A method and device for extracting and distributing information suitable for power supply station

Abstract

The present invention provides an information extraction and distribution method and device suitable for a power supply station, which compares the current language text with the reference language text, and divides the language text information of the current system user into regions using the change detection segmentation technology, that is, calculates the similarity between the current language text and the reference language text, obtains the unchanged region and the changed region, searches for the keywords of the reference language text in the unchanged region, retains the existing keywords, and extracts the keywords in the changed region at the same time, and then uses the hash function to associate the keywords with the business information. The present invention divides the current language text using the change detection segmentation technology, retains the keywords existing in the unchanged region, and only needs to extract the keywords from the changed region to complete the recognition of the language text, which reduces the extraction range of the keywords, makes the extraction of the keywords more accurate, and improves the business management efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of business systems in the electric power industry, and in particular to an information extraction and distribution method and device suitable for a power supply station. Background Art

[0002] The power industry plays a vital role in a country's economic and social development. Power companies will face more and more external supervision and social attention. With the continuous deepening of the national power system reform, the internal and external environment faced by power supply companies has undergone tremendous changes. The new environment and new situation not only bring new challenges to power supply companies, but also new opportunities, and put forward higher new requirements for the operation and management of power supply companies. As the most basic unit of power supply enterprises, the power supply station is the most direct unit for power grid operation and management of power supply bureaus to face customers. It is responsible for the power supply, line equipment maintenance, meter reading and collection management, electricity fee recovery, line loss management, and customer service tasks of power supply companies at the grassroots level. The quality of the management of grassroots power supply stations directly affects the internal economic benefits and external social image of power supply companies, especially county-level power supply companies. Therefore, it is the basis for ensuring the orderly and harmonious development of power supply companies to do a good job in the management of grassroots power supply stations.

[0003] However, in the existing business management process of power supply stations, due to the wide range and complexity of power supply station business and the large number of people involved, the management efficiency of business information of users and staff is low and cannot meet the work needs of users and internal staff. Existing business management methods include business data visualization, business integration and other technologies. Although they have improved the business management efficiency of power supply stations, due to factors such as excessive business data, the accuracy of business management is low and cannot be widely used. Summary of the invention

[0004] In view of this, the present invention aims to solve the problem of low business management efficiency due to excessive data in the existing business system.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] The present invention provides an information extraction and distribution method applicable to a power supply station, comprising the following steps:

[0007] Obtain the language text and reference language text provided by the current system user;

[0008] Calculating the similarity between the language text provided by the current system user and the corresponding part of the reference language text;

[0009] Calculate the average value of the similarity of each corresponding part, and determine whether the average value is greater than a preset similarity threshold. If so, classify the corresponding part of the language text provided by the current system user into an unchanged area; if not, classify it into a changed area;

[0010] Extract keywords from the reference language text and search for keywords in unchanged areas of the language text provided by the current system user. If the keywords exist, retain them;

[0011] Extract keywords from the changed areas in the language text provided by the current system user;

[0012] Based on the retained keywords and the extracted keywords, a hash function is used to associate the keywords with the business information.

[0013] Furthermore, obtaining the language text and reference language text provided by the current system user specifically includes:

[0014] Get the language text provided by the current system user;

[0015] Get the current sequence number of the current system user in the system;

[0016] Based on the current sequence number, the language text of the system user before the current system user is obtained as a reference language text.

[0017] Furthermore, based on the current sequence value, obtaining the language text of the system user before the current system user as a reference language text specifically includes:

[0018] When the current sequence number is greater than 1 and less than or equal to the first threshold, the language texts of all system users whose sequence numbers are less than the current sequence number are used as reference texts;

[0019] When the current sequence number is greater than the first threshold and less than or equal to the second threshold, randomly select system users whose sequence numbers are less than the current sequence number and whose number is equal to the first threshold, and obtain the corresponding language texts as reference language texts;

[0020] When the current sequence number is greater than or equal to the second threshold, an unchanged region in the language text of a system user is randomly selected from all system users whose sequence numbers are less than the second threshold, and the unchanged region is compared with the unchanged region in the language text of other system users whose sequence numbers are less than the second threshold for similarity, and the unchanged region in the currently selected language text is removed from all changed parts and used as the reference language text.

