CN-115563955-B - Text processing method, device, equipment and medium

CN115563955BCN 115563955 BCN115563955 BCN 115563955BCN-115563955-B

Abstract

The invention discloses a text processing method, a text processing device, text processing equipment and a text processing medium. The method comprises the steps of determining multiple groups of target record texts from at least two candidate record texts according to vehicle types recorded by the record texts, extracting features of the target record texts according to semantic extraction rules, determining candidate fault description information and candidate fault grades recorded by each target record text in each group of target record texts, determining target fault grades and target fault description information of each group of target record texts according to the number of the target record texts of each candidate fault grade in each group of target record texts and the candidate fault description information, and visually displaying text processing results according to the target fault grades and the target fault description information of each group of target record texts. According to the technical scheme, fault information of different vehicle types in the recorded text can be accurately extracted and visualized, and subsequent analysis of potential risks of the vehicle is facilitated.

Inventors

WANG ZHAOLIN
DING GUANYUAN
HUI SHU
GUO FUQI
HUANG JIATONG
ZHENG TONG
ZHANG WENJUAN

Assignees

中国第一汽车股份有限公司

Dates

Publication Date: 20260508
Application Date: 20220921

Claims (7)

1. A text processing method, comprising: Determining a plurality of groups of target record texts from at least two candidate record texts according to vehicle types recorded by the record texts, wherein the record texts are texts for recording fault conditions of vehicles of different vehicle types in a historical operation process; extracting features of the target record texts according to semantic extraction rules, and determining candidate fault description information and candidate fault grades of each target record text record in each group of target record texts, wherein the semantic extraction rules are preset matching rules for each candidate fault grade; Determining target fault levels and target fault description information of each group of target record texts according to the number of target record texts of each candidate fault level in each group of target record texts and the candidate fault description information; according to the target fault grade and the target fault description information of each group of target record texts, visually displaying the text processing results; The feature extraction is performed on the target record text according to a semantic extraction rule, and candidate fault description information and candidate fault grades of each target record text record in each group of target record texts are determined, including: determining whether each target record text in each group of target record texts meets preset semantic extraction rules or not according to each target record text; If yes, extracting features of each target record text according to semantic extraction rules met by each target record text, and determining candidate fault description information of each target record text; Determining the candidate fault level of each target record text as the fault level to which the semantic extraction rule which is satisfied by the candidate fault level; the feature extraction of each target record text according to the semantic extraction rule satisfied by each target record text comprises the following steps: extracting a first keyword group in the target record text if the fault level of the semantic extraction rule met by the target record text is a first fault level, wherein the first keyword group comprises at least one of barrier nouns, collision verbs, vehicle non-key part nouns, destruction verbs and negatives; And if the fault level of the semantic extraction rule met by the target record text is a second fault level, extracting a second keyword group in the target record text, wherein the second keyword group comprises at least one of a non-risk noun, a low-risk adjective, a collision verb, a car noun, a vehicle key part noun and a harmless adjective.
2. The method of claim 1, wherein determining the target fault level and the target fault description information for each set of target record texts based on the number of target record texts for each candidate fault level in each set of target record texts and the candidate fault description information, comprises: determining the proportional relation between the target record text of each candidate fault level and the total target record text according to the number of the target record texts of each candidate fault level in each group of target record texts; determining target fault levels of each group of target record texts according to the proportional relation; And integrating candidate fault description information of each target record text in each group of target record texts to serve as target fault description information of the group of target record texts.
3. The method of claim 1, wherein visually presenting the results of the text processing based on the target fault level and the target fault description information of each set of target record text, comprises: and visually displaying the text processing result according to the target vehicle type associated with each group of target record texts, the target fault level and the target fault description information of each group of target record texts.
4. A method according to claim 3, further comprising: acquiring a vehicle type to be displayed, and determining a fault level to be displayed and fault description information to be displayed, which correspond to the vehicle type to be displayed; And visually displaying the text processing result according to the vehicle type to be displayed, the fault level to be displayed and the fault description information to be displayed.
5. A text processing apparatus, comprising: The target text determining module is used for determining a plurality of groups of target recorded texts from at least two candidate recorded texts according to vehicle types recorded by the recorded texts, wherein the recorded texts are texts for recording fault conditions of vehicles of different vehicle types in a historical operation process; The candidate information determining module is used for extracting the characteristics of the target record texts according to semantic extraction rules, and determining candidate fault description information and candidate fault grades of each target record text record in each group of target record texts, wherein the semantic extraction rules are preset matching rules for each candidate fault grade; The target information determining module is used for determining target fault levels and target fault description information of each group of target record texts according to the number of target record texts of each candidate fault level in each group of target record texts and the candidate fault description information; The visualization module is used for carrying out visual display on the text processing result according to the target fault grade and the target fault description information of each group of target record texts; Wherein the candidate information determination module comprises: The judging unit is used for determining whether each target record text in each group of target record texts meets preset semantic extraction rules or not; The candidate information determining unit is used for extracting the characteristics of the target record texts according to semantic extraction rules met by the target record texts if the target record texts are in the same category, and determining candidate fault description information and candidate fault grades of the target record texts in the target record texts; the candidate information determining unit is specifically configured to: according to semantic extraction rules satisfied by each target record text, extracting features of each target record text, and determining candidate fault description information of each target record text; Determining the candidate fault level of each target record text as the fault level to which the semantic extraction rule which is satisfied by the candidate fault level; the feature extraction of each target record text according to the semantic extraction rule satisfied by each target record text comprises the following steps: extracting a first keyword group in the target record text if the fault level of the semantic extraction rule met by the target record text is a first fault level, wherein the first keyword group comprises at least one of barrier nouns, collision verbs, vehicle non-key part nouns, destruction verbs and negatives; And if the fault level of the semantic extraction rule met by the target record text is a second fault level, extracting a second keyword group in the target record text, wherein the second keyword group comprises at least one of a non-risk noun, a low-risk adjective, a collision verb, a car noun, a vehicle key part noun and a harmless adjective.
6. An electronic device, the electronic device comprising: At least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the text processing method of any of claims 1-4.
7. A computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, the computer instructions for causing a processor to perform the text processing method of any of claims 1-4 when executed.

