Industrial Brain Data Analysis Platform Based on Industrial Knowledge Graph

Abstract

The invention discloses an industrial brain data analysis platform based on an industrial knowledge graph, belonging to the technical field of relay protection, including an enterprise data collection module, an industrial graph construction module, an upstream and downstream association matching module, and a priority recommendation determination module; the enterprise data collection module collects data information of enterprises settled in the platform, including enterprise names, enterprise business licenses, enterprise locations, enterprise invoices, bills of exchange, and check data, establishes files of enterprises through #imgabs0#, records the data information of the settled enterprises, and transmits the data into the industrial graph construction module. The industrial brain data analysis platform based on the industrial knowledge graph of the invention can realize enterprise classification, extraction of electronic bill data, and association of upstream and downstream enterprises as a whole, facilitating enterprises settled in the platform to discover potential suppliers and distributors, and enhancing the practicability and functionality of the platform.

Description

Technical Field

[0001] The present invention belongs to the technical field of electronic data processing, and specifically is an industrial brain data analysis platform based on industrial knowledge graph. Background Art

[0002] Electronic bills, also known as electronic tickets, are a form of voucher that digitizes traditional paper bills. They use computer technology and network communication methods to record and store bill information in digital form to achieve bill generation, transmission, payment, redemption, settlement, and archiving. With the rapid development of information technology, transactions between enterprises are becoming increasingly frequent and complex. Electronic bills such as invoices, bills of exchange, and checks have become an important part of corporate financial management.

[0003] Electronic invoices not only record detailed transaction information, but also include rich business data, such as transaction amount, type of purchased goods, transaction time, etc. The existing data processing and analysis platform cannot associate enterprises with upstream and downstream companies based on the electronic invoice information of business transactions and the type of enterprise. At the same time, it cannot assist enterprises in selecting suppliers and distributors based on their own scale, and cannot deeply explore the potential value of data. There are problems of low practicality and functionality;

[0004] In response to the above, this case proposes an industrial brain data analysis platform based on industrial knowledge graph to solve the above technical problems. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes an industrial brain data analysis platform based on an industrial knowledge graph to solve the above technical problems by improving the detection method and processing method.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] The industry brain data analysis platform based on the industry knowledge graph includes enterprise data collection module, industry graph construction module, upstream and downstream association matching module, and priority recommendation and judgment module;

[0008] The enterprise data collection module collects data information of enterprises settled in the platform, including enterprise name, enterprise business license, enterprise location, enterprise invoice, bill of exchange, and check data. Establish enterprise files, record the data information of settled enterprises, and transfer data into the industrial map construction module;

[0009] The industrial map construction module classifies the bill data of enterprises based on the types of enterprises, extracts different data of different bills for different types of enterprises, generates labels and constructs an industrial map;

[0010] The upstream and downstream association matching module determines the enterprise credit rating by combining the enterprise's own electronic bill data information, conducts upstream and downstream association matching analysis on the enterprise, and generates upstream files and downstream files for different enterprises;

[0011] The priority recommendation and determination module calculates the enterprise purchasing power and productivity levels by combining the recent electronic bill information of enterprises in the upstream and downstream files, and sorts the enterprises in the upstream and downstream files in combination with the monthly order quantity of the enterprises in the files.

[0012] Furthermore, the industrial map construction module classifies the bill data of enterprises based on the types of enterprises, extracts different data of different bills for different types of enterprises, generates labels to construct an industrial map, and its specific steps are as follows:

[0013] Based on the business licenses of the enterprises settled on the platform collected by the enterprise data collection module, the enterprises are classified into different types, including basic raw material industries, manufacturing industries, financial service industries, commercial service industries, and other industries. The collection scope of the electronic bill information of different types of enterprises is determined, the characters in the electronic bills are extracted through OCR technology, and keyword fields are respectively extracted by combining keyword matching in NLP technology with the types of enterprises, where:

[0014] For basic raw material industries, extract the commodity name, amount, date, customer name, and bill status in the sales invoices in the enterprise's historical electronic bills;

[0015] For manufacturing industries, extract the commodity name, amount, date, customer name in the sales invoices in the enterprise's historical electronic bills, the commodity name, amount, date, customer name in the purchase invoices, and the bill status;

[0016] For financial service industries, extract the bill amount and check amount in the enterprise's historical electronic bills;

[0017] For commercial service industries, extract the commodity name, amount, date, customer name in the purchase invoices in the enterprise's historical electronic bills, and the bill status;

[0018] For other industries, no electronic bill data extraction is performed;

[0019] Combine the bill data extracted from different types of enterprises in the platform to generate enterprise labels and establish an industrial map.

