Material data management and processing method based on AI technology

Abstract

The invention belongs to the field of data management and processing, and particularly relates to an AI technology-based material data management and processing method, which comprises the following steps of S1, firstly defining material categories, then defining material names, and transmitting the defined material categories and material names to a material knowledge base for storage; s2, obtaining a material name in the result file, matching a category from a material knowledge base according to the name, and automatically classifying the result file into a subject of the category if the corresponding category is matched; s3, pushing a material name and a knowledge base if the category is not matched, performing semantic recognition and standardization processing on the material through AI, then matching the category, and pushing the material name and the category if the category is matched; according to the method, semantic recognition is utilized, the recognition range is expanded, meanwhile, material data are standardized, a manual intervention part can be automatically updated to a knowledge base, the accuracy rate is high, the efficiency is high, and the cost is low.

Description

Technical Field

[0001] The present invention relates to the field of data management and processing technology, and in particular to a material data management and processing method based on AI technology. Background Art

[0002] Material data refers to information about the various properties and characteristics of materials, including but not limited to chemical composition, physical properties, mechanical properties, thermal properties, electrical properties, etc. These data are of great significance for the research, development, application and safe use of materials.

[0003] There are many types of materials and their names are not standardized, which makes it impossible to group many materials. Manual identification of material types is required. Manual grouping is labor-intensive, with high error rates, low efficiency and high costs. Summary of the invention

[0004] The purpose of the present invention is to solve the shortcomings of the prior art, such as the wide variety of materials and non-standard names, which result in many materials being unable to be grouped, requiring manual identification of material types, large manual grouping workload, high error rate, low efficiency and high cost, and propose a material data management and processing method based on AI technology.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] The material data management and processing method based on AI technology includes the following steps:

[0007] S1. First define the material category, then define the material name, and transfer the defined material category and material name to the material knowledge base for storage;

[0008] S2. Obtain the material name in the achievement document, match the category from the material knowledge base according to the name, and automatically classify the achievement document into the category subject if the corresponding category is matched;

[0009] S3: If no category is matched, the material name and knowledge base will be pushed. The material will be semantically recognized and standardized through AI, and then the category will be matched. If the category is matched, the material name and category will be pushed;

[0010] S4. Receive the pushed material name and category, and automatically classify the achievement document into subjects;

[0011] If no category is matched in S5 and S3, the material name is pushed and the material is regarded as unclassified material. Then the categories are manually classified and the manually classified materials and categories are transferred to the knowledge base for storage.

[0012] Preferably, in S3, before semantic recognition and standardization of materials are performed by AI, material information is established, a standard material format is defined, and storage is performed.

[0013] Preferably, in S3, semantic recognition of materials is performed by AI as follows: the text is segmented according to certain rules and divided into standard names, specifications and models, each word or phrase is tagged with the part of speech, its grammatical role and meaning in the sentence are indicated, the structure in the sentence is analyzed, the grammatical relationship and dependency relationship between each word or phrase are determined, the meaning and context of the text are analyzed, the concepts, entities and events in the text are understood, and the intention and purpose of the text are determined based on the semantic information of the text.

[0014] Preferably, in S1, the material category is first defined, and then the material name is defined, the material name is associated with the material category, and the defined material category and material name are transmitted to the material knowledge base for storage.

[0015] Preferably, in S2, the material name is input into the knowledge base to check whether there is an associated material category for the material name. If there is an associated category, it means that the category is matched, otherwise, the category is not matched.

[0016] Preferably, in S1, the defined material category and material name are transmitted to a material knowledge base, and the material knowledge base receives the material category and material name, then identifies and stores them after identification.

[0017] Preferably, in S1, the storage capacity of the knowledge base is monitored, a storage capacity threshold is set, and when the monitored storage capacity is greater than the capacity threshold, an early warning is issued.

[0018] Preferably, in S2, during matching, multiple matching channels are established, the material data groups stored in the knowledge base are transmitted to the multiple matching channels, the material names in the obtained result files are transmitted to the multiple matching channels, and matching is performed simultaneously.

[0019] Preferably, when any one of the multiple matching channels matches a category, the matching is ended, the matching channel is closed, and the data in the matching channel is cleared.

