004] In order to solve or partially solve the problems existing in the related art, the present application provides a time standardization method, device, equipment, storage medium and program product, which can improve the accuracy of time recognition.

[0005] The first aspect of the present application provides a time standardization method, comprising: By performing semantic analysis on the user instruction, identifying the time type corresponding to each part of the time content in the user instruction; each part of the time content represents time positioning information of different dimensions; Each part of the time content is standardized according to the time type to obtain standardized time.

[0006] Furthermore, in the above method, the step of performing semantic analysis on the user instruction to identify the time type corresponding to each part of the time content in the user instruction includes: The user instruction is input into a pre-trained time type recognition model, so that the time type recognition model performs semantic analysis on the user instruction, identifies and outputs the time type corresponding to each part of the time content in the user instruction.

[0007] Furthermore, in the above method, the time type recognition model is trained with the first user instruction containing time content as a positive sample and the second user instruction not containing time content as a negative sample, with the goal of identifying the time types corresponding to each part of the time content in the positive sample.

[0008] Furthermore, in the above method, the time type recognition model includes a BERT model.

[0009] Furthermore, in the above method, the step of normalizing the various parts of the time content according to the time type to obtain the standardized time includes: performing normalization processing on each part of the time content according to the conversion logic corresponding to the time type to obtain normalization results corresponding to each part of the time content; The normalized results corresponding to the various parts of the time content are combined together to obtain the normalized time.

[0010] Furthermore, in the above method, the standardization of each part of the time content according to the conversion logic corresponding to the time type includes: For the first part of the time content whose time type is a relative time type, determining the sum of the relative offset days and the current date as the normalized result corresponding to the first part; and / or, for a second portion of the time content whose time type is an absolute time type, determining the absolute time as a normalization result corresponding to the second portion; and / or, for the third part of the time content whose time type is the lunar calendar time type, determining the Gregorian calendar time corresponding to the lunar calendar time as the normalization result corresponding to the third part; And / or, for the fourth part of the time content whose time type is a time period type, the time range corresponding to the time period is determined as the standardized result corresponding to the fourth part.

[0011] A second aspect of the present application provides a time standardization device, comprising: an identification module configured to perform semantic analysis on a user instruction to identify a time type corresponding to each part of a time content in the user instruction; each part of the time content represents time location information of different dimensions; The standardization module is used to standardize each part of the time content according to the time type to obtain standardized time.

[0012] A third aspect of the present application provides an electronic device, including: processor; and The memory stores executable codes thereon, and when the executable codes are executed by the processor, the processor is caused to execute the method described above.

[0013] A fourth aspect of the present application provides a computer-readable storage medium having executable code stored thereon. When the executable code is executed by a processor of an electronic device, the processor is caused to execute the method described above.

[0014] A fifth aspect of the present application provides a computer program product, which includes computer instructions, and when the computer instructions are executed by a processor, implements the method described above.

[0015] The technical solution provided by this application may include the following beneficial results: The technical solution provided in this application performs semantic analysis on user instructions to identify the time type corresponding to each part of the time content in the user instructions. Each part of the time content represents time positioning information of different dimensions. Then, each part of the time content is standardized according to the time type to obtain standardized time. In this way, spoken user instructions can be standardized to improve the accuracy of time recognition.

[0016] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The above and other objects, features and advantages of the present application will become more apparent by describing in more detail exemplary embodiments of the present application in conjunction with the accompanying drawings, wherein like reference numerals generally represent like components in the exemplary embodiments of the present application.

[0018] Figure 1 Schematic diagram of a feasible application scenario of the time standardization method shown in the embodiment of the present application; Figure 2 1 is a flow chart of a time standardization method shown in an embodiment of the present application; Figure 3 Schematic diagram of the structure of a time standardization device shown in an embodiment of the present application; Figure 4 It is a structural diagram of an electronic device shown in an embodiment of the present application. DETAILED DESCRIPTION

[0019] The following describes embodiments of the present application in more detail with reference to the accompanying drawings. Although the accompanying drawings illustrate embodiments of the present application, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. Rather, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0020] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. As used in this application and the appended claims, the singular forms "a," "an," "the," and "the" are intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0021] It should be understood that although the terms "first", "second", "third", etc. may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of this application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0022] In related technologies, voice assistants perform information queries or related operations based on the date specified in the user's instructions when executing commands such as weather query, itinerary planning, and smart reminders. However, users often use colloquial time expressions, resulting in low time recognition accuracy and inability to effectively execute the corresponding query or related operations.

