Industrial application rule matching method

Abstract

The invention relates to the technical field of rule matching, and discloses an industry application rule matching method, which comprises the following steps of S1, establishing a plurality of industry behavior categories, and classifying industry application rules into different industry behavior categories; s2, according to the industry application rule under each type of industry behavior, establishing an index for each type of industry behavior, and associating the established index with the corresponding industry behavior type; and S3, recording the description of the currently occurring industry behaviors, establishing representative phrases for the currently occurring industry behaviors, and associating the established representative phrases with the currently occurring behaviors. Compared with an existing industry application rule matching method, the method has the advantages that the most conforming industry application rule can be quickly and accurately matched for the currently occurring industry behavior, so that disputes generated by the currently occurring industry behavior can be more quickly handled, and the industry development can be promoted to tend to normal operation activities.

Description

Technical Field

[0001] The present invention relates to the technical field of rule matching, and more specifically discloses an industry application rule matching method. Background Art

[0002] Industry application rules refer to a series of unified regulations and standards formulated within an industry to regulate industry behavior, ensure product or service quality, maintain market order, and promote the healthy development of the industry. These rules and standards usually cover multiple aspects such as technology, management, and services. They are formulated by industry authorities or industry associations and are issued and implemented after approval through relevant procedures. Industry application rules are the necessary management basis for the healthy development of the industry, and they provide a theoretical basis for the management and supervision of the industry. When a dispute occurs over industry behavior, the industry behavior in which the dispute occurs is usually matched with the corresponding industry application rules to handle the dispute. When matching the corresponding industry application rules, experienced industry managers or industry application rule formulation departments usually dispatch staff who are familiar with industry application rules to match the disputed industry behavior with the corresponding industry application rules. For industries with small scale and a small number of industry rules, this method can match the corresponding industry application rules relatively quickly, thereby completing the handling of industry behavior disputes as soon as possible. However, for industries with large scale and a large number of industry rules, this method cannot quickly and accurately match the corresponding industry rules for dispute handling, resulting in the inability to quickly handle industry behavior disputes. Summary of the invention

[0003] The present invention mainly provides an industry application rule matching method, which can solve the problem that corresponding industry application rules cannot be matched quickly and accurately in response to industry behavior disputes.

[0004] To solve the above technical problem, according to one aspect of the present invention, more specifically, an industry application rule matching method comprises the following steps:

[0005] S1. Establish multiple industry behavior categories and classify industry application rules into different industry behavior categories;

[0006] S2. Create an index for each type of industry behavior based on the industry application rules under each type of industry behavior, and associate the created index with the corresponding industry behavior category;

[0007] S3. Record the description of the current industry behavior, create a representative phrase for the current industry behavior, and associate the created representative phrase with the current behavior;

[0008] S4, comparing the representative phrase of the current industry behavior in step S3 with the index of each type of industry behavior established in step S2, and finally finding the industry behavior category that is closest to the current industry behavior;

[0009] S5. After finding the industry behavior category that is closest to the current industry behavior, the industry rules under that industry behavior category will be fed back to the management staff.

[0010] Furthermore, step S1 is specifically as follows: determining industry behaviors and classifying industry behaviors into multiple categories, and then classifying a number of industry application rules into different industry behavior categories according to the management scope of the industry application rules, and saving them.

[0011] Furthermore, in step S2, when indexing each type of industry behavior, each industry rule under each type of industry behavior is segmented, and each industry rule is divided into several entries. If each entry satisfies:

[0012]

[0013] The current term is selected as a component of the index; where n i,j is the number of times the current term i appears in each type of industry behavior j, Z j is the number of entries in each type of industry behavior j, n i∈r,j is the number of industry rules that contain the current term i in each type of industry behavior j, Z r,j is the total number of rules in each category of industry behavior j, and δ is the preset limit threshold; finally, all entries that meet the above conditions are integrated together to form an index of the current category of industry behavior, and at the same time, an association is established.

[0014] Furthermore, in step S3, recording the description of the current industry behavior includes one of text input recording and voice recording. When performing voice recording, the descriptive voice is collected by a microphone, and then the descriptive voice collected by the microphone is converted into text through voice recognition technology and recorded.

[0015] Furthermore, in step S3, when establishing a representative phrase for the current behavior, the pre-trained CNN neural network algorithm is used to remove conjunctions, modal particles, and words without specific meaning in the description of the industry behavior that has occurred, and then the description of the industry behavior that has occurred is segmented, and all the words after segmentation are aggregated to form a representative phrase for the current industry behavior that has occurred.

