Weather information providing method and device

Abstract

A weather information providing method performed by a weather information providing device according to an embodiment comprises the steps of: collecting weather information; generating a natural language response by inputting a natural language request to a trained language model when the natural language request for requesting the weather information is input from a user; and providing the natural language response to the user in response to the request for the weather information.

Description

Method and device for providing weather information

[0001] The present invention relates to a method and device for providing weather information.

[0002] Recent advancements in smartphones and artificial intelligence have made it easier for users to access weather information anytime, anywhere. Currently, most weather information services provide predefined weather data formats based on the user's location. These services typically provide basic information such as temperature, wind direction, wind speed, precipitation amount, probability of precipitation, and humidity. Some also provide additional information, such as fine dust concentration and UV index.

[0003] However, current weather information systems have several limitations. First, most systems fail to address users' specific needs. For example, even when users seek detailed weather conditions for specific activities, they often only provide general information. Furthermore, current systems generally lack natural language processing capabilities, making it difficult to accurately understand and provide appropriate answers to user queries in everyday language.

[0004] Therefore, there is a need to develop a new weather information provision device that can accurately understand users' natural language requests and provide personalized weather information.

[0005] The problem that the present invention seeks to solve is to provide a method for providing weather information.

[0006] However, the problems to be solved by the present invention are not limited to those mentioned above, and other problems to be solved that are not mentioned can be clearly understood by a person having ordinary skill in the art to which the present invention pertains from the description below.

[0007] Hereinafter, specific means for achieving the purpose of the present disclosure will be described.

[0008] A method for providing weather information performed by a weather information providing device according to one embodiment may include: a step of collecting weather information; a step of inputting a natural language request requesting weather information into a pre-trained language model to generate a natural language response when a natural language request requesting weather information is input from a user; and a step of providing a natural language response to the user in response to the request for weather information.

[0009] In one embodiment, the step of generating a natural language response may include: inputting a natural language request into a language model to generate a task command; using the task command to generate a task result from previously collected weather information; and inputting the task result into the language model to generate a natural language response.

[0010] In one embodiment, the step of generating a work command may include: tokenizing a natural language request to generate a plurality of tokens; embedding the plurality of tokens; extracting the user's intent and key entities based on the embedding; and generating the work command using the intent and key entities.

[0011] In one embodiment, the step of collecting weather information may include a step of removing noise from the collected weather information using an interpolation-based noise removal algorithm.

[0012] In one embodiment, the method may further include a step of inputting the collected weather information into a predetermined recurrent neural network (RNN) model to detect whether there is an anomaly in the collected weather information; and a step of generating a notification about the anomaly if an anomaly is detected.

[0013] In one embodiment, the method may further include the step of providing a notification to the user or a predetermined weather control server.

[0014] In one embodiment, the method may further include a step of allocating resources used in executing the weather information providing method according to preset allocation criteria.

[0015] In one embodiment, the allocation criteria may include task priorities and resource requirements.

[0016] In one embodiment, the step of allocating resources may include the step of comparing resources required for executing the weather information providing method with available resources within the weather information providing device; and if the available resources are less than the resources required for executing the weather information providing method, the step of allocating resources used for executing the weather information providing method according to the task priority.

[0017] In one embodiment, the method further includes a step of encrypting the weather information using an encryption algorithm selected according to the importance of the weather information, and in the step of generating a natural language response, the natural language response can be generated using the encrypted weather information.

[0018] A weather information providing device according to another embodiment includes a memory in which a weather information providing program is stored; and a processor controlling the memory, wherein the processor executes the weather information providing program to collect weather information, and when a natural language request requesting weather information is input from a user, the processor inputs the natural language request into a pre-learned language model to generate a natural language response, and provides the natural language response to the user in response to the request for weather information.

[0019] In some embodiments, accessibility to weather information for non-expert users can be improved by responding in natural language to requests for weather information entered by users in natural language.

[0020] According to some embodiments, an efficient weather information providing device can be efficiently operated by allocating resources used in a weather information providing program according to preset allocation criteria.

[0021] In some embodiments, the security of the storage and processing of weather information can be enhanced by encrypting weather information to generate natural language responses and ensuring the integrity of the weather information.

