CN-122021580-A - Chart generation method, apparatus, electronic device and computer readable storage medium

CN122021580ACN 122021580 ACN122021580 ACN 122021580ACN-122021580-A

Abstract

The application discloses a chart generation method, a chart generation device, electronic equipment and a computer readable storage medium, and relates to the technical field of semiconductors, wherein the method comprises the following steps: and obtaining chart description information, wherein the chart description information is used for indicating to generate a chart of the semiconductor field, semantic representation information corresponding to the chart description information is generated based on a field knowledge graph, the field knowledge graph corresponds to the semiconductor field, target chart data is queried according to the semantic representation information, and a target chart corresponding to the target chart data is constructed. Semantic understanding and control during chart generation are conducted by introducing a domain knowledge graph in the semiconductor domain, so that chart data and business accuracy of the chart during chart generation in the semiconductor domain are improved, and accuracy of chart generation in the semiconductor domain is improved. And the chart is automatically generated through chart description, so that the learning cost of a user on complex information such as chart structures and the like is reduced, and the user experience and chart generation efficiency are improved.

Inventors

RU XINGXING
XU MINGGUANG
WEI TAO
YI WENTAO

Assignees

格创东智(武汉)科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260128

Claims (10)

1. A chart generation method, the method comprising: acquiring chart description information, wherein the chart description information is used for indicating the generation of a chart in the semiconductor field; Generating semantic representation information corresponding to the chart description information based on a domain knowledge graph, wherein the domain knowledge graph corresponds to the semiconductor domain; Querying target chart data according to the semantic representation information; and constructing a target chart corresponding to the target chart data.
2. The graph generation method according to claim 1, wherein the constructing the target graph corresponding to the target graph data includes: Identifying data type information of the target chart data; performing feature extraction on the chart data according to the data type information to obtain data feature information; Screening a target chart type according to the data characteristic information; and constructing a target chart corresponding to the target chart type based on the target chart data and the data characteristic information.
3. The chart generation method according to claim 2, wherein the screening the target chart type according to the data characteristic information includes: determining first matching score information of the data characteristic information aiming at least one preset chart type based on a domain chart rule, wherein the domain chart rule corresponds to the semiconductor domain; outputting second matching score information of the data characteristic information aiming at least one preset chart type through a target model; for each preset chart type, fusing first matching score information and second matching score information corresponding to the preset chart type to obtain fused score information; and screening the target chart type from the preset chart types according to the fusion score information corresponding to each preset chart type.
4. The chart generation method according to claim 3, wherein the domain chart rule includes a configuration condition, a preset chart type, and a corresponding basic confidence score, wherein the basic confidence score is obtained based on an evaluation of a degree of matching of the configuration condition with the preset chart type by industry experience of the semiconductor domain; The determining, based on the domain graph rule, first matching score information of the data feature information for at least one preset graph type includes: Carrying out association matching on the configuration conditions according to the data characteristic information to obtain a condition matching result; and screening at least one preset chart type according to the condition matching result, and determining a basic confidence score corresponding to the preset chart type as the first matching score information.
5. The chart generating method according to claim 2, wherein the constructing the target chart corresponding to the target chart type based on the target chart data and the data characteristic information includes: Matching configuration description information from a ECharts knowledge base and a domain chart knowledge base according to the target chart data, the data characteristic information and the target chart type; Outputting configuration description information, the target chart data, the data characteristic information and configuration structure information corresponding to the target chart type information through a target model; And generating the target chart according to the configuration structure information.
6. The chart generation method according to claim 1, wherein the target chart is generated based on configuration structure information corresponding to the target chart data; After the target chart corresponding to the target chart data is constructed, the method further comprises the following steps: if the supplementary description information is detected, determining intention change information according to the supplementary description information and the chart description information; positioning a target configuration node from the configuration structure information according to the intention change information, and determining updated configuration information of the target configuration node according to the intention change information; Updating the configuration structure information according to the target configuration node and the updating configuration information to obtain updating structure information; And updating the target chart according to the updated structure information to obtain a target optimization chart.
7. The graph generation method according to claim 1, wherein the querying the target graph data according to the semantic representation information includes: Generating a target query statement conforming to the structure of the target database through a target model based on the field mapping dictionary and the semantic representation information; And inquiring the target chart data from the target database according to the target inquiry statement.
8. A chart generating apparatus, characterized in that the apparatus comprises: The acquisition module is used for acquiring chart description information, wherein the chart description information is used for indicating to generate a chart in the semiconductor field; The generation module is used for generating semantic representation information corresponding to the chart description information based on a domain knowledge graph, and the domain knowledge graph corresponds to the semiconductor domain; The query module is used for querying target chart data according to the semantic representation information; and the construction module is used for constructing a target chart corresponding to the target chart data.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps in the chart generation method of any of claims 1-7 when the computer program is executed.
10. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements the steps in the chart generation method according to any of claims 1-7.