[0023] Furthermore, the time description in user commands often only constitutes a portion of the instruction, requiring identification and standardization of the time portion. This process also requires adaptation to diverse expressions such as the lunar calendar, absolute time, and relative time. However, current technologies for standardizing spoken time have significant limitations. They are not only poorly scalable, but also ineffective at handling unconventional or highly colloquial time expressions.

[0024] In response to the above problems, an embodiment of the present application provides a time standardization method, which can standardize spoken user instructions and improve the accuracy of time recognition.

[0025] The technical solutions of the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0026] Figure 1 The following shows a possible application scenario of the time standardization method, such as Figure 1 In the scenario shown, a client and a server are set up.

[0027] The client can be any electronic device with network access capabilities. Specifically, for example, the client can be a desktop computer, tablet computer, laptop computer, smartphone, digital assistant, smart wearable device, shopping guide terminal, television, etc. Smart wearable devices include, but are not limited to, smart bracelets, smart watches, smart glasses, smart helmets, smart necklaces, etc. Alternatively, the client can be software that can run on an electronic device.

[0028] A server can be an electronic device with certain computing and processing capabilities. It can include a network communication module, a processor, and memory. Of course, a server can also refer to software running on an electronic device. A server can also be a distributed server, a system with multiple processors, memories, network communication modules, and the like operating in coordination. Alternatively, a server can be a server cluster formed by multiple servers. Alternatively, with the advancement of science and technology, a server can also be a new technological means capable of implementing the functions described in the embodiments of this specification. For example, it can be a new form of "server" based on quantum computing.

[0029] The client and server can communicate via a target network. The target network can be any type of network. For example, the target network can be a single network that can be subdivided into multiple subnetworks. The target network, or multiple subnetworks contained within the target network, can specifically be at least one of a cellular mobile network (e.g., 2G, 3G, 4G, or 5G), Zigbee, Wi-Fi, or Bluetooth, or any combination of at least one of these networks and other networks.

[0030] In the above-mentioned feasible application scenario, the client receives user instructions and sends them to the server. The server then performs semantic analysis on the user instructions to identify the time type corresponding to each time component in the instruction. Each time component represents time location information of different dimensions. The server then normalizes each time component according to the time type to obtain a standardized time. This setup allows for the standardization of spoken user instructions, improving the accuracy of time recognition.

[0031] Furthermore, the embodiments of the present application provide a time standardization method that can be executed by an electronic device. The electronic device can be any device with data and instruction processing capabilities, such as a laptop, tablet computer, desktop computer, mobile device (e.g., mobile phone, personal digital assistant, dedicated messaging device) and other types of user terminals, or a combination of any two or more of these electronic devices, or a server. Figure 2 As shown, the method includes: S101 : Identify the time type corresponding to each part of the time content in the user instruction by performing semantic analysis on the user instruction.

[0032] The user instructions described above refer to requests issued to the intelligent system in natural language to cause it to perform specific tasks. The intelligent system includes voice assistants or smart devices, and this embodiment does not limit this. Natural language includes spoken language or text, and this embodiment does not limit this.

[0033] A user command may or may not contain a time element. For example, "Remind me to get a haircut at 8:00 AM tomorrow" is a user command that contains a time element, while "Close the curtains" is a user command that does not contain a time element.

[0034] In the embodiments of the present application, semantic analysis is performed on the user instruction to identify the time type corresponding to each part of the time content in the user instruction. More specifically, if the user instruction includes time content, the identification result is the time type corresponding to each part of the time content in the user instruction; if the user instruction does not include time content, the identification result is empty.

[0035] The various parts of the aforementioned time content represent time location information of different dimensions within the time content. These dimensions may include date, time period, and moment dimensions. For example, "tomorrow at 8:00 AM" includes "tomorrow," "morning," and "8:00 AM," while "next Wednesday evening" includes "next Wednesday" and "evening." "Tomorrow" and "next Wednesday" represent time location information in the date dimension, "morning" and "evening" represent time location information in the time period dimension, and "8:00 AM" represents time location information in the moment dimension.

[0036] The above-mentioned time types include relative time type, absolute time type, time period type and special time type, among which special time includes special time terms such as lunar calendar time type, week type, quarter type, etc.