[0022] However, the effects that can be obtained from the present disclosure are not limited to the effects mentioned above, and other effects that are not mentioned can be clearly understood by a person having ordinary skill in the technical field to which the present disclosure belongs from the description below.

[0023] The following drawings attached to this specification illustrate preferred embodiments of the present disclosure and, together with the detailed description of the invention, serve to further understand the technical idea of ​​the present disclosure, and therefore, the present disclosure should not be interpreted as being limited to matters described in such drawings.

[0024] Figure 1 is a block diagram illustrating a weather information providing device according to one embodiment.

[0025] Figure 2 is a block diagram conceptually illustrating the functions of a weather information provision program according to one embodiment.

[0026] FIG. 3 is a diagram illustrating a method for generating a natural language response using a language model according to one embodiment.

[0027] FIG. 4 is a drawing for explaining a method for collecting weather information by a weather information collection unit according to one embodiment.

[0028] FIG. 5 is a diagram for explaining a method for a resource allocation unit to allocate resources of a weather information provision device according to one embodiment.

[0029] Figure 6 is a flowchart for explaining the function of an encryption unit according to one embodiment.

[0030] FIG. 7 is a flowchart illustrating a weather information providing method performed by a weather information providing device according to one embodiment.

[0031] The various embodiments described in this specification are exemplified for the purpose of clearly explaining the technical concept of the present disclosure and are not intended to be limited to specific embodiments. The technical concept of the present disclosure includes various modifications, equivalents, alternatives, and embodiments selectively combined from all or part of the embodiments described herein. Furthermore, the scope of the technical concept of the present disclosure is not limited to the various embodiments presented below or the specific descriptions thereof.

[0032] Terms used herein, including technical or scientific terms, unless otherwise defined, may have the meaning commonly understood by one of ordinary skill in the art to which this disclosure belongs.

[0033] As used herein, expressions such as "includes," "may include," "comprises," "may have," "have," and "may have" indicate the presence of a target feature (e.g., a function, operation, or component), but do not exclude the presence of other additional features. In other words, such expressions should be understood as open-ended terms that imply the possibility of including a second embodiment.

[0034] In this specification, singular expressions include plural expressions unless the context clearly indicates otherwise. Furthermore, plural expressions include singular expressions unless the context clearly indicates otherwise. When a part of the specification is said to include a component, this does not exclude other components, but rather implies that other components may be included, unless otherwise specifically stated.

[0035] Also, the term 'module' or 'part' used in the specification means a software or hardware component, and the 'module' or 'part' performs certain roles. However, the 'module' or 'part' is not limited to software or hardware. The 'module' or 'part' may be configured to reside on an addressable storage medium and may be configured to execute one or more processors. Thus, as an example, the 'module' or 'part' may include at least one of components such as software components, object-oriented software components, class components, and task components, processes, functions, attributes, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, or variables. The functionality provided within the components and 'modules' or 'parts' may be combined into a smaller number of components and 'modules' or 'parts', or further separated into additional components and 'modules' or 'parts'.

[0036] According to one embodiment of the present disclosure, a 'module' or 'unit' may be implemented as a processor and a memory. 'Processor' should be broadly construed to include a general-purpose processor, a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a controller, a microcontroller, a state machine, and the like. In some circumstances, a 'processor' may also refer to an application-specific integrated circuit (ASIC), a programmable logic device (PLD), a field-programmable gate array (FPGA), and the like. A 'processor' may also refer to a combination of processing devices, such as, for example, a combination of a DSP and a microprocessor, a combination of multiple microprocessors, a combination of one or more microprocessors in conjunction with a DSP core, or any other such combination of configurations. In addition, 'memory' should be broadly construed to include any electronic component capable of storing electronic information. 'Memory' may refer to various types of processor-readable media, such as random access memory (RAM), read-only memory (ROM), non-volatile random access memory (NVRAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable PROM (EEPROM), flash memory, magnetic or optical data storage, registers, etc. Memory is said to be in electronic communication with the processor if the processor can read information from, and / or write information to, the memory. Memory integrated in a processor is in electronic communication with the processor.

[0037] As used herein, the expressions “first,” “second,” or “first,” “second,” etc., unless the context indicates otherwise, are used to refer to multiple similar objects and to distinguish one object from another, and do not limit the order or importance among the objects.