[0016] Furthermore, in step S4, when matching the industry behavior category that is closest to the currently occurring industry behavior, the representative phrase of the occurring industry behavior is compared with the index of each category of industry behavior, and each component in each category of industry behavior index is compared with each word in the representative phrase of the occurring industry behavior. If:

[0017]

[0018] The industry behavior category corresponding to the current index is the industry category closest to the current industry behavior; among them, pt y k Index each component in y for each industry behavior category, w l is each word in the representative phrase of the industry behavior that occurred, A is the total number of components of the index y for each industry behavior category, and S is the total number of words in the representative phrase of the industry behavior that occurred.

[0019] Furthermore, in step S5, in the process of feeding back to the user the industry application rules under the industry behavior category that is closest to the currently occurring industry behavior, the sentence volume of each industry application rule is obtained, and the sentence volume in the description of the currently occurring industry behavior is obtained, and then the industry application rules are reorganized according to the following conditions and fed back to the user:

[0020] sort(|N st -N r |)

[0021] Among them, N st is the number of sentences describing the current industry behavior, N r To make the statements in each industry application rule coherent, sort() is a sorting function, and the default is descending sorting. This allows the rules that best fit the current industry behavior to be placed in the front for users to directly reference.

[0022] Furthermore, the association of the index with the corresponding industry behavior category in step S2 and the association of the representative phrase with the currently occurring behavior in step S3 are both performed in the form of pointers.

[0023] The beneficial effect of the industry application rule matching method of the present invention is as follows: multiple industry behavior categories are established for the industry behaviors occurring in the current industry, and the industry application rules are classified into different industry behavior categories according to the management scope of the industry application rules, and then an index is established for each type of industry behavior according to the industry application rules under each type of industry behavior. Then the current industry behavior is recorded, and a representative phrase is established for it. By matching and comparing the index with the representative phrase, the industry application rules for the current industry behavior can be found quickly and accurately. Compared with the existing industry application rule matching method, this method can quickly and accurately match the current industry behavior to the most suitable industry application rules, so that disputes arising from the current industry behavior can be handled more quickly, and then the industry development can be promoted to normal operation activities. In addition, when the industry application rules for the current industry behavior are fed back to the management personnel, the industry application rules will be fed back to the management personnel in sequence according to the degree of fit to the current industry behavior, so that the management personnel can further refer to the most suitable industry application rules. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The present invention is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0025] Figure 1 A schematic diagram of the method flow chart. DETAILED DESCRIPTION

[0026] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0027] According to one aspect of the present invention, Figure 1 As shown, an industry application rule matching method is provided. First, the industry behavior is determined and divided into multiple categories. Then, according to the management orientation of the industry application rules, the industry application rules are classified into different industry behavior categories. One category of industry behavior corresponds to several industry application rules, forming an industry behavior category-industry application rule table and storing it. The details are as follows:

[0028]

[0029]

[0030] Then, each industry rule under each type of industry behavior is segmented, and each industry rule is divided into several terms. If each term satisfies:

[0031]

[0032] The current term is selected as a component of the index; where n i,j is the number of times the current term i appears in each type of industry behavior j, Z j is the number of entries in each type of industry behavior j, n i∈r,j is the number of industry rules that contain the current term i in each type of industry behavior j, Z r,j is the total number of rules in each category of industry behavior j, and δ is the preset limit threshold; finally, all entries that meet the above conditions are integrated together to form the index of the current category of industry behavior, and at the same time, the association is established in the form of a pointer, and the industry behavior category-industry application rule table is updated. As shown below:

[0033]

[0034] The description of the current industry behavior is then recorded through text input or voice recognition (voice recognition picks up audio through a microphone and converts the audio into text), and then through a pre-trained CNN neural network algorithm (this algorithm is an existing algorithm, and the CNN model is trained by preparing training set data and test set data. The specific means are currently known and will not be repeated here), the conjunctions, modal particles, and words without specific meanings in the description of the industry behavior are removed, and then the description of the industry behavior is segmented, and all the words after segmentation are aggregated to form a representative phrase of the current industry behavior, and the representative phrase is associated with the description of the current industry behavior in the form of a pointer.