[0037] Among them, the relative time type refers to the type of dynamic time expression based on the current time and calculated by offset, such as "three weeks later", "tomorrow", "the day after tomorrow", etc.; the absolute time type refers to the type of fixed time point or date independent of the current time, which can be directly parsed into machine format, such as "May 20, 2025", "13:14 on May 20, 2026", etc.; the time period type is used to describe a continuous interval within a day rather than an exact moment, and needs to be mapped to a time range, such as "morning", "afternoon", "evening", etc.; the special time type refers to a complex time expression type that requires calling external rules or resources, such as the lunar calendar conversion library and quarterly calculation table, to be standardized, such as "the 23rd day of the twelfth lunar month" and "the end of Q3".

[0038] For example, "tomorrow morning at 8 o'clock" includes three parts: "tomorrow", "morning" and "eight o'clock". The time type corresponding to "tomorrow" is a relative time type, the time type corresponding to "morning" is a time period type, and the time type corresponding to "eight o'clock" is an absolute time type; "next Wednesday evening" includes two parts: "next Wednesday" and "evening". The time type corresponding to "next Wednesday" is a special time type, and the time type corresponding to "evening" is a time period type.

[0039] In the embodiments of the present application, semantic analysis is performed on user commands to identify the time types corresponding to the various time content components in the user commands. More specifically, a time category recognition model can be trained and used to perform semantic analysis on user commands to obtain the time types corresponding to the various time content components in the user commands.

[0040] A large number of user instructions can be obtained as training samples, and the time types corresponding to the various parts of the time content in the user instructions can be marked as training labels. During training, the training samples are input into the time category recognition model to obtain the prediction results output by the time category recognition model. By comparing the prediction results output by the time category recognition model with the training labels, the loss value of the time category recognition model is determined. With the goal of reducing the loss value of the time category recognition model, the parameters of the time category recognition model are adjusted until the parameters of the time category recognition model meet the requirements and the time category recognition model training is completed.

[0041] The temporal category recognition model can use a large language model. To reduce time consumption, a lightweight BERT (Bidirectional Encoder Representations from Transformers) model can also be used.

[0042] The BERT model can be used to perform named entity recognition (NER) classification tasks to identify the time types corresponding to the various parts of the time content in user instructions.

[0043] The user instruction is input into a pre-trained time type recognition model so that the time type recognition model performs semantic analysis on the user instruction, identifies and outputs the time type corresponding to each part of the time content in the user instruction.

[0044] S102: Standardize each part of the time content according to the time type to obtain standardized time.

[0045] After determining the time type corresponding to each part of the time content in the user instruction, each part of the time content is standardized according to the time type to obtain a standardized time.

[0046] The above-mentioned standardization process converts the time content into a unified, accurate and computable machine format, thereby supporting the efficient operation of the upper-level business logic.

[0047] In some embodiments, a pre-trained normalization model can be used to normalize the various parts of the time content. A large amount of time content can be obtained, and the various parts of the time content and the time types corresponding to the various parts of the time content can be used as training samples, and the results of the normalization of the training samples are used as training labels. During training, the training samples are input into the normalization model to obtain the prediction results output by the normalization model. By comparing the prediction results output by the normalization model and the training labels, the loss value of the normalization model is determined. The parameters of the normalization model are adjusted with the goal of reducing the loss value of the normalization model until the various parameters of the normalization model meet the requirements and the normalization model training is completed. The above-mentioned normalization model can be obtained based on training of any neural network model, or based on training of a pre-trained model, such as a pre-trained large model similar to ChatGPT.

[0048] Input any part of the time content and the time type corresponding to that part into the trained normalization model to obtain the normalization result for that part output by the normalization model. Combine the normalization results corresponding to each part of the time content to obtain the normalized time.

[0049] For example, let's say the user input command is "Submit a report at 3:00 PM the day after tomorrow." The time content in the user command includes "the day after tomorrow," "the afternoon," and "3:00 PM." If the current date is July 1, 2025, after processing by the normalization model, "the day after tomorrow" is normalized to "July 3, 2025," "the afternoon" is normalized to "12:00-18:00," and "3:00 PM" is normalized to "3:00 PM." The normalization results corresponding to each part of the time content are combined to obtain the normalized time "3:00 PM, July 3, 2025."

[0050] In the above embodiment, semantic analysis is performed on the user instructions to identify the time types corresponding to the various parts of the time content in the user instructions. The various parts of the time content represent time positioning information of different dimensions. Then, the various parts of the time content are standardized according to the time type to obtain standardized time. In this way, the spoken user instructions can be standardized to improve the accuracy of time recognition.