[0038] As used herein, the expressions "A, B, and C," "A, B, or C," "A, B, and / or C," or "at least one of A, B, and C," "at least one of A, B, or C," "at least one of A, B, and / or C," "at least one selected from A, B, and C," "at least one selected from A, B, or C," "at least one selected from A, B, and / or C," and the like can mean each listed item or all possible combinations of the listed items. For example, "at least one selected from A and B" can refer to (1) A, (2) at least one of A, (3) B, (4) at least one of B, (5) at least one of A and at least one of B, (6) at least one of A and B, (7) at least one of B and A, and (8) both A and B.

[0039] The expression "based on" as used herein is used to describe one or more factors that influence a decision, act of judgment, or action described in a phrase or sentence containing the expression, and this expression does not exclude additional factors that influence the decision, act of judgment, or action.

[0040] As used herein, the expression that a component (e.g., a first component) is “connected” or “connected” to another component (e.g., a second component) may mean that the component is directly connected or connected to the other component, as well as connected or connected via a new other component (e.g., a third component).

[0041] The expression "configured to" used herein may have the meanings of "set to", "having the ability to", "modified to", "made to", "capable of", etc., depending on the context. The expression is not limited to the meaning of "specifically designed in hardware", and for example, a processor configured to perform a specific operation may mean a generic-purpose processor that can perform the specific operation by executing software.

[0042] Hereinafter, various embodiments of the present disclosure will be described with reference to the attached drawings. In the attached drawings and the description of the drawings, identical or substantially equivalent components may be assigned the same reference numerals. Furthermore, in the description of various embodiments below, duplicate descriptions of identical or corresponding components may be omitted, but this does not mean that the corresponding components are not included in the embodiments.

[0043] Figure 1 is a block diagram illustrating a weather information providing device according to one embodiment.

[0044] Referring to FIGS. 1 and 2, a weather information providing device (100) is provided to provide a natural language response related to weather information when a natural language input is received from a user, and may include a processor (110), a transmitter / receiver (120), and a memory (130).

[0045] The processor (110) can control the overall operation of the weather information providing device (100).

[0046] The processor (110) can collect weather information in real time using a transceiver (120).

[0047] The memory (130) may store a weather information providing program (200) and information necessary for executing the weather information providing program (200).

[0048] In this specification, a weather information providing program (200) may mean software including commands programmed to collect weather information in real time, receive a natural language request from a user, input the input natural language request into a language model to generate a natural language response, provide the generated natural language response to the user, allocate resources used for executing the weather information providing program (200) according to preset allocation criteria, and encrypt weather information using an encryption algorithm selected according to the importance of the weather information.

[0049] The processor (110) can load the weather information providing program (200) and information necessary for executing the weather information providing program (200) from the memory (130) to execute the weather information providing program (200).

[0050] The processor (110) executes a weather information providing program (200) to collect weather information in real time, and when a natural language request is input from a user, inputs the input natural language request into a language model to generate a natural language response, provides the generated natural language response to the user, allocates resources used for executing the weather information providing program (200) according to preset allocation criteria, and encrypts the weather information using an encryption algorithm selected according to the importance of the weather information.

[0051] The functions and / or operations of the weather information provision program (200) will be examined in detail with reference to Fig. 2.

[0052] FIG. 2 is a block diagram conceptually illustrating the function of a weather information providing program according to one embodiment, FIG. 3 is a diagram for explaining a method for generating a natural language response using a language model according to one embodiment, FIG. 4 is a diagram for explaining a method for a weather information collecting unit according to one embodiment to collect weather information, FIG. 5 is a diagram for explaining a method for a resource allocation unit according to one embodiment to allocate resources of a weather information providing device, and FIG. 6 is a flowchart for explaining the function of an encryption unit according to one embodiment.

[0053] Referring to FIGS. 1 and 2, the weather information provision program (200) may include a natural language processing unit (210), a weather information collection unit (220), a weather information analysis unit (230), a resource allocation unit (240), and a data encryption unit (250).

[0054] The natural language processing unit (210), weather information collection unit (220), weather information analysis unit (230), resource allocation unit (240), and data encryption unit (250) illustrated in FIG. 2 conceptually divide the functions of the weather information providing program (200) to easily explain the functions of the weather information providing program (200), but are not limited thereto. According to embodiments, the functions of the natural language processing unit (210), weather information collection unit (220), weather information analysis unit (230), resource allocation unit (240), and data encryption unit (250) can be merged / separated, and implemented as a series of commands included in one program.