[0020] Furthermore, the step of combining the bill data extracted from different types of enterprises in the platform to generate enterprise labels and establish an industrial map is as follows:

[0021] Create a keyword list for each category based on the known commodity names. By using the fuzzy matching method and the Jaccard similarity, match the data extracted from the e-bills of different types of enterprises with the keyword list, where:

[0022] For the basic raw material industry, divide the product type labels according to the commodity names in the sales invoices, including metal and ore categories, non-metal ore categories, chemical raw material categories, building material categories, agricultural raw material categories, energy raw material categories, rare and precious raw material categories, and other basic raw material categories. Based on the commodity names in the enterprise sales invoices, record the product type labels in the enterprise archives of the basic raw material industry;

[0023] For the manufacturing industry, divide the raw material type labels according to the commodity names in the purchase invoices, including metal and ore categories, non-metal ore categories, chemical raw material categories, building material categories, agricultural raw material categories, energy raw material categories, rare and precious raw material categories, and other basic raw material categories. At the same time, divide the product type labels according to the commodity names in the sales invoices, including mechanical equipment categories, electronic and electrical equipment categories, automotive and parts categories, chemical product categories, metal product categories, building materials and household categories, textile and clothing categories, food and packaging categories, medical device categories, and other manufacturing categories. Based on the commodity names in the enterprise purchase invoices and sales invoices, record the raw material type labels and product type labels in the enterprise archives of the manufacturing industry;

[0024] For the commercial service industry, divide the product type labels according to the commodity names in the purchase invoices, including mechanical equipment categories, electronic and electrical equipment categories, automotive and parts categories, chemical product categories, metal product categories, building materials and household categories, textile and clothing categories, food and packaging categories, medical device categories, and other manufacturing categories. Record the product type labels in the enterprise archives of the commercial service industry.

[0025] Furthermore, the upstream and downstream association matching module combines the e-bill data information of the enterprise itself to determine the enterprise credit rating, conducts upstream and downstream association matching analysis on the enterprise, and generates the upstream archives and downstream archives of different enterprises. The specific steps are as follows:

[0026] For enterprises in the basic raw material industry, manufacturing industry, and commercial service industry, determine the enterprise credit rating based on the bill status;

[0027] For enterprises in the basic raw material industry, manufacturing industry, financial service industry, and commercial service industry, combine the e-bill data information and credit rating of the enterprise, and conduct upstream and downstream association matching analysis of the enterprises within the platform based on the enterprise type respectively, and generate the upstream archives and downstream archives of different enterprises.

[0028] Furthermore, for enterprises in the basic raw material industry, manufacturing industry, and commercial service industry, determining the enterprise credit rating based on the bill status, the specific steps are as follows:

[0029] Based on the e-bill data identified in the industrial map construction module, collect the within the year and to the draft data status of the enterprise in the year of , and classify the credit rating of the current enterprise. The draft data status includes overdue and normal. The specific steps are as follows:

[0030] When the enterprise there is no overdue in the draft data status within the year, and to there is no overdue in the draft data status of the enterprise in the year of , record the enterprise label as excellent credit in the enterprise file;

[0031] When the enterprise there is no overdue in the draft data status within the year, and to there is overdue in the draft data status of the enterprise in the year of , record the enterprise label as average credit in the enterprise file;

[0032] When the enterprise there is overdue in the draft data status within the year, record the enterprise label as poor credit in the enterprise file.

[0033] Furthermore, for enterprises in the basic raw material industry, manufacturing industry, financial service industry, and commercial service industry, combining the enterprise e-bill data information, enterprise purchasing power, productivity, and credit rating, conduct an upstream and downstream association matching analysis of enterprises within the platform based on the enterprise type, and generate upstream and downstream files for different enterprises. The specific steps are as follows:

[0034] For the financial service industry, count the draft and check amount data of financial service industry enterprises within the platform in the past 2 years, respectively obtain the maximum draft amount, minimum draft amount, maximum check amount, and minimum check amount, and calculate the maximum and minimum transactions of the current enterprise:

[0035] ;

[0036] ;

[0037] Obtain the enterprise customer range by combining the maximum and minimum transactions of the current enterprise :

[0038] ;

[0039] Count the purchase invoice amounts of the basic raw material industry, manufacturing industry, and commercial service industry in the past 2 years, respectively obtain the maximum purchase amount and minimum purchase amount, and obtain the enterprise transaction amount range , according to the current financial service industry enterprise's , establish the downstream files of enterprises, and classify the enterprise files into the downstream files of the current financial service enterprises;