[0021] Furthermore, the similarity specifically includes:

[0022] Glyph similarity and semantic similarity.

[0023] Furthermore, the similarity between the language text provided by the current system user and the corresponding part of the reference language text is calculated as follows:

[0024] The similarity formula is used to calculate the similarity between the language text provided by the current system user and the corresponding part of the reference language text. The similarity formula is as follows:

[0025]

[0026] Where z is the error parameter, A is the glyph similarity, B is the semantic similarity, and 1 is the preset similarity threshold.

[0027] Furthermore, the error parameter ranges from 0.8 to 1.

[0028] Furthermore, the system users include:

[0029] Users and staff.

[0030] Furthermore, extracting keywords from the reference language text specifically includes:

[0031] When the reference language text is obtained from a language text provided by a user, the extracted keywords include the user, the user name, and the specific name of the service handled by the user;

[0032] When the reference language text is obtained from the language text provided by the staff, the extracted keywords include meter information, material information and approval information.

[0033] In a second aspect, the present invention provides an information extraction and distribution device applicable to a power supply station, the device comprising a processor and a memory:

[0034] The memory is used to store the computer program and send the instructions of the computer program to the processor;

[0035] The processor executes the information extraction and distribution method applicable to a power supply station according to the instructions of the computer program.

[0036] In a third aspect, the present invention provides a computer storage medium having a computer program stored thereon, and when the computer program is executed by a processor, an information extraction and distribution method applicable to a power supply station as in the first aspect is implemented.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] The present invention provides an information extraction and distribution method and device suitable for a power supply station, which compares the current language text with the reference language text, and divides the language text information of the current system user into regions using the change detection segmentation technology, that is, calculates the similarity between the current language text and the reference language text, obtains the unchanged region and the changed region, searches for the keywords of the reference language text in the unchanged region, retains the existing keywords, and extracts the keywords in the changed region at the same time, and then uses the hash function to associate the keywords with the business information. The present invention divides the current language text using the change detection segmentation technology, retains the keywords existing in the unchanged region, and only needs to extract the keywords from the changed region to complete the recognition of the language text, which reduces the extraction range of the keywords, makes the extraction of the keywords more accurate, and improves the business management efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0040] Figure 1 A flowchart of an information extraction and distribution method applicable to a power supply station is provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0041] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] The power industry plays a vital role in a country's economic and social development. Power companies will face more and more external supervision and social attention. With the continuous deepening of the national power system reform, the internal and external environment faced by power supply companies has undergone tremendous changes. The new environment and new situation not only bring new challenges to power supply companies, but also new opportunities, and put forward higher new requirements for the operation and management of power supply companies. As the most basic unit of power supply enterprises, the power supply station is the most direct unit for power grid operation and management of power supply bureaus to face customers. It is responsible for the power supply, line equipment maintenance, meter reading and collection management, electricity fee recovery, line loss management, and customer service tasks of power supply companies at the grassroots level. The quality of the management of grassroots power supply stations directly affects the internal economic benefits and external social image of power supply companies, especially county-level power supply companies. Therefore, it is the basis for ensuring the orderly and harmonious development of power supply companies to do a good job in the management of grassroots power supply stations.

[0043] However, in the existing business management process of power supply stations, due to the wide range and complexity of power supply station business and the large number of people involved, the management efficiency of business information of users and staff is low and cannot meet the work needs of users and internal staff. Existing business management methods include business data visualization, business integration and other technologies. Although they have improved the business management efficiency of power supply stations, due to factors such as excessive business data, the accuracy of business management is low and cannot be widely used.