[0051] As an optional implementation, the time type recognition model in the above embodiment is trained with the first user instruction containing time content as the positive sample and the second user instruction not containing time content as the negative sample, with the goal of identifying the time types corresponding to each part of the time content in the positive sample.

[0052] In addition to training the time type recognition model according to the description of the above embodiment, the embodiment of the present application also provides a training method to improve the accuracy of the recognition results of the time type recognition model. The specific training process is as follows: A large number of first user instructions containing time content are obtained as positive samples, such as "Submit the report at 3 pm the day after tomorrow", "Steam buns three days later", "Open the curtains at 9 o'clock", etc.; second user instructions that do not contain time content are obtained as negative samples, such as "Turn on the air conditioner", "Turn off the water heater", "Lower the car windows", "Close the car doors", etc.

[0053] The training labels for negative samples are empty, while the training labels for positive samples are the time types corresponding to the various parts of the first user instruction. Data can be annotated using the BIO (Begin, Inside, Outside) annotation method, which is used for named entity recognition tasks and delineates entity boundaries through sequence annotation.

[0054] In some embodiments, specific marking instructions can be referred to Table 1.

[0055]

[0056] Table 1 During training, positive and negative samples are input into the temporal classification model to obtain predictions. The model's loss is determined by comparing the predictions with the corresponding training labels. The model's parameters are then adjusted to minimize this loss until all parameters meet the requirements. Training is complete. Comparative ablation experiments can also be conducted to improve model generalization.

[0057] In the above embodiment, the time category recognition model is trained by using positive samples and negative samples, which can improve the accuracy of the recognition result of the time category recognition model.

[0058] As an optional embodiment, the steps of the above embodiment standardize various parts of the time content according to the time type to obtain the standardized time, which may specifically include the following steps: According to the conversion logic corresponding to the time type, each part of the time content is standardized to obtain a standardized result corresponding to each part of the time content; the standardized results corresponding to each part of the time content are combined together to obtain the standardized time.

[0059] In the embodiment of the present application, different time types correspond to different conversion logics.

[0060] More specifically, for the first portion of the time content where the time type is relative, the sum of the relative offset days and the current date is determined as the normalized result for the first portion. For example, if the input instruction is "Please submit the report in three days", the time type corresponding to "three days later" is relative, the relative offset days is three, and if the current date is July 1, 2025, the normalized result for this portion is July 4, 2025.

[0061] For the second part of the time content where the time type is absolute, the absolute time is determined as the normalized result corresponding to the second part. For example, if the input instruction is "Submit the report at 3:00 PM tomorrow," and the time type corresponding to "3:00" is absolute, the normalized result corresponding to this part is 15:00.

[0062] For the third part of the time content that has a lunar calendar time type, the Gregorian calendar time corresponding to the lunar calendar time is determined as the standardized result corresponding to the third part. For example, if the input instruction is "Remind me to make dumplings on New Year's Eve this year", and the time type corresponding to "New Year's Eve" is a lunar calendar time type, the lunar calendar conversion library can be called to determine that the Gregorian calendar time corresponding to "New Year's Eve" is "February 16, 2026", and the standardized result corresponding to this part is February 16, 2026.

[0063] For the fourth part of the time content, where the time type is a time period, the time range corresponding to the time period is determined as the standardized result corresponding to the fourth part. For example, if the input instruction is "Submit the report at 3:00 PM tomorrow," and the time type corresponding to "PM" is a time range, the standardized result corresponding to this part is 12:00-18:00.

[0064] By combining the normalized results corresponding to each part of the time content, the normalized time can be obtained.

[0065] For example, let's say the user input command is "Go wash the car tomorrow at 6:00 PM." The time content in the user command includes "tomorrow," "afternoon," and "6:00 PM." If the current date is July 1, 2025, after processing through the normalization model, "tomorrow" is normalized to "July 1, 2025," "afternoon" is normalized to "12:00-18:00," and "6:00 PM" is normalized to "6:00 PM." The normalization results corresponding to each part of the time content are combined to obtain the normalized time "July 2, 2025, 6:00 PM."

[0066] In the above embodiment, various parts of the time content can be quickly standardized according to the conversion logic corresponding to the time type, thereby effectively improving the processing efficiency.