[0040] For the manufacturing industry, establish downstream files and upstream files in the enterprise files respectively. Based on the raw material type labels in the enterprise files, classify the enterprise files with the same product type labels as the raw material type labels into the upstream files of the current enterprise. Based on the product type labels in the enterprise files, classify the enterprise files with the same raw material type labels as the product type labels into the downstream files of the current enterprise;

[0041] For the basic raw material industry, establish downstream files in the enterprise files. Based on the product type labels in the enterprise files, classify the enterprise files with the same raw material type labels as the product type labels into the downstream files of the current enterprise;

[0042] For the commercial service industry, establish upstream files in the enterprise files. Based on the raw material type labels in the enterprise files, classify the enterprise files with the same product type labels as the raw material type labels into the upstream files of the current enterprise;

[0043] Based on the current enterprise creditworthiness, further screen the upstream and downstream files of the enterprise. The specific steps are as follows:

[0044] When the credit in the current enterprise file label is excellent, retain all the enterprise files in the upstream or downstream files of the current enterprise;

[0045] When the credit in the current enterprise file label is average, retain the enterprises with the enterprise labels of average credit or poor credit in the upstream or downstream files of the current enterprise;

[0046] When the credit in the current enterprise file label is poor, retain the enterprises with the enterprise label of poor credit in the upstream or downstream files of the current enterprise.

[0047] Furthermore, the priority recommendation determination module calculates the enterprise purchasing power and productivity levels by combining the recent e-bill information of enterprises in the upstream and downstream files, and sorts the enterprises in the upstream and downstream files by combining the monthly order quantity of enterprises in the files. The specific steps are as follows:

[0048] Based on the amounts in the purchase invoices and sales invoices of enterprises in the basic raw material industry, manufacturing industry, and commercial service industry within the platform, determine the enterprise purchasing power and productivity. The specific steps are as follows:

[0049] For the basic raw material industry and the manufacturing industry, the sales invoice amounts of enterprises with consistent product type labels in the enterprise archives within the statistical platform are counted over a two-year period. The average annual sales amount is calculated, and based on the distribution of the amounts, the enterprise productivity levels are divided. The specific algorithm formula is as follows:

[0050] For enterprises under different product type labels, calculate the average annual sales amount :

[0051] ;

[0052] Based on the average annual sales amounts of enterprises with the same product type labels within the platform, obtain the maximum average annual sales amount and the minimum average annual sales amount , and divide the interval width :

[0053] ;

[0054] Among them, represents the number of segments. According to and , calculate the upper and lower limits of each productivity level. The algorithm formula is as follows:

[0055] ;

[0056] ;

[0057] Based on the average annual sales amounts of different enterprises, divide the enterprises into different intervals, and record the upper limit amount and the lower limit amount of the interval in the enterprise archives;

[0058] For the manufacturing industry and the commercial service industry, the purchase invoice amounts of enterprises with consistent product type labels in the enterprise archives within the statistical platform are counted over a two-year period. The average annual purchase amount is calculated, and based on the distribution of the amounts, the enterprise purchasing power levels are divided. The specific algorithm formula is as follows:

[0059] For enterprises under different product type labels, calculate the average annual purchase amount :

[0060] ;

[0061] Based on the average annual purchase amounts of enterprises with the same product type labels within the platform, obtain the maximum average annual purchase amount and the minimum average annual purchase amount , and divide the interval width :

[0062] ;

[0063] Among them, Represents the number of segments, according to and , calculate the upper and lower limits of each purchasing power level, and the algorithm formula is:

[0064] ;

[0065] ;

[0066] Based on the average annual purchase amount of different enterprises, the enterprises are divided into different ranges, and the upper and lower limits of the ranges are recorded in the enterprise files;

[0067] Based on the purchasing power and productivity level of enterprises, combined with the product information of enterprise purchase invoices and sales invoices, the upstream or downstream files of enterprises in the basic raw materials industry, manufacturing industry, and commercial service industry are screened twice, and the enterprises in the upstream and downstream files are ranked based on the number of monthly orders of the enterprises in the files.