[0044] In the daily work process, a large number of query language texts are received from customers and staff inside the power grid every day. In the query process of the system, the text is generally analyzed by extracting keywords. However, since the language texts provided by customers and staff have little difference in other parts except the specific information to be queried, if keyword retrieval is performed on all language text information, the workload will be very large, the operation requirements of the system are high, and the query efficiency is also reduced. For this reason, the present invention uses change detection segmentation technology to segment the part of the language text that has changed, divide it into change areas, and only perform keyword extraction on the change areas to reduce the amount of business data that needs to be retrieved and improve work efficiency.

[0045] An embodiment of an information extraction and distribution method applicable to a power supply station of the present invention is described in detail below.

[0046] See also Figure 1 This embodiment provides an information extraction and distribution method applicable to a power supply station, including:

[0047] S101: Acquire the language text and reference language text provided by the current system user;

[0048] It should be noted that since the information that users (customers) and internal staff of the power grid need to query is quite different, their language texts need to be considered separately. When performing information query, the system first needs to input personal information. For the language text provided by the first user or internal staff of the power grid, keywords need to be extracted from the entire text information.

[0049] Specifically, when the user is a user, the extracted keywords include the user name and the specific name of the service handled by the user (such as meter reading, meter damage, meter installation, payment information, etc.).

[0050] When they are staff members, the extracted keywords include meter information (electricity charges, meter calibration data, meter data, ledgers, etc.), material information (material, price, buying and selling, etc.), approval information (process progress of various material reports), etc.

[0051] In addition, when obtaining the reference language text, it is necessary to obtain the current sequence number of the current system user in the system, and based on the current sequence number, obtain the language text of the system user before the current system user as the reference language text.

[0052] Specifically, when the current sequence number is greater than 1 and less than or equal to the first threshold, the language texts of all system users whose sequence numbers are less than the current sequence number are used as reference texts;

[0053] When the current sequence number is greater than the first threshold and less than or equal to the second threshold, randomly select system users whose sequence numbers are less than the current sequence number and whose number is equal to the first threshold, and obtain the corresponding language texts as reference language texts;

[0054] When the current sequence number is greater than or equal to the second threshold, an unchanged region in the language text of a system user is randomly selected from all system users whose sequence numbers are less than the second threshold, and the unchanged region is compared with the unchanged region in the language text of other system users whose sequence numbers are less than the second threshold for similarity, and the unchanged region in the currently selected language text is removed from all changed parts and used as the reference language text.

[0055] S102: Calculating the similarity between the language text provided by the current system user and the corresponding part of the reference language text;

[0056] It is understandable that, according to the previous step, there can be multiple reference texts. When calculating similarity, it is necessary to calculate between the current language text and multiple reference texts to obtain multiple sets of similarity results. Each time the similarity is calculated, it is calculated between the corresponding parts of the two language texts.

[0057] The specific calculation process is to set the information feature of the current language text to M(x, y) and the information feature of the reference language text to O(x, y). Among them, x refers to the glyph of natural language, y refers to the language meaning, and X is set to 1 when the glyphs are inconsistent. For the language meaning, since the semantic similarity is distributed between 0 and 100%, the semantic similarity is expressed proportionally in the form of 0 to 1.

[0058] The language text information is first segmented according to the sentence components, and the corresponding parts of the first language text information and the second language text information are analyzed for glyph and semantic similarity. The glyph similarity value is recorded as A, and the semantic similarity value is recorded as B. If it satisfies When the search is complete, the part is divided into the unchanged area, and the unchanged information corresponding to the unchanged area (composed of parts divided according to sentence components, in addition to the keywords required for search, there are other useless sentence components, which have no substantive meaning for keyword retrieval of the language text) is temporarily stored.

[0059] The error parameter z is introduced into the above similarity calculation formula, the main purpose of which is to correct the calculation error. The value range of the error parameter z is 0.8-1, and 0.8 is taken in this embodiment.