[0055] When a natural language request requesting weather information is input from a user, the natural language processing unit (210) can interpret the input natural language request and provide a natural language response.

[0056] More specifically, the natural language processing unit (210) receives a natural language request requesting weather information from a user in natural language, inputs the input natural language request into a language model to generate a work command, transmits the generated work command to a weather information analysis unit (230) to be described later to receive a work result for the work command, inputs the work result into a language model to generate a natural language response, and provides the generated natural language response to the user.

[0057] Here, the language model may be a neural network model trained to understand a user's request from input natural language. In some embodiments, the language model may be a neural network model based on Bidirectional Encoder Representations from Transformers (BERT).

[0058] Referring further to Figure 3, when a language model receives a natural language request from a user (S300), it can preprocess and normalize the natural language request (S310). In some embodiments, when preprocessing the natural language request, the language model can remove noise from the natural language request and separate it into sentence units.

[0059] The language model can tokenize a normalized natural language request and embed the tokenized natural language request (S320).

[0060] More specifically, a language model can break down a normalized natural language request into tokens (e.g., word units). At this point, the language model can segment the words into smaller subword units to convert them into token sequences. The language model can then add special tokens to the tokenized natural language request. For example, the language model can add the [CLS] token to the beginning of the tokenized natural language request and the [SEP] token to the end.

[0061] The language model can then perform embedding, which converts each token into a fixed-size vector. In some embodiments, the embedding performed by the language model may include token embedding, which converts each token into a vector representing its meaning, segment embedding, which converts each token into a vector representing its position in a sentence, and position embedding, which converts each token into a vector providing information about its order. Accordingly, the language model can generate a final embedding that includes token embedding, segment embedding, and position embedding.

[0062] The language model can analyze the user's intention using embedding (S330) and extract key entities (S335).

[0063] For example, if the natural language request is "Acquire temperature data every 5 minutes and upload it to the server," the language model can use the embedding of the natural language request to analyze that the user's intent is "collect and transmit data," and extract that the key entities for this are "5 minutes (time interval)" and "temperature (data type)."

[0064] Although FIG. 3 illustrates that the language model performs the process of analyzing the user's intent (S330) and the process of extracting key entities (S335) simultaneously or in parallel, this is not limited thereto. That is, depending on the embodiment, the language model may sequentially perform the process of analyzing the user's intent (S330) and the process of extracting key entities (S335). For example, the language model may analyze the user's intent (S330) and then extract key entities (S335), or conversely, the language model may extract key entities (S335) and then analyze the user's intent (S330).

[0065] Thereafter, the language model can generate a task command using the analyzed user intent and key entities (S340). For example, the language model can generate a task command that includes the user intent and key entities. Here, the task command may refer to a command to be input into the weather information analysis unit (230), described below, to provide weather information to the user.

[0066] Finally, when the language model receives analysis results from the weather information analysis unit (230) in response to a work command, it can generate a natural language response using the analysis results (S350).

[0067] Meanwhile, depending on the embodiment, the language model may be a neural network model pre-trained to process multiple languages. Therefore, even if a user inputs a natural language request in various languages, such as Korean, English, Japanese, or Chinese, the language model can provide a natural language response to the user in the language they input.

[0068] The weather information collection unit (220) can collect weather information in real time and store the collected weather information.

[0069] More specifically, the weather information collection unit (220) can collect weather information in real time, preprocess the collected weather information, and store the preprocessed weather information.

[0070] According to an embodiment, the weather information collected by the weather information collection unit (220) may include at least one of temperature, humidity, fine dust, weather (clear, cloudy, rain, snow, etc.), and detailed weather (typhoon, storm, heavy rain, heavy snow, tsunami, wave, cold wave, dry, etc.).

[0071] According to an embodiment, the weather information collection unit (220) may collect weather information regardless of a natural language request, but may also collect weather information if a natural language request is input. That is, the natural language request may be about past weather information, or may be about future weather information. Accordingly, the weather information collection unit (220) collects and stores weather information at a predetermined time interval regardless of a natural language request, and if a natural language request is input, weather information may be collected and stored according to the time interval, type, and method corresponding to the request.