[0068] Furthermore, based on the purchasing power and productivity level of the enterprise, combined with the product information of the enterprise's purchase invoice and sales invoice, the upstream files or downstream files of the basic raw material industry, manufacturing industry, and commercial service industry are screened twice, and the correlation score is calculated in combination with the enterprise's transaction frequency, and the upstream files or downstream files retained enterprises in the enterprise files are arranged in descending order of correlation score. The specific steps are:

[0069] For upstream archives of manufacturing and commercial service enterprises, extract the commodity names in the current enterprise purchase invoices and the commodity names in the enterprise sales invoices in the upstream archives, and establish a collection of enterprise purchase commodities And the collection of goods sold by each enterprise in the upstream archive ,in Remove duplicate product names from the current company's purchase invoices. Remove duplicate product names from sales invoices of various enterprises in the upstream files of the current enterprise. is the number of enterprises in the upstream archive, when If it is not an empty set, the relevant enterprises are retained in the upstream archive. If it is an empty set, the relevant enterprises will be removed from the upstream archives;

[0070] For the downstream archives of basic raw material industries and manufacturing enterprises, extract the commodity names in the current enterprise sales invoices and the commodity names in the enterprise purchase invoices in the downstream archives, and establish the enterprise sales commodity collection And the collection of goods purchased by each enterprise in the downstream archives ,in Remove duplicate product names from the current company's sales invoices. For the de-duplicated commodity names in the purchase invoices of each enterprise in the downstream archives of the current enterprise, For the number of enterprises in the downstream archives, when is not an empty set, relevant enterprises are retained in the downstream archives. When is an empty set, relevant enterprises are excluded from the downstream archives;

[0071] Respectively count the number of sales invoices within two years of different retained enterprises in the upstream archives and the number of purchase invoices within two years of different retained enterprises in the downstream archives, and calculate the average monthly order volume. The algorithm formula is as follows:

[0072] For the upstream archives: ;

[0073] For the downstream archives: ;

[0074] For the enterprises in the upstream archives, classify the different enterprises in the upstream archives according to the productivity level, and in each classification, sort them in descending order based on the average monthly order volume of different enterprises;

[0075] For the enterprises in the downstream archives, classify the different enterprises in the downstream archives according to the purchasing power level, and in each classification, sort them in descending order based on the average monthly order volume of different enterprises.

[0076] Compared with the prior art, the beneficial effects of the present invention are:

[0077] 1. In the present invention, based on the types of electronic bills of enterprises, for different types of enterprises, different data are extracted by selecting different types of electronic bills, and the enterprise archives within the platform are labeled and recorded to obtain an industrial map, assisting the enterprises within the platform to establish upstream and downstream associations with each other, and extracting the trading behaviors and business models of the enterprises;

[0078] 2. In the present invention, by obtaining the bill information of the enterprises within the platform, establishing upstream and downstream archives in combination with the types of different enterprises, and screening according to the credit level and commodity name, it is ensured that the upstream and downstream archives meet the actual needs of the current enterprise. At the same time, for the upstream archives of the enterprise, through the sorting of the productivity level and the average monthly order volume, it can assist the enterprise to preferentially select potential suppliers with high productivity and stable order volume. For the downstream archives of the enterprise, through the sorting of the purchasing power level and the average monthly order volume, it can assist the enterprise to discover high-quality customers;

[0079] 3. In the present invention, by automatically identifying and matching upstream and downstream enterprises for the enterprises settled in the platform according to the internal enterprise electronic bill data, manual intervention is reduced, the upstream and downstream relationships between enterprises are identified, and the supply chain network of enterprises is helped to be constructed. At the same time, enterprises can optimize their own supply chain structures according to the upstream and downstream relationships provided by the platform, select more suitable suppliers or customers, reduce costs, and enhance the practicability and functionality of the platform;

[0080] 4. In the present invention, by analyzing different enterprise types specifically, for basic raw material industries, manufacturing industries, financial service industries, commercial service industries, and other industries, their bill data is analyzed specifically to dig out information that better meets the actual needs of enterprises, and the upstream and downstream enterprise matching analysis is carried out in combination with the bill scale, purchased goods, and credit conditions of the enterprise itself, so as to objectively and systematically evaluate potential suppliers and distributors for the enterprises settled in the platform;

[0081] The entire industrial brain data analysis platform based on the industrial knowledge graph can realize enterprise classification, extraction of electronic bill data, and association of upstream and downstream enterprises, which is convenient for the enterprises settled in the platform to discover potential suppliers and distributors, and enhances the practicability and functionality of the platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0082] Figure 1 It is a block diagram of the industrial brain data analysis platform based on the industrial knowledge graph of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0083] Next, the technical solutions of the present invention will be described clearly and completely in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0084] Embodiment 1: As Figure 1 shown, the industrial brain data analysis platform based on the industrial knowledge graph includes an enterprise data collection module, an industrial graph construction module, an upstream and downstream association matching module, and a priority recommendation determination module;

[0085] The enterprise data collection module collects the data information of the enterprises settled in the platform, including enterprise names, enterprise business licenses, enterprise locations, enterprise invoices, bills of exchange, and check data, and establishes the files of the enterprises, records the data information of the enterprises settled in, and transmits the data into the industrial graph construction module;