[0060] S103: Calculate the average value of the similarity of each corresponding part, and determine whether the average value is greater than a preset similarity threshold. If so, classify the corresponding part of the language text provided by the current system user into an unchanged area; if not, classify it into a changed area;

[0061] It should be noted that since there are multiple reference texts, for a certain part of the current language text, there may be multiple similarity results calculated, so the multiple similarity results of the corresponding part of the current language text are calculated and the average value is taken. When the average value is greater than the preset similarity threshold, the part corresponding to the average value is divided into the unchanged area.

[0062] S104: extracting keywords from the reference language text and searching for keywords in unchanged areas of the language text provided by the current system user, and retaining the keywords if they exist;

[0063] S105: extracting keywords from the changed area in the language text provided by the current system user;

[0064] S106: Based on the retained keywords and the extracted keywords, the keywords are associated with the business information using a hash function;

[0065] It should be noted that for users, the association includes associating electricity charges with the electricity charges in the user's name, and associating meter damage with the treatment measures when the meter is damaged. For staff, the associated information includes approval information pointing to the project's work progress, and meter data pointing to all meter data tables in the area where the staff is responsible.

[0066] The present embodiment provides an information extraction and distribution method and device suitable for a power supply station, by comparing the current language text with the reference language text, using the change detection segmentation technology to divide the language text information of the current system user into regions, that is, calculating the similarity between the current language text and the reference language text, obtaining the unchanged region and the changed region, searching the keywords of the reference language text in the unchanged region, retaining the existing keywords, and extracting the keywords in the changed region at the same time, and then using the hash function to associate the keywords with the business information. The present invention divides the current language text using the change detection segmentation technology, retains the keywords existing in the unchanged region, and only needs to extract the keywords from the changed region to complete the recognition of the language text, which reduces the extraction range of keywords, makes the extraction of keywords more accurate, and improves the efficiency of business management.

[0067] In addition, this method also distinguishes the business that needs to be handled by users and power system staff, making the management of business information more targeted; it uses segmented form to transform reference texts to ensure the accuracy of business management in various situations.

[0068] The following is a detailed introduction to an information extraction and distribution method applicable to a power supply station provided in this embodiment with reference to examples.

[0069] Set the first threshold to 10 and the second threshold to 100.

[0070] When the current sequence number of the current system user is 1, all keywords in his language text are extracted and business information is associated using a hash function;

[0071] When the current sequence number n of the current system user is any one of 2-10, the language texts of the previous n-1 system users are used as references to extract the change information in the current language text using the change detection segmentation technology.

[0072] Specifically, the similarity Z1, Z2, ... Z between the current language text and the reference language text is calculated according to the similarity formula. n-1, and calculate the average value between these similarities. When the average value is greater than the preset similarity threshold, the corresponding current language text part is divided into the unchanged area, and the keywords in the reference language text are extracted and searched in the unchanged area of ​​the current language text. If the keyword exists, it is retained. The remaining area of ​​the current language text is the changed area, and the keywords can be extracted from the changed area.

[0073] When the current sequence number n of the current system user is greater than 10 and less than or equal to 100, the language texts of 10 system users are randomly selected from the first n-1 system users as reference texts, and the regions are divided and keywords are extracted according to the above method.

[0074] When the current sequence number n of the current system user is greater than 100, one of the unchanged areas corresponding to the language texts provided by the first 100 system users is selected, and then compared with the unchanged areas of the language texts of the remaining system users, and the changed part is proposed with reference to the aforementioned similarity calculation scheme, and the language text corresponding to the currently selected unchanged area is used as the final reference language text P. Subsequent system users all use the reference language text P as a reference to divide areas and extract keywords.

[0075] After extracting the keywords, a hash function is used to match the keywords with the business information.

[0076] Furthermore, in order to ensure the accuracy of the division of unchanged areas and changed areas in the business management system, the reference text needs to be continuously updated according to the above steps of calculating similarity to ensure the validity of the reference text and improve the accuracy of information extraction.