[0067] Corresponding to the aforementioned application function implementation method embodiment, the present application also provides a time standardization device, electronic device, computer-readable storage medium, computer program product and corresponding embodiments.

[0068] Figure 3 It is a structural diagram of a time standardization device shown in an embodiment of the present application.

[0069] See also Figure 3 The time standardization device of the above embodiment includes: The identification module 100 is used to identify the time type corresponding to each part of the time content in the user instruction by performing semantic analysis on the user instruction; each part of the time content represents time positioning information of different dimensions; The standardization module 110 is used to standardize each part of the time content according to the time type to obtain a standardized time.

[0070] Furthermore, the recognition module 100 of the above embodiment, when performing semantic analysis on the user instruction to identify the time type corresponding to each part of the time content in the user instruction, is specifically configured to: The user instruction is input into a pre-trained time type recognition model so that the time type recognition model performs semantic analysis on the user instruction, identifies and outputs the time type corresponding to each part of the time content in the user instruction.

[0071] Furthermore, the time type recognition model of the above embodiment is trained with the first user instruction containing time content as the positive sample and the second user instruction not containing time content as the negative sample, with the goal of identifying the time type corresponding to each part of the time content in the positive sample.

[0072] Furthermore, the time type recognition model in the above embodiment includes a BERT model.

[0073] Furthermore, the standardization module 110 of the above embodiment, when performing standardization processing on each part of the time content according to the time type to obtain the standardized time, is specifically used to: According to the conversion logic corresponding to the time type, each part of the time content is standardized to obtain the standardized results corresponding to each part of the time content; the standardized results corresponding to each part of the time content are combined together to obtain the standardized time.

[0074] Furthermore, the standardization module 110 of the above embodiment, when performing standardization processing on each part of the time content according to the conversion logic corresponding to the time type to obtain the standardization results corresponding to each part of the time content, is specifically used to: For the first part of the time content whose time type is a relative time type, the sum of the relative offset days and the current date is determined as the standardized result corresponding to the first part; and / or, for the second part of the time content whose time type is an absolute time type, the absolute time is determined as the standardized result corresponding to the second part; and / or, for the third part of the time content whose time type is a lunar time type, the Gregorian calendar time corresponding to the lunar time is determined as the standardized result corresponding to the third part; and / or, for the fourth part of the time content whose time type is a time period type, the time range corresponding to the time period is determined as the standardized result corresponding to the fourth part.

[0075] Regarding the apparatus in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated again here.

[0076] Figure 4 It is a structural diagram of an electronic device shown in an embodiment of the present application.

[0077] See also Figure 4 , the electronic device includes a memory 200 and a processor 210; The memory 200 is connected to the processor 210 and is used to store programs; The processor 210 is configured to implement part or all of the above-mentioned methods by running the program stored in the memory 200 .

[0078] Specifically, the electronic device may further include: a bus, a communication interface 220 , an input device 230 and an output device 240 .

[0079] The processor 210, the memory 200, the communication interface 220, the input device 230 and the output device 240 are interconnected via a bus. A bus may include a pathway that transfers information between components of a computer system.

[0080] The processor 210 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0081] The processor 210 may include a main processor, and may also include a baseband chip, a modem, and the like.

[0082] Memory 200 may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage. ROM may store static data or instructions required by processor 210 or other computer modules. Permanent storage may be a readable and writable storage device. Permanent storage may be a non-volatile storage device that retains stored instructions and data even when the computer is powered off. In some embodiments, the permanent storage device utilizes a mass storage device (e.g., a magnetic or optical disk, flash memory). In other embodiments, the permanent storage device may be a removable storage device (e.g., a floppy disk, optical drive). System memory may be a readable and writable storage device or a volatile readable and writable storage device, such as dynamic random access memory (DRAM). System memory may store some or all instructions and data required by the processor during operation. Furthermore, memory 200 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (e.g., DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), as well as magnetic disks and / or optical disks. In some embodiments, the memory 200 may include a readable and / or writable removable storage device, such as a compact disc (CD), a read-only digital versatile disc (e.g., DVD-ROM, double-layer DVD-ROM), a read-only Blu-ray disc, an ultra-density optical disc, a flash memory card (e.g., SD card, mini SD card, Micro-SD card, etc.), a magnetic floppy disk, etc. Computer-readable storage media do not include carrier waves and transient electronic signals transmitted wirelessly or wired.