Description

Text processing method, device, equipment and medium Technical Field The present invention relates to the field of vehicles, and in particular, to a text processing method, apparatus, device, and medium. Background With the development of vehicle technology, the importance of the safety of the functions of the automobiles seems to be higher and higher, various types of faults can occur in the running process of the vehicles of different automobile types, and the fault record text contains rich valuable information. The text content is deeply analyzed, so that guidance can be provided for product investigation, planning, research and development, frequent fault analysis and early warning. Therefore, how to better process the text of the recorded text after the vehicle is in fault, accurately extract valuable information in the text and visualize the valuable information, so that related personnel can know the potential risk of the product, and the aim of improving the safety of the vehicle is the problem to be solved urgently at present. Disclosure of Invention The invention provides a text processing method, a device, equipment and a medium, which can accurately extract fault information of different vehicle types in a recorded text and visualize the fault information, thereby being beneficial to subsequent analysis of potential risks of vehicles. According to an aspect of the present invention, there is provided a text processing method including: Determining a plurality of groups of target record texts from at least two candidate record texts according to vehicle types recorded by the record texts, wherein the record texts are texts for recording fault conditions of vehicles of different vehicle types in a historical operation process; According to the semantic extraction rule, extracting the characteristics of the target record texts, and determining candidate fault description information and candidate fault grades of each target record text record in each group of target record texts; Determining target fault grades and target fault description information of each group of target record texts according to candidate fault grades recorded by each target record text in each group of target record texts; And visually displaying the text processing result according to the target fault grade and the target fault description information of each group of target record texts. According to another aspect of the present invention, there is provided a text processing apparatus including: The target text determining module is used for determining a plurality of groups of target recorded texts from at least two candidate recorded texts according to vehicle types recorded by the recorded texts, wherein the recorded texts are texts for recording fault conditions of vehicles of different vehicle types in a historical operation process; the candidate information determining module is used for extracting the characteristics of the target record texts according to the semantic extraction rule and determining candidate fault description information and candidate fault grades of each target record text record in each group of target record texts; The target information determining module is used for determining target fault levels and target fault description information of each group of target record texts according to candidate fault levels recorded by each target record text in each group of target record texts; and the visualization module is used for carrying out visual display on the text processing result according to the target fault grade and the target fault description information of each group of target record texts. According to another aspect of the present invention, there is provided an electronic apparatus including: At least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the text processing method according to any one of the embodiments of the present invention. According to another aspect of the present invention, there is provided a computer readable storage medium storing computer instructions for causing a processor to execute a text processing method according to any one of the embodiments of the present invention. According to the technical scheme of the embodiment of the invention, multiple groups of target record texts are determined from at least two candidate record texts according to the vehicle type recorded by the record texts, feature extraction is carried out on the target record texts according to semantic extraction rules, candidate fault description information and candidate fault grades recorded by each target record text in each group of target record texts are determined, the target fault grade and the target fault description information of each group of target record texts are determined according to the number of t