[0072] The weather information collection unit (220) can preprocess weather information using a predetermined preprocessing method. For example, the weather information collection unit (220) can preprocess weather information by inputting weather information into an interpolation-based noise removal algorithm to remove noise from the weather information.

[0073] Additionally, depending on the embodiment, the weather information collection unit (220) may generate a notification regarding an anomaly detected in the collected weather information. Depending on the embodiment, the weather information collection unit (220) may display the notification to the user, but may also transmit the notification to a predetermined weather control server (not shown).

[0074] For example, referring further to FIG. 4, the weather information collection unit (220) can collect weather information at a predetermined time interval (S400) and preprocess the collected weather information (S410).

[0075] The weather information collection unit (220) can detect whether there is an anomaly in the preprocessed weather information (S420). For example, the weather information collection unit (220) can detect anomalies in the weather information by inputting the weather information into a recurrent neural network (RNN) model, such as a Long Short-Term Memory (LSTM). In this specification, the occurrence of an anomaly in the weather information does not correspond to noise, but may refer to a case where the weather changes rapidly and requires notification to the user and / or the weather control server.

[0076] Accordingly, when detecting whether there is an anomaly in the weather information, the weather information collection unit (220) can not only detect whether there is an anomaly in the weather information already collected, but can also predict whether an anomaly will occur using the weather information already collected. To this end, according to an embodiment, the weather information collection unit (220) can input weather information into a time series prediction model, such as the prophet library, to predict weather information after a predetermined time interval, and determine whether there is an anomaly in the predicted weather information.

[0077] If an abnormality is detected in the weather information ('Yes' in S420), the weather information collection unit (220) can generate a notification including information about the abnormality (S425).

[0078] The weather information collection unit (220) can store weather information (S430) if no abnormality is detected in the weather information ('No' in S420).

[0079] The weather information collection unit (220) can transmit the stored weather information and / or notifications to a predetermined weather control server (S440).

[0080] When the weather information analysis unit (230) receives a work command from the natural language processing unit (210), it can analyze the weather information collected (stored) according to the work command and generate a work result according to the analysis result.

[0081] According to an embodiment, the weather information analysis unit (230) may extract weather information corresponding to the user's request from among the collected weather information using the user's intention and key entities included in the work command, and may generate a work result using the extracted weather information.

[0082] For example, if the user's intention is "data collection and transmission" and the main entities are "5 minutes (time interval)" and "temperature (data type)", the weather information analysis unit (230) can collect temperature at 5-minute intervals and generate work results.

[0083] The weather information analysis unit (230) can transmit the generated work results to the natural language processing unit (210).

[0084] The resource allocation unit (240) can allocate resources used for executing the weather information provision program (200) according to preset allocation criteria in order to efficiently use the resources being used.

[0085] Here, the resource may include at least one of a processor, memory (e.g., volatile memory such as RAM), storage space (e.g., HDD, SDD, database, external storage space, etc.), and network bandwidth.

[0086] Depending on the embodiment, the preset allocation criteria may include task priorities and resource requirements.

[0087] For example, the resource allocation unit (240) compares the resources required for executing the weather information providing program (200) with the available resources within the weather information providing device (100), and if the available resources are less than the resources required for executing the weather information providing program (200), the resources can be allocated to the weather information providing program (200) according to a preset task priority.

[0088] For example, referring further to FIG. 5, the resource allocation unit (240) can analyze available resources within the weather information provision device (100) by monitoring allocated resources (S500).

[0089] If the resources required to execute the weather information providing program (200) are greater than the available resources in the weather information providing device (100), the resource allocation unit (240) can allocate resources to the weather information providing program (200) according to the preset task priority (S510).

[0090] After allocating resources, the resource allocation unit (240) can evaluate the performance of the weather information provision device (100) and determine whether optimization is necessary based on the results of the evaluation (S520).

[0091] If the evaluation result indicates that optimization of resources is necessary ('Yes' in S520), the resource allocation unit (240) can reallocate resources to the weather information provision program (200) according to the preset work priority (S530).

[0092] After resource allocation (S510) or reallocation (S530), the resource allocation unit (240) can monitor the allocated or reallocated resources and re-analyze the available resources within the weather information provision device (100) (S500).