[0083] The input device 230 may include a device for receiving data and information input by a user, such as a keyboard, a mouse, a camera, a scanner, a light pen, a voice input device, a touch screen, a pedometer, or a gravity sensor.

[0084] Output device 240 may include devices that allow information to be output to a user, such as a display screen, printer, speakers, etc.

[0085] The communication interface 220 may include any transceiver or similar device for communicating with other devices or communication networks, such as Ethernet, Radio Access Network (RAN), Wireless Local Area Network (WLAN), etc.

[0086] The processor 210 executes the program stored in the memory 200 and calls other devices to implement part or all of the methods described above.

[0087] Furthermore, the method according to the present application may also be implemented as a computer program product, which includes computer program code instructions for executing some or all of the steps of the method described above. Optionally, the computer program may be stored on a computer-readable storage medium or in the cloud; the computer device's processor reads the computer program from the computer-readable storage medium or the cloud.

[0088] The computer program product may be written in any combination of one or more programming languages ​​to implement the program code of the embodiments of the present application, including object-oriented programming languages ​​such as Java, C++, and conventional procedural programming languages ​​such as C or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0089] The computer program product may be implemented in hardware, software, or a combination thereof. In one embodiment, the computer program product is implemented as a computer storage medium. In another embodiment, the computer program product is implemented as a software product, such as a software development kit (SDK).

[0090] Alternatively, the present application can also be implemented as a computer-readable storage medium (or non-transitory machine-readable storage medium or machine-readable storage medium), which stores executable code (or computer program or computer instruction code) and, when executed by a processor of an electronic device (or server, etc.), enables the processor to perform part or all of the steps of the above-mentioned method according to the present application.

[0091] The computer-readable storage medium may be any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0092] The embodiments of the present application have been described above. The above description is illustrative, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or improvements to the technology in the market, or to enable other persons skilled in the art to understand the embodiments disclosed herein.

Claims

1. A time standardization method, characterized in that: include: By performing semantic analysis on the user instruction, identifying the time type corresponding to each part of the time content in the user instruction; each part of the time content represents time positioning information of different dimensions; Each part of the time content is standardized according to the time type to obtain standardized time.

2. The time standardization method according to claim 1, characterized in that: The step of performing semantic analysis on the user instruction to identify the time type corresponding to each part of the time content in the user instruction includes: The user instruction is input into a pre-trained time type recognition model, so that the time type recognition model performs semantic analysis on the user instruction, identifies and outputs the time type corresponding to each part of the time content in the user instruction.

3. The time standardization method according to claim 2, characterized in that: The time type recognition model is trained with a first user instruction containing time content as a positive sample and a second user instruction not containing time content as a negative sample, with the goal of identifying the time types corresponding to each part of the time content in the positive sample.

4. The time standardization method according to claim 2, characterized in that: The time type recognition model includes a BERT model.

5. The time standardization method according to claim 1, characterized in that: The step of normalizing each part of the time content according to the time type to obtain the standardized time includes: performing normalization processing on each part of the time content according to the conversion logic corresponding to the time type to obtain normalization results corresponding to each part of the time content; The normalized results corresponding to the various parts of the time content are combined together to obtain the normalized time.

6. The time standardization method according to claim 5, characterized in that: The standardization of each part of the time content according to the conversion logic corresponding to the time type includes: For the first part of the time content whose time type is a relative time type, determining the sum of the relative offset days and the current date as the normalized result corresponding to the first part; and / or, for a second portion of the time content whose time type is an absolute time type, determining the absolute time as a normalization result corresponding to the second portion; and / or, for the third part of the time content whose time type is the lunar calendar time type, determining the Gregorian calendar time corresponding to the lunar calendar time as the normalization result corresponding to the third part; And / or, for the fourth part of the time content whose time type is a time period type, the time range corresponding to the time period is determined as the standardized result corresponding to the fourth part.

7. A time standardization device, characterized in that: include: an identification module, configured to identify the time type corresponding to each part of the time content in the user instruction by performing semantic analysis on the user instruction; Each part of the time content represents time positioning information of different dimensions; The standardization module is used to standardize each part of the time content according to the time type to obtain standardized time.

8. An electronic device, characterized in that: include: processor; as well as A memory having executable codes stored thereon, which, when executed by the processor, causes the processor to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that An executable code is stored thereon, and when the executable code is executed by a processor of an electronic device, the processor is caused to execute the method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The computer program product comprises computer instructions, and when the computer instructions are executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images