[0077] The above is a detailed introduction to an embodiment of an information extraction and distribution method applicable to a power supply station of the present invention. The following is a detailed introduction to an embodiment of an information extraction and distribution device applicable to a power supply station of the present invention.

[0078] This embodiment provides an information extraction and distribution device applicable to a power supply station, the device including a processor and a memory:

[0079] The memory is used to store the computer program and send the instructions of the computer program to the processor;

[0080] The processor executes an information extraction and distribution method applicable to a power station as described in the above embodiment according to the instructions of the computer program.

[0081] The above is a detailed introduction to an embodiment of an information extraction and distribution device applicable to a power supply station of the present invention. The following is a detailed introduction to an embodiment of a computer storage medium of the present invention.

[0082] This embodiment provides a computer storage medium, on which a computer program is stored. When the computer program is executed by a processor, an information extraction and distribution method applicable to a power supply station as in the aforementioned embodiment is implemented.

[0083] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for extracting and distributing information applicable to a power supply station, characterized in that: The steps include: Obtain the language text and reference language text provided by the current system user; Calculating the similarity between the language text provided by the current system user and the corresponding part of the reference language text; Calculate the average value of the similarity of each corresponding part, and determine whether the average value is greater than a preset similarity threshold. If so, classify the corresponding part of the language text provided by the current system user into an unchanged area; if not, classify it into a changed area; Extracting keywords from the reference language text and searching for the keywords in unchanged areas of the language text provided by the current system user, and retaining the keywords if they exist; Extracting keywords from the changed areas in the language text provided by the current system user; Based on the retained keywords and the extracted keywords, a hash function is used to associate the keywords with the business information.

2. The information extraction and distribution method applicable to a power supply station according to claim 1, characterized in that: The obtaining of the language text and the reference language text provided by the current system user specifically includes: Get the language text provided by the current system user; Obtaining the current sequence number of the current system user in the system; Based on the current sequence number, the language text of the system user before the current system user is obtained as a reference language text.

3. The information extraction and distribution method applicable to a power supply station according to claim 2, characterized in that: The acquiring, based on the current sequence number, the language text of the system user before the current system user as the reference language text specifically includes: When the current sequence number is greater than 1 and less than or equal to a first threshold, the language texts of all system users whose sequence numbers are less than the current sequence number are used as reference texts; When the current sequence number is greater than a first threshold and less than or equal to a second threshold, randomly selecting system users whose sequence numbers are less than the current sequence number and whose number is equal to the first threshold, and obtaining corresponding language texts as reference language texts; When the current sequence number is greater than or equal to a second threshold, an unchanged region in the language text of a system user is randomly selected from all system users whose sequence numbers are less than the second threshold, and a similarity comparison is performed with unchanged regions in the language texts of other system users whose sequence numbers are less than the second threshold. The unchanged region in the currently selected language text is removed from all changed parts and used as a reference language text.

4. The information extraction and distribution method applicable to a power supply station according to claim 1, characterized in that: The similarity specifically includes: Glyph similarity and semantic similarity.

5. The information extraction and distribution method applicable to a power supply station according to claim 1, characterized in that: The current system users specifically include: Users and staff.

6. The information extraction and distribution method applicable to a power supply station according to claim 5, characterized in that: The extracting of keywords from the reference language text specifically includes: When the reference language text is obtained from a language text provided by a user, the extracted keywords include the user name and the specific name of the service handled by the user; When the reference language text is obtained from a language text provided by a staff member, the extracted keywords include meter information, material information and approval information.

7. An information extraction and distribution device suitable for a power supply station, characterized in that: The device comprises a processor and a memory: The memory is used to store a computer program and send instructions of the computer program to the processor; The processor executes the information extraction and distribution method applicable to a power supply station according to any one of claims 1 to 6 according to the instructions of the computer program.

8. A computer storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for extracting and distributing information applicable to a power supply station as described in any one of claims 1 to 6 is implemented.

Images