[0035] Then compare the representative phrases of the previous industry behaviors with the index of each type of industry behaviors, and compare each component in the index of each type of industry behaviors with each word in the representative phrases of the industry behaviors that occurred. If:

[0036]

[0037] The industry behavior category corresponding to the current index is the industry category closest to the current industry behavior; Index each component in y for each industry behavior category, w l is each word in the representative phrase of the industry behavior that has occurred, A is the total number of components of each industry behavior category index y, and S is the total number of words in the representative phrase of the industry behavior that has occurred. The industry rules under the industry behavior category that is closest to the current industry behavior are reorganized according to the following conditions and fed back to the user:

[0038] sort(|N st -N r |)

[0039] Among them, Nst is the number of sentences describing the current industry behavior, N r To make the statements in each industry application rule coherent, sort() is a sorting function, and the default is descending sorting. This allows the rules that best fit the current industry behavior to be placed in the front for users to directly reference.

[0040] Of course, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by ordinary technicians in this technical field within the essential scope of the present invention also fall within the protection scope of the present invention.

Claims

1. A method for matching industry application rules, characterized in that: The method specifically comprises the following steps: S1. Establish multiple industry behavior categories and classify industry application rules into different industry behavior categories; S2. Create an index for each type of industry behavior based on the industry application rules under each type of industry behavior, and associate the created index with the corresponding industry behavior category; S3. Record the description of the current industry behavior, create a representative phrase for the current industry behavior, and associate the created representative phrase with the current behavior; S4, comparing the representative phrase of the current industry behavior in step S3 with the index of each type of industry behavior established in step S2, and finally finding the industry behavior category that is closest to the current industry behavior; S5. After finding the industry behavior category that is closest to the current industry behavior, the industry rules under that industry behavior category will be fed back to the management staff.

2. The industry application rule matching method according to claim 1, characterized in that: The step S1 specifically includes: determining industry behaviors and classifying the industry behaviors into multiple categories, and then classifying a number of industry application rules into different industry behavior categories according to the management scope of the industry application rules, and saving them.

3. The industry application rule matching method according to claim 1, characterized in that: In step S2, when indexing each type of industry behavior, each industry rule under each type of industry behavior is segmented, and each industry rule is divided into several terms. If each term satisfies: The current term is selected as a component of the index; where n i,j is the number of times the current term i appears in each type of industry behavior j, Z j is the number of entries in each type of industry behavior j, n i∈r,j is the number of industry rules that contain the current term i in each type of industry behavior j, Z r,j is the total number of rules in each category of industry behavior j, and δ is the preset limit threshold; finally, all entries that meet the above conditions are integrated together to form an index of the current category of industry behavior, and at the same time, an association is established.

4. The industry application rule matching method according to claim 1, characterized in that: In the step S3, recording the description of the current industry behavior includes one of text input recording and voice recording. When performing voice recording, the descriptive voice is collected by a microphone, and then the descriptive voice collected by the microphone is converted into text through voice recognition technology and recorded.

5. The industry application rule matching method according to claim 1, characterized in that: In step S3, when establishing a representative phrase for the current behavior, the pre-trained CNN neural network algorithm is used to remove conjunctions, modal particles, and words without specific meaning in the description of the industry behavior that has occurred, and then the description of the industry behavior that has occurred is segmented, and all the words after segmentation are aggregated to form a representative phrase for the current industry behavior that has occurred.

6. The industry application rule matching method according to claim 1, characterized in that: In step S4, when matching the industry behavior category that is closest to the currently occurring industry behavior, the representative phrase of the occurring industry behavior is compared with the index of each type of industry behavior, and each component in each type of industry behavior index is compared with each word in the representative phrase of the occurring industry behavior. If: The industry behavior category corresponding to the current index is the industry category closest to the current industry behavior; Index each component in y for each industry behavior category, w l is each word in the representative phrase of the industry behavior that occurred, A is the total number of components of the index y for each industry behavior category, and S is the total number of words in the representative phrase of the industry behavior that occurred.

7. The industry application rule matching method according to claim 1, characterized in that: In step S5, in the process of feeding back to the user the industry application rules under the industry behavior category that is closest to the currently occurring industry behavior, the sentence volume of each industry application rule is obtained, and the sentence volume in the description of the currently occurring industry behavior is obtained, and then the industry application rules are reorganized according to the following conditions and fed back to the user: sort(|N st -N r |) Among them, N st is the number of sentences describing the current industry behavior, N r To make the statements in each industry application rule coherent, sort() is a sorting function, and the default is descending sorting. This allows the rules that best fit the current industry behavior to be placed in the front for users to directly reference.

8. The industry application rule matching method according to claim 1, characterized in that: The association of the index with the corresponding industry behavior category in step S2 and the association of the representative phrase with the currently occurring behavior in step S3 are both performed in the form of pointers.

Images