Description

Chart generation method, apparatus, electronic device and computer readable storage medium Technical Field The present application relates to the field of semiconductor technologies, and in particular, to a chart generating method, a chart generating device, an electronic device, and a computer readable storage medium. Background In the semiconductor manufacturing industry, equipment operating state, production line efficiency, yield analysis, defect statistics, parameter fluctuation monitoring and the like all depend on a large amount of data visualization charts. Chart systems are an important tool for engineers to conduct yield analysis, SPC (statistical process control) monitoring and abnormal trend discovery. However, existing chart generation systems lack an index hierarchy in the semiconductor field, resulting in queried chart data and charts often lacking business accuracy. Disclosure of Invention The embodiment of the application provides a chart generation method, a chart generation device, electronic equipment and a computer readable storage medium, which can improve the accuracy of chart generation in the semiconductor field, reduce the learning cost of a user and improve the user experience. In a first aspect, an embodiment of the present application provides a chart generating method, including: acquiring chart description information, wherein the chart description information is used for indicating the generation of a chart in the semiconductor field; Generating semantic representation information corresponding to the chart description information based on a domain knowledge graph, wherein the domain knowledge graph corresponds to the semiconductor domain; Querying target chart data according to the semantic representation information; and constructing a target chart corresponding to the target chart data. In a second aspect, an embodiment of the present application further provides a chart generating apparatus, including: The acquisition module is used for acquiring chart description information, wherein the chart description information is used for indicating to generate a chart in the semiconductor field; The generation module is used for generating semantic representation information corresponding to the chart description information based on a domain knowledge graph, and the domain knowledge graph corresponds to the semiconductor domain; The query module is used for querying target chart data according to the semantic representation information; and the construction module is used for constructing a target chart corresponding to the target chart data. Optionally, in some embodiments of the present application, the constructing a target chart corresponding to the target chart data includes: Identifying data type information of the target chart data; performing feature extraction on the chart data according to the data type information to obtain data feature information; Screening a target chart type according to the data characteristic information; and constructing a target chart corresponding to the target chart type based on the target chart data and the data characteristic information. Optionally, in some embodiments of the present application, the screening the target chart type according to the data characteristic information includes: determining first matching score information of the data characteristic information aiming at least one preset chart type based on a domain chart rule, wherein the domain chart rule corresponds to the semiconductor domain; outputting second matching score information of the data characteristic information aiming at least one preset chart type through a target model; for each preset chart type, fusing first matching score information and second matching score information corresponding to the preset chart type to obtain fused score information; and screening the target chart type from the preset chart types according to the fusion score information corresponding to each preset chart type. Optionally, in some embodiments of the present application, the domain chart rule includes a configuration condition, a preset chart type, and a corresponding basic confidence score, where the basic confidence score is obtained based on an evaluation of a matching degree of the configuration condition and the preset chart type by industry experience of the semiconductor domain; The determining, based on the domain graph rule, first matching score information of the data feature information for at least one preset graph type includes: Carrying out association matching on the configuration conditions according to the data characteristic information to obtain a condition matching result; and screening at least one preset chart type according to the condition matching result, and determining a basic confidence score corresponding to the preset chart type as the first matching score information. Optionally, in some embodiments of the present application, the constructing a t