[0086] The industrial map construction module classifies the bill data of enterprises based on the types of enterprises, extracts different data of different bills for different types of enterprises, and generates labels to construct an industrial map. The specific steps are as follows:

[0087] Based on the business licenses of the platform-registered enterprises collected by the enterprise data collection module, the enterprises are classified into different types, including basic raw material industries, manufacturing industries, financial service industries, commercial service industries, and other industries. Determine the collection scope of electronic bill information for different types of enterprises, extract the characters in the electronic bill through OCR technology, and based on keyword matching in NLP technology combined with the enterprise type, extract key fields respectively, where:

[0088] For basic raw material industries, extract the commodity name, amount, date, customer name, and bill status in the sales invoices in the enterprise's historical electronic bills;

[0089] For manufacturing industries, extract the commodity name, amount, date, customer name in the sales invoices in the enterprise's historical electronic bills, the commodity name, amount, date, customer name in the purchase invoices, and the bill status;

[0090] For financial service industries, extract the bill amount and check amount in the enterprise's historical electronic bills;

[0091] For commercial service industries, extract the commodity name, amount, date, customer name in the purchase invoices in the enterprise's historical electronic bills, and the bill status;

[0092] For other industries, no electronic bill data extraction is performed;

[0093] It should be noted that by using Tesseract OCR to convert bill images or PDF files into text, different types of keywords and regular expressions are defined according to the enterprise type, and the keywords and regular expressions are used to match and extract key data from the text extracted by OCR. Among them, the basic raw material industry mainly includes enterprises engaged in the extraction, processing and sales of raw materials, such as mining companies, agricultural companies, forestry companies, etc. The main daily business direction is to provide raw materials for downstream enterprises. The manufacturing industry includes enterprises that produce products by processing raw materials, such as machinery manufacturing, electronic equipment, clothing production, etc. The main daily business direction is to provide products for downstream enterprises and obtain raw materials from upstream enterprises. The financial service industry includes enterprises that provide financial services, such as banks, insurance companies, securities companies, etc. The main daily business direction is to provide checks, bills of exchange, etc. The commercial service industry includes enterprises that provide commercial services, such as retail, wholesale, logistics, etc. The main business direction is to obtain products from upstream enterprises for sale. Other industries such as software development, Internet services, education, medical care, etc., due to their business models being quite different from those of the manufacturing industry, basic raw material industry or financial service industry, usually involve service contracts, technology development contracts, software license agreements, etc., and involve less electronic bill data.

[0094] Combined with the bill data extracted from different types of enterprises in the platform, generate enterprise labels to establish an industrial map. The specific steps are as follows:

[0095] According to the known commodity names, create a keyword list for each category, and use the fuzzy matching method to match the data extracted from the electronic bills of different types of enterprises with the keyword list by using the Jaccard similarity. Among them:

[0096] For the basic raw material industry, divide the product type labels according to the commodity names in the sales invoices, including metal and ore categories, non-metal ore categories, chemical raw material categories, building material categories, agricultural raw material categories, energy raw material categories, rare and precious raw material categories, other basic raw material categories. Based on the commodity names in the enterprise sales invoices, record the product type labels in the enterprise archives of the basic raw material industry;

[0097] For the manufacturing industry, divide the raw material type labels according to the commodity names in the purchase invoices, including metal and ore categories, non-metal ore categories, chemical raw material categories, building material categories, agricultural raw material categories, energy raw material categories, rare and precious raw material categories, other basic raw material categories. At the same time, divide the product type labels according to the commodity names in the sales invoices, including machinery and equipment categories, electronic and electrical equipment categories, automotive and parts categories, chemical product categories, metal product categories, building materials and household categories, textile and clothing categories, food and packaging categories, medical device categories, other manufacturing categories. Based on the commodity names in the enterprise purchase invoices and sales invoices, record the raw material type labels and product type labels in the enterprise archives of the manufacturing industry;

[0098] For the commercial service industry, product type labels are classified according to the product names in the purchase invoices, including machinery and equipment, electronic and electrical equipment, automobiles and parts, chemical products, metal products, building materials and household items, textiles and clothing, food and packaging, medical devices, and other manufacturing products, and the product type labels are recorded in the enterprise files of commercial service enterprises.

[0099] It should be noted that the metal and ore categories include but are not limited to iron ore, copper ore, and aluminum ore; the non-metal ore categories include but are not limited to limestone and quartz sand; the chemical raw material categories include but are not limited to petrochemical raw materials, organic chemical raw materials, and inorganic chemical raw materials; the building material categories include but are not limited to cement and sand; the energy raw material categories include but are not limited to coal and petroleum; the rare and precious raw material categories include but are not limited to rare earth elements and diamonds; and the other basic raw material categories include but are not limited to water resources and salts.

[0100] The machinery and equipment category includes but is not limited to machine tools and compressors; the electronic and electrical equipment category includes but is not limited to refrigerators, washing machines, air conditioners, and televisions; the automobiles and parts category includes but is not limited to passenger cars, commercial vehicles, engines, and gearboxes; the chemical products category includes but is not limited to plastic products and rubber products; the metal products category includes but is not limited to steel pipes, steel plates, and steel bars; the building materials and household items category includes but is not limited to doors and windows, floors, and tiles; the textiles and clothing category includes but is not limited to cotton yarn, chemical fiber, and wool spinning; the food and packaging category includes but is not limited to canned foods, frozen foods, and beverages; the medical devices category includes but is not limited to CT machines, medical device accessories, and medical gauze; and the other manufacturing products category includes but is not limited to office supplies, sports equipment, toys, and handicrafts.

[0101] Embodiment 2: The upstream and downstream association matching module determines the enterprise credit rating by combining the enterprise's own electronic bill data information, conducts upstream and downstream association matching analysis on the enterprise, and generates the upstream files and downstream files of different enterprises. The specific steps are as follows:

[0102] For enterprises in the basic raw material industry, manufacturing industry, and commercial service industry, the enterprise credit rating is determined based on the bill status. The specific steps are as follows:

[0103] Based on the electronic bill data identified in the industrial map construction module, collect the within the year and to the bill data status of the current enterprise for the year of

[0104] When the enterprise has no overdue bill data status within the year, and from If there is no overdue in the annual bill data status, record the enterprise label as "excellent credit" in the enterprise file;

[0105] When the enterprise has no overdue in the annual bill data status within to and there is overdue in the annual bill data status, record the enterprise label as "medium credit" in the enterprise file;

[0106] When the enterprise has overdue in the annual bill data status within

[0107] It should be noted that usually takes the value of 1, that is, the credit rating of enterprises within the platform is classified according to the bill status of relevant enterprises within 1 year and the bill status within 2 - 5 years;

[0108] For enterprises in basic raw material industries, manufacturing industries, financial service industries, and commercial service industries, combined with enterprise electronic bill data information and credit ratings, upstream and downstream association matching analysis of enterprises within the platform is carried out based on enterprise types respectively, and upstream and downstream files of different enterprises are generated. The specific steps are as follows:

[0109] For the financial service industry, count the 2-year bill and check amount data of financial service industry enterprises within the platform, obtain the maximum bill amount, minimum bill amount, maximum check amount, and minimum check amount respectively, and calculate the maximum and minimum current enterprise transactions:

[0110] ;

[0111] ;

[0112] Obtain the enterprise customer range by combining the maximum and minimum current enterprise transactions :

[0113] ;

[0114] Count the 2-year purchase invoice amounts of basic raw material industries, manufacturing industries, and commercial service industries, obtain the maximum purchase amount and minimum purchase amount respectively, and obtain the enterprise transaction amount range , according to the of the current financial service industry enterprise, establish the downstream file of the enterprise, and incorporate the enterprise files into the downstream file of the current financial service industry enterprise;

[0115] It should be noted that The procurement amounts of relevant enterprises in the internal basic raw material industry, manufacturing industry, and commercial service industry of the representative platform are within the range of the business handling amounts of current financial service enterprises within two years, and they are potential downstream customers.

[0116] For the manufacturing industry, establish downstream files and upstream files in the enterprise file respectively. Based on the raw material type labels in the enterprise file, divide the enterprise files with the same product type labels as the raw material type labels into the upstream file of the current enterprise. Based on the product type labels in the enterprise file, divide the enterprise files with the same raw material type labels as the product type labels into the downstream file of the current enterprise;

[0117] For the basic raw material industry, establish a downstream file in the enterprise file. Based on the product type labels in the enterprise file, divide the enterprise files with the same raw material type labels as the product type labels into the downstream file of the current enterprise;

[0118] For the commercial service industry, establish an upstream file in the enterprise file. Based on the raw material type labels in the enterprise file, divide the enterprise files with the same product type labels as the raw material type labels into the upstream file of the current enterprise;

[0119] Based on the current enterprise creditworthiness, further screen the upstream and downstream files of the enterprise. The specific steps are as follows:

[0120] When the credit in the current enterprise file label is excellent, retain all enterprise files in the upstream or downstream file of the current enterprise;

[0121] When the credit in the current enterprise file label is average, retain the enterprise labels with average or poor credit in the upstream or downstream file of the current enterprise;

[0122] When the credit in the current enterprise file label is poor, retain the enterprise labels with poor credit in the upstream or downstream file of the current enterprise.

[0123] It should be noted that by screening the upstream or downstream files of the current enterprise based on the enterprise creditworthiness, the higher the creditworthiness of the enterprise, the stronger the selectivity for the upstream or downstream files, which helps to positively promote the enterprise to abide by the agreement and improve the enterprise creditworthiness.

[0124] Example 3: The priority recommendation determination module, combines the recent electronic bill information of enterprises in the upstream and downstream files, calculates the enterprise purchasing power and productivity levels, and combines the monthly order quantities of enterprises in the files to sort the enterprises in the upstream and downstream files. The specific steps are as follows:

[0125] Based on the amounts in the purchase invoices and sales invoices of enterprises in the internal basic raw material industry, manufacturing industry, and commercial service industry on the platform, determine the purchasing power and productivity of the enterprises. The specific steps are as follows:

[0126] For the basic raw material industry and manufacturing industry, count the sales invoice amounts of enterprises with the same product type label in the enterprise archives on the platform within 2 years, calculate the average annual sales amount, and based on the distribution of the amount sizes, divide the enterprise productivity levels. The specific algorithm formula is:

[0127] Calculate the average annual sales amount for enterprises under different product type labels :

[0128] ;

[0129] Based on the average annual sales amounts of enterprises with the same product type label on the platform, obtain the maximum value of the average annual sales amount and the minimum value of the average annual sales amount , and divide the interval width :

[0130] ;

[0131] Among them, represents the number of segments. According to and , calculate the upper and lower limits of each productivity level. The algorithm formula is:

[0132] ;

[0133] ;

[0134] Based on the average annual sales amounts of different enterprises, divide the enterprises into different intervals, and record the upper limit amount and lower limit amount of the interval in the enterprise archives;

[0135] For the manufacturing industry and commercial service industry, count the purchase invoice amounts of enterprises with the same product type label in the enterprise archives on the platform within 2 years, calculate the average annual purchase amount, and based on the distribution of the amount sizes, divide the enterprise purchasing power levels. The specific algorithm formula is:

[0136] Calculate the average annual purchase amount for enterprises under different product type labels :

[0137] ;

[0138] Based on the average annual purchase amounts of enterprises with the same product type label on the platform, obtain the maximum value of the average annual purchase amount and the minimum value of the average annual purchase amount , and divide the interval width :

[0139] ;

[0140] Among them, represents the number of segments. According to and , calculate the upper and lower limits of each purchasing power level, and its algorithm formula is:

[0141] ;

[0142] ;

[0143] Based on the average annual purchase amount of different enterprises, divide the enterprises into different intervals, and record the upper limit amount and lower limit amount of the interval in the enterprise's file;

[0144] It should be noted that usually takes the value of 1. The number of segments can refer to the standards in different industries and be set by consulting experts in relevant industries. By dividing the intervals of the productivity and purchasing power of different types of enterprises and marking the upper and lower limits, the upstream and downstream files of enterprises can be further optimized.

[0145] Based on the enterprise's purchasing power and productivity levels, combined with the product information in the enterprise's purchase invoices and sales invoices, conduct secondary screening on the upstream or downstream files of enterprises in the basic raw material industry, manufacturing industry, and commercial service industry. Combine the monthly order quantity of enterprises in the file to sort the enterprises in the upstream and downstream files. The specific steps are as follows:

[0146] For the upstream files of manufacturing and commercial service enterprises, extract the commodity names in the current enterprise's purchase invoices and the commodity names in the sales invoices of enterprises in the upstream file to establish the enterprise's purchased commodity set and the sold commodity sets of each enterprise in the upstream file , where is the distinct commodity names in the current enterprise's purchase invoice, is the distinct commodity names in the sales invoices of each enterprise in the current enterprise's upstream file, is the number of enterprises in the upstream file. When is not an empty set, retain the relevant enterprises in the upstream file. When is an empty set, eliminate the relevant enterprises in the upstream file;

[0147] For the downstream files of the basic raw material industry, manufacturing and commercial service enterprises, extract the commodity names in the current enterprise's sales invoices and the commodity names in the purchase invoices of enterprises in the downstream file to establish the enterprise's sold commodity set and the purchased commodity sets of each enterprise in the downstream file , where It is the distinct product names in the current enterprise's sales invoices. It is the distinct product names in the purchase invoices of each enterprise in the downstream files of the current enterprise. It is the number of enterprises in the downstream files. When it is not an empty set, relevant enterprises are retained in the downstream files. When it is an empty set, relevant enterprises are excluded from the downstream files.

[0148] It should be noted that by establishing sets for the products of the enterprise and the products in the upstream or downstream files, it can be ensured that there are business overlap points between the enterprises in the upstream or downstream files and the current enterprise.

[0149] Respectively count the number of sales invoices within two years of different retained enterprises in the upstream file and the number of purchase invoices within two years of different retained enterprises in the downstream file, and calculate the average monthly order quantity. The algorithm formula is:

[0150] For the upstream file: ;

[0151] For the downstream file: ;

[0152] For the enterprises in the upstream file, classify the different enterprises in the upstream file according to the productivity level, and in each classification, sort them in descending order based on the average monthly order quantity of different enterprises.

[0153] For the enterprises in the downstream file, classify the different enterprises in the downstream file according to the purchasing power level, and in each classification, sort them in descending order based on the average monthly order quantity of different enterprises.

[0154] It should be noted that for the upstream file of the enterprise, through the sorting of productivity level and average monthly order quantity, suppliers with high productivity and stable order volume can be preferentially selected to improve the reliability and efficiency of the supply chain. For the downstream file, through the sorting of purchasing power level and average monthly order quantity, the purchasing power and needs of customers can be better understood for precise marketing and customer classification.

[0155] In the embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation; the modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the method of this embodiment.

[0156] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An industrial brain data analysis platform based on an industrial knowledge graph, characterized in that: It includes enterprise data collection module, industry map construction module, upstream and downstream association matching module, and priority recommendation and judgment module; The enterprise data collection module collects data information of enterprises settled in the platform, including enterprise name, enterprise business license, enterprise location, enterprise invoice, bill of exchange, and check data, establishes enterprise files through MySQL, records data information of settled enterprises, and transmits data to the industry map construction module; The industrial map construction module classifies the bill data of enterprises based on the types of enterprises, extracts different data of different bills for different types of enterprises, generates labels and constructs industrial maps. The specific steps are as follows: Based on the business licenses of the enterprises settled in the platform collected by the enterprise data collection module, the enterprises are divided into different types, including basic raw material industry, manufacturing industry, financial services industry, commercial services industry, and other industries. The scope of electronic bill information collection for different types of enterprises is determined, and the characters in the electronic bills are extracted through OCR technology. Based on the keyword matching in NLP technology combined with the enterprise type, the key fields are extracted respectively, including: For the basic raw materials industry, extract the product name, amount, date, customer name, and bill status from the sales invoice in the company's historical electronic bills; For the manufacturing industry, extract the product name, amount, date, customer name in the sales invoice, the product name, amount, date, customer name, and bill status in the purchase invoice from the company's historical electronic bills; For the financial services industry, extract the amount of bills and checks from the company's historical electronic bills; For commercial services, extract the product name, amount, date, customer name, and bill status from the purchase invoice in the company's historical electronic bills; For other industries, electronic invoice data extraction is not performed; Combine the bill data extracted from different types of enterprises on the platform to generate enterprise tags and establish an industry map; The upstream and downstream association matching module combines the electronic bill data information of the enterprise itself to determine the enterprise credit rating, conducts upstream and downstream association matching analysis on the enterprise, and generates upstream files and downstream files of different enterprises. The specific steps are as follows: For enterprises in the basic raw materials industry, manufacturing industry and commercial service industry, the credit rating of the enterprise is determined based on the status of the bill of exchange. The specific steps are as follows: Based on the electronic bill data identified in the industry graph building module, the bill data status within n years and n+1 to n+4 years of the above-mentioned types of enterprises are collected to classify the credit rating of the current enterprises, where the bill data status includes overdue and normal; For enterprises in the basic raw material industry, manufacturing industry, financial services industry and commercial service industry, we combine the electronic bill data information and credit rating of the enterprises to conduct upstream and downstream correlation matching analysis of the enterprises on the platform based on the enterprise type, generate upstream and downstream files for different enterprises, and further screen the upstream and downstream files of the enterprises based on the current enterprise credit rating; The priority recommendation judgment module combines the recent e-bill information of enterprises in the upstream and downstream files, calculates the enterprise purchasing power and productivity levels, and combines the monthly order quantity of enterprises in the files to sort the enterprises in the upstream and downstream files, including the following steps; For basic raw material industries and manufacturing industries, count the sales invoice amounts of enterprises with the same product type labels in the enterprise files within the platform in the past two years, calculate the average annual sales amount, and based on the amount size distribution, divide the enterprise productivity levels; For manufacturing industries and commercial service industries, count the purchase invoice amounts of enterprises with the same product type labels in the enterprise files within the platform in the past two years, calculate the average annual purchase amount, and based on the amount size distribution, divide the enterprