CN-122019236-A - Method and device for processing abnormal information, electronic device, storage medium and program

CN122019236ACN 122019236 ACN122019236 ACN 122019236ACN-122019236-A

Abstract

The present disclosure provides a method, an apparatus, an electronic device, a storage medium, and a program for processing abnormality information. The method comprises the steps of responding to multiple items of attribute information of a current abnormal condition, packaging the multiple items of attribute information into multiple items of attribute data in a data structure body, generating natural sentences by utilizing at least one item of target attribute data in the attribute data of the data structure body, wherein the natural sentences are used for describing all items of target attribute data, judging whether the natural sentence representation meets preset display standards or not, and displaying the natural sentences in response to the fact that the display standards are met. According to the embodiment, the attribute information of the abnormal condition is packaged into the data structure body, so that all attribute information is unified into the standardized expression form, all the attribute data standardized in the expression form in the data structure body can be used as effective context information in the process of reasoning natural sentences, identification errors of the attribute information in different forms are avoided, and a more accurate result is obtained.

Inventors

LI DONGYANG
DENG XI

Assignees

北京懂车族科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260202

Claims (10)

1. A processing method of abnormal information includes: In response to detecting the pieces of attribute information of the current abnormal situation, encapsulating the pieces of attribute information into pieces of attribute data in a data structure; generating a natural sentence by utilizing at least one item of target attribute data in each item of attribute data of the data structure, wherein the natural sentence is used for describing each item of target attribute data; judging whether the natural sentence representation accords with a preset display standard or not; The natural sentence is presented in response to determining that the presentation criterion is met.
2. The method of claim 1, wherein the natural language sentence comprises a hint sentence, and The generating a natural sentence using at least one item of target attribute data in each item of attribute data of the data structure includes: Determining a prompt word comprising a plurality of instructions, wherein the plurality of instructions comprise at least one content instruction pointing to target attribute data and at least one description instruction of constraint statement format and semantic content; Determining target attribute data corresponding to each content indication from the data structure body according to each content indication; and generating the prompt statement pointing to each item of target attribute data according to each item of description indication.
3. The method of claim 2, wherein the target attribute data includes error codes, scene data, interface data, operation data corresponding to the current abnormal situation, the description indication includes a format indication for constraining word count and/or sentence format, and a language indication for constraining semantic content, and The generating the prompt statement pointing to each item of target attribute data according to each item of description instruction comprises the following steps: mapping the error code into a corresponding preset semantic tag; And generating the prompt statement which is used for describing the semantic tag, the scene data, the interface data and the operation data and accords with the description instruction and the language instruction by using a preset large language model.
4. The method of claim 2, wherein the natural language sentence further comprises an operation language sentence, the plurality of indications further comprises at least one guide indication defining an operation action, and After the generating of the hint statement that points to target attribute data according to the description indications, the method further includes: Determining a target operation action corresponding to the current abnormal condition from a plurality of preset operation actions based on various target attribute data of the current abnormal condition; and generating the operation statement pointing to each target operation action according to each guiding instruction.
5. The method of claim 1, wherein the determining whether the natural language sentence representation meets a preset presentation criterion comprises: determining the semantic readability degree of the natural sentence and taking the semantic readability degree as a first semantic index; determining the emotion aggressiveness of the natural sentence and taking the emotion aggressiveness as a second semantic index; Determining the guiding degree of the natural sentence and taking the guiding degree as a third semantic index; Determining semantic consistency between the natural sentence and each item of target attribute data, and taking the semantic consistency as a fourth semantic index; Determining weighted results of the first semantic index, the second semantic index, the third semantic index and the fourth semantic index; And in response to determining that the weighted result is greater than or equal to a preset evaluation threshold, determining that the natural sentence representation meets the presentation criterion.
6. The method of claim 5, wherein after the determining the weighted results of the first, second, third, and fourth semantic metrics, the method further comprises: And responding to the fact that the weighted result is smaller than the evaluation threshold, selecting a candidate sentence corresponding to the current abnormal condition from the plurality of candidate sentences according to a preset mapping relation between the plurality of preset candidate sentences and various abnormal conditions, and determining the candidate sentence as the natural sentence.
7. An abnormality information processing apparatus comprising: a packaging module configured to package pieces of attribute information into pieces of attribute data in a data structure in response to detecting the pieces of attribute information of the current abnormal situation; A generation module configured to generate a natural sentence using at least one item of target attribute data in each item of attribute data of the data structure, the natural sentence being used to describe each item of target attribute data; the judging module is configured to judge whether the natural sentence representation accords with a preset display standard or not; and a presentation module configured to present the natural sentence in response to determining that the presentation criterion is met.
8. An electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon, The one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of any of claims 1-6.
9. A computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by one or more processors implements the method of any of claims 1-6.
10. A computer program product comprising computer program instructions which, when run on a computer, cause the computer to perform the method of any of claims 1-6.

Description

Method and device for processing abnormal information, electronic device, storage medium and program Technical Field The embodiment of the application relates to the technical field of abnormal error detection, in particular to a method and a device for processing abnormal information, electronic equipment, a storage medium and a program. Background In the process of running a related business system or executing a related task process through an electronic device or a processor, a user often generates errors or anomalies due to improper operation or system problems of the user, and displays Error prompts or anomalies to the user, wherein the Error prompts or anomalies can be SERVER INTERAL Error, unexpect Error unexpected Error, messy codes or the like. Since such error cues or exception cues are usually expressed in English, terms of art, code language or messy codes, such error cues or exception cues are very obscure to users of non-expert technicians, so that users often have difficulty in understanding the specific error or exception pointed by the error cues or exception cues, and also cannot know the cause of the error or exception, so that the processing operation of relieving the error or exception is difficult to be performed smoothly. Disclosure of Invention In view of this, embodiments of the present disclosure provide a method, an apparatus, an electronic device, a storage medium, and a program for processing exception information. In a first aspect, an embodiment of the present disclosure provides a method for processing exception information, including: In response to detecting the pieces of attribute information of the current abnormal situation, encapsulating the pieces of attribute information into pieces of attribute data in a data structure; generating a natural sentence by utilizing at least one item of target attribute data in each item of attribute data of the data structure, wherein the natural sentence is used for describing each item of target attribute data; judging whether the natural sentence representation accords with a preset display standard or not; In response to determining that the presentation criteria are met, the natural sentence is presented. In some alternative embodiments, generating a natural sentence using at least one item of target genus data in each item of attribute data of the data structure includes: Determining a prompt word comprising a plurality of instructions, wherein the plurality of instructions comprise at least one content instruction pointing to target attribute data and at least one description instruction of constraint statement format and semantic content; Determining target attribute data corresponding to each content indication from the data structure body according to each content indication; generating a prompt statement pointing to each item of target attribute data according to each item of description indication. In some optional embodiments, the target attribute data includes an error code, scene data, interface data, and operation data corresponding to the current abnormal situation, and the description instruction includes a format instruction for restricting the number of words and/or the sentence format, and a language instruction for restricting the semantic content; correspondingly, generating a prompt statement pointing to each item of target attribute data according to each item of description indication comprises: mapping the error code into a corresponding preset semantic label; And generating prompt sentences for describing the semantic tags, the scene data, the interface data and the operation data by using a preset large language model, wherein the prompt sentences accord with description indication and language indication. In some alternative embodiments, the natural language sentence further comprises an operation sentence, and the plurality of indications further comprises at least one guidance indication for defining an operation action; accordingly, after generating a hint statement that points to each item of target attribute data according to each item of description indication, the method further includes: Determining a target operation action corresponding to the current abnormal condition from a plurality of preset operation actions based on various target attribute data of the current abnormal condition; And generating operation sentences pointing to the target operation actions according to the guiding instructions. In some alternative embodiments, determining whether the natural sentence representation meets a preset presentation criterion includes: determining the semantic readability degree of a natural sentence and taking the semantic readability degree as a first semantic index; determining the emotion aggressiveness of the natural sentence and taking the emotion aggressiveness as a second semantic index; determining the guiding degree of the natural sentence and taking the guiding degree as a third semantic