[0020] Preferably, in S3, the material is standardized as follows: the value of the original data is scaled to a specific interval, usually [0,1], and the conversion function is X1=X-min / max-min, where X1 is the standardized value, X is the original value, min and max are the minimum and maximum values ​​in the data set, respectively.

[0021] In the present invention, the beneficial effects of the material data management and processing method based on AI technology are:

[0022] The present invention utilizes semantic recognition to expand the recognition scope and standardize material data at the same time. The manual intervention part will be automatically updated to the knowledge base with high accuracy, high efficiency and low cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a flow chart of the material data management and processing method based on AI technology proposed in the present invention. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0025] Embodiment 1

[0026] Reference Figure 1 , a material data management and processing method based on AI technology, comprising the following steps:

[0027] S1. First define the material category, then define the material name, and transfer the defined material category and material name to the material knowledge base for storage;

[0028] S2. Obtain the material name in the achievement document, match the category from the material knowledge base according to the name, and automatically classify the achievement document into the category subject if the corresponding category is matched;

[0029] S3: If no category is matched, the material name and knowledge base will be pushed. The material will be semantically recognized and standardized through AI, and then the category will be matched. If the category is matched, the material name and category will be pushed;

[0030] S4. Receive the pushed material name and category, and automatically classify the achievement document into subjects;

[0031] If no category is matched in S5 and S3, the material name is pushed and the material is regarded as unclassified material. Then the categories are manually classified and the manually classified materials and categories are transferred to the knowledge base for storage.

[0032] In this embodiment, in S3, before the semantic recognition and standardization of the material are performed by AI, the material information is established, the standard format of the material is defined, and the material is stored.

[0033] In this embodiment, in S3, semantic recognition of materials is performed by AI as follows: the text is segmented according to certain rules and divided into standard names, specifications and models, each word or phrase is tagged with a part of speech, its grammatical role and meaning in the sentence are indicated, the structure in the sentence is analyzed, the grammatical relationship and dependency relationship between each word or phrase are determined, the meaning and context of the text are analyzed, the concepts, entities and events in the text are understood, and the intention and purpose of the text are determined based on the semantic information of the text.

[0034] In this embodiment, in S1, the material category is first defined, and then the material name is defined, the material name is associated with the material category, and the defined material category and material name are transmitted to the material knowledge base for storage.

[0035] In this embodiment, in S2, the material name is input into the knowledge base to check whether there is an associated material category for the material name. If there is an associated category, it means that the category is matched, otherwise, the category is not matched.

[0036] In this embodiment, in S1, the defined material category and material name are transmitted to the material knowledge base, and the material knowledge base receives the material category and material name, then identifies and stores them.

[0037] In this embodiment, in S1, the storage capacity of the knowledge base is monitored, and a storage capacity threshold is set. When the monitored storage capacity is greater than the capacity threshold, an early warning is issued.

[0038] In this embodiment, in S2, during matching, multiple matching channels are established, the material data stored in the knowledge base are grouped and transmitted to the multiple matching channels, the material names in the obtained result files are transmitted to the multiple matching channels, and matching is performed simultaneously.

[0039] In this embodiment, when any one of the multiple matching channels matches a category, the matching is ended, the matching channel is closed, and the data in the matching channel is cleared.

[0040] In this embodiment, in S3, the material is standardized as follows: the value of the original data is scaled to a specific interval, usually [0,1], and the conversion function is X1=X-min / max-min, where X1 is the standardized value, X is the original value, and min and max are the minimum and maximum values ​​in the data set, respectively.

[0041] Embodiment 2

[0042] The difference between this embodiment and the first embodiment is that:

[0043] The material data management and processing method based on AI technology includes the following steps:

[0044] S1. First define the material category, then define the material name, and transfer the defined material category and material name to the material knowledge base for storage;

[0045] S2. Obtain the material name in the result file, match the category from the material knowledge base according to the name, and automatically classify the result file into the category subject if the corresponding category is matched. Meanwhile, monitor the matching time, and issue an early warning when the matching time is too long.

[0046] S3: If no category is matched, the material name and knowledge base will be pushed. The material will be semantically recognized and standardized through AI, and then the category will be matched. If the category is matched, the material name and category will be pushed;

[0047] S4. Receive the pushed material name and category, and automatically classify the achievement document into subjects;

[0048] If no category is matched in S5 and S3, the material name is pushed and the material is regarded as unclassified material. Then the categories are manually classified and the manually classified materials and categories are transferred to the knowledge base for storage.

[0049] Embodiment 3

[0050] The difference between this embodiment and the first embodiment is that:

[0051] The material data management and processing method based on AI technology includes the following steps:

[0052] S1. First define the material category, then define the material name, and transfer the defined material category and material name to the material knowledge base for storage;

[0053] S2. Obtain the material name in the achievement document, match the category from the material knowledge base according to the name, and automatically classify the achievement document into the category subject if the corresponding category is matched;

[0054] S3: If no category is matched, the material name and knowledge base will be pushed. The material will be semantically recognized and standardized through AI, and then the category will be matched. If the category is matched, the material name and category will be pushed;

[0055] S4. Receive the pushed material name and category, and automatically classify the achievement document into subjects;

[0056] If no category is matched in S5 or S3, the material name is pushed and the material is regarded as unclassified. The manually classified materials and categories are then transferred to the knowledge base for storage. The manually classified materials and category data are registered for traceability and marked with the identity information of the manual classifier for subsequent traceability.

[0057] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A material data management and processing method based on AI technology, characterized in that: The following steps are involved: S1. First define the material category, then define the material name, and transfer the defined material category and material name to the material knowledge base for storage; S2. Obtain the material name in the achievement document, match the category from the material knowledge base according to the name, and automatically classify the achievement document into the category subject if the corresponding category is matched; S3: If no category is matched, the material name and knowledge base will be pushed. The material will be semantically recognized and standardized through AI, and then the category will be matched. If the category is matched, the material name and category will be pushed; S4. Receive the pushed material name and category, and automatically classify the achievement document into subjects; If no category is matched in S5 and S3, the material name is pushed and the material is regarded as unclassified material. Then the categories are manually classified and the manually classified materials and categories are transferred to the knowledge base for storage.

2. The material data management and processing method based on AI technology according to claim 1 is characterized in that: In S3, before semantic recognition and standardization of materials are performed by AI, material information is established, the standard format of materials is defined, and storage is performed.

3. The material data management and processing method based on AI technology according to claim 2 is characterized in that: In S3, semantic recognition of materials by AI is performed as follows: the text is segmented according to certain rules and divided into standard names, specifications and models, each word or phrase is tagged with the part of speech, its grammatical role and meaning in the sentence are indicated, the structure in the sentence is analyzed, the grammatical relationship and dependency relationship between each word or phrase are determined, the meaning and context of the text are analyzed, the concepts, entities and events in the text are understood, and the intention and purpose of the text are determined based on the semantic information of the text.

4. The material data management and processing method based on AI technology according to claim 3 is characterized in that: In S1, the material category is first defined, and then the material name is defined, the material name is associated with the material category, and the defined material category and material name are transmitted to the material knowledge base for storage.

5. The material data management and processing method based on AI technology according to claim 4 is characterized in that: In S2, the material name is input into the knowledge base to check whether there is an associated material category for the material name. If there is an associated category, it means that the category is matched, otherwise, the category is not matched.

6. The material data management and processing method based on AI technology according to claim 5 is characterized in that: In S1, the defined material category and material name are transmitted to the material knowledge base, and the material knowledge base receives the material category and material name, then identifies and stores them.

7. The material data management and processing method based on AI technology according to claim 6 is characterized in that: In S1, the storage capacity of the knowledge base is monitored, and a storage capacity threshold is set. When the monitored storage capacity is greater than the capacity threshold, an early warning is issued.

8. The material data management and processing method based on AI technology according to claim 7 is characterized in that: In S2, during matching, multiple matching channels are established, the material data stored in the knowledge base are grouped and transmitted to the multiple matching channels, the material names in the obtained result files are transmitted to the multiple matching channels, and matching is performed simultaneously.

9. The material data management and processing method based on AI technology according to claim 8 is characterized in that: When any of the multiple matching channels matches a category, the matching ends, the matching channel is closed, and the data in the matching channel is cleared.

10. The material data management and processing method based on AI technology according to claim 9 is characterized in that: In S3, the material is standardized as follows: the value of the original data is scaled to a specific interval, usually [0,1], and the conversion function is X1=X-min / max-min, where X1 is the standardized value, X is the original value, and min and max are the minimum and maximum values ​​in the data set, respectively.

Images