[0093] According to an embodiment, the resource allocation unit (240) may allocate resources using Docker containers and Kubernetes. For example, the resource allocation unit (240) may set resource-specific limits using Docker containers and perform load balancing and auto scaling using Kubernetes.

[0094] The encryption unit (250) can ensure the confidentiality and integrity of weather information when transmitting and storing weather information.

[0095] More specifically, the encryption unit (250) can encrypt weather information using an encryption algorithm selected according to the importance of the weather information, and can cause the weather information analysis unit (230) to generate a work result using the encrypted weather information. For example, the encryption unit (250) can generate a work result using the encrypted weather information using the homomorphic encryption (HE) technique.

[0096] In addition, the encryption unit (250) can verify the integrity of weather information using blockchain technology, enable weather information to be safely transmitted using a predetermined communication protocol, and control access to weather information using a predetermined access control method.

[0097] For example, referring further to FIG. 6, the encryption unit (250) can classify weather information into multiple groups based on importance (S600) and encrypt the weather information using an encryption algorithm corresponding to the classified group (S610). Here, the encryption algorithm may include AES-256-GCM and ChaCha20-Poly1305.

[0098] The encryption unit (250) can generate and manage an encryption key using a hardware security module (HSM) (S620).

[0099] The encryption unit (250) can verify the integrity of weather information using blockchain technology (S630) and enable weather information to be transmitted safely through TLS 1.3 and mutual authentication (S640).

[0100] In addition, the encryption unit (250) can use Role Based Access Control (RBAC) to allow only authorized users to access weather information (S650).

[0101] Meanwhile, although FIG. 6 illustrates that the encryption process (S600 to S620), the integrity verification process (S630), the secure transmission process (S640), and the access control process (S650) of the weather information are performed in sequence, the present invention is not limited thereto. That is, depending on the embodiment, at least two of the encryption process (S600 to S620), the integrity verification process (S630), the secure transmission process (S640), and the access control process (S650) may be performed simultaneously or in parallel. Alternatively, depending on the embodiment, at least some of the encryption process (S600 to S620), the integrity verification process (S630), the secure transmission process (S640), and the access control process (S650) may be performed in a different order from FIG. 6. For example, the encryption process (S600 to S620) and the integrity verification process (S630) may be performed in parallel, or the integrity verification process (S630) may be performed after the access control process (S650).

[0102] FIG. 7 is a flowchart illustrating a weather information providing method performed by a weather information providing device according to one embodiment.

[0103] Referring to FIG. 1, FIG. 2, and FIG. 7, the weather information collection unit (220) can collect weather information and store the collected weather information (S700).

[0104] When the natural language processing unit (210) receives a natural language request requesting weather information from a user (S710), it can input the input natural language request into a language model to generate a work command (S720).

[0105] The weather information analysis unit (230) can extract weather information corresponding to a user's request from among the collected weather information using a work command, and generate a work result using the extracted weather information (S730).

[0106] The natural language processing unit (210) can use the work results to input them into a language model to generate a natural language response and provide the generated natural language response to the user (S740).

[0107] In one embodiment, accessibility to weather information for non-expert users can be improved by responding in natural language to requests for weather information entered by users in natural language.

[0108] According to one embodiment, by allocating resources used in a weather information provision program according to preset allocation criteria, an efficient weather information provision device can be efficiently operated.

[0109] According to one embodiment, the security of storage and processing of weather information can be strengthened by encrypting weather information to generate a natural language response and ensuring the integrity of the weather information.

[0110] It is obvious that each step or operation of the method according to the embodiments of the present disclosure can be performed by a computer including one or more processors according to the execution of a computer program stored in a computer-readable recording medium.

[0111] The computer-executable instructions stored in the aforementioned recording medium can be implemented through a computer program programmed to perform each corresponding step, and such a computer program can be stored in a computer-readable recording medium and executed by a processor. The computer-readable recording medium may be a non-transitory readable medium. In this case, the non-transitory readable medium means a medium that semi-permanently stores data and can be read by a device, rather than a medium that stores data for a short period of time, such as a register, a cache, or a memory. Specifically, the programs for performing the various methods described above can be stored and provided in a non-transitory readable medium, such as semiconductor memory devices such as erasable programmable read-only memories (EPROMs), electrically erasable programmable read-only memories (EEPROMs), and flash memory devices, magnetic disks such as internal hard disks and removable disks, optical-magnetic disks, and non-volatile memories including CD-ROMs and DVD-ROM disks.

[0112] The methods according to various examples disclosed in this document may be provided as a computer program product. The computer program product may be distributed in the form of a machine-readable storage medium (e.g., compact disc read-only memory (CD-ROM)) or online through an application store (e.g., Play Store™). In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily generated in a storage medium, such as the memory of a manufacturer's server, an application store's server, or a relay server.

[0113] As described above, those skilled in the art will appreciate that the present disclosure can be implemented in other specific forms without altering the technical spirit or essential characteristics thereof. Therefore, the above-described embodiments should be understood as illustrative in all respects and not restrictive. The scope of the present disclosure is defined by the following claims rather than the detailed description, and all changes or modifications derived from the meaning and scope of the claims and equivalent concepts should be construed as being included within the scope of the present disclosure.

[0114] The features and advantages described in this specification are not exhaustive, and many additional features and advantages will become apparent to those skilled in the art upon review of the drawings, specification, and claims. Furthermore, it should be noted that the language used in this specification has been primarily selected for readability and instructional purposes, and may not be intended to delineate or circumscribe the subject matter of the present disclosure.

[0115] The above description of the embodiments of the present disclosure has been presented for illustrative purposes. It is not intended to be exhaustive or to limit the disclosure to the precise form disclosed. Those skilled in the art will appreciate that numerous modifications and variations are possible in light of the above disclosure.

[0116] Therefore, the scope of this disclosure is not limited by the detailed description, but is defined by any claims of the application based on this description. Accordingly, the disclosure of embodiments of this disclosure is illustrative and does not limit the scope of this disclosure, which is set forth in the following claims.

Claims

1. In a method for providing weather information performed by a weather information providing device, Steps to collect weather information; When a natural language request requesting the above weather information is input from a user, a step of inputting the natural language request into a pre-trained language model to generate a natural language response; and A step of providing the natural language response to the user in response to a request for the weather information. How to provide weather information.

2. In paragraph 1, The step of generating the above natural language response is: A step of inputting the natural language request into the language model to generate a work command; A step of generating a work result from the collected weather information using the above work command; and A step of inputting the above work result into the language model and generating the natural language response. How to provide weather information.

3. In paragraph 2, The steps for generating the above work command are: A step of tokenizing the above natural language request to generate multiple tokens; A step of embedding the above plurality of tokens; A step of extracting the user's intention and main entities according to the above embedding; and A step of generating the work command using the above intention and the above main entity. How to provide weather information.

4. In paragraph 1, The steps for collecting the above weather information are: A step of removing noise from the collected weather information using an interpolation-based noise removal algorithm is included. How to provide weather information.

5. In paragraph 1, A step of inputting the collected weather information into a predetermined recurrent neural network (RNN) model to detect whether there is an abnormality in the collected weather information; and If the above abnormality is detected, further comprising a step of generating a notification for the above abnormality. How to provide weather information.

6. In paragraph 5, Further comprising a step of providing the above notification to the user or a predetermined weather control server. How to provide weather information.

7. In paragraph 1, Further comprising a step of allocating resources used for executing the above weather information provision method according to preset allocation criteria. How to provide weather information.

8. In paragraph 7, The above allocation criteria include task priorities and resource requirements. How to provide weather information.

9. In paragraph 8, The step of allocating the above resources is: A step of comparing the resources required for executing the above weather information providing method with the available resources within the weather information providing device; and If the available resources are less than the resources required to execute the weather information providing method, a step of allocating resources used to execute the weather information providing method according to the task priority is included. How to provide weather information.

10. In paragraph 1, Further comprising a step of encrypting the weather information using an encryption algorithm selected according to the importance of the weather information, In the step of generating the above natural language response, Generating the natural language response using the encrypted weather information above. How to provide weather information.

11. Memory where the weather information provision program is stored; and including a processor that controls the above memory, The above processor executes the above weather information providing program, Collect weather information, When a natural language request requesting the above weather information is entered from a user, the natural language request is input into a pre-trained language model to generate a natural language response, Providing the natural language response to the user in response to the request for the above weather information. Weather information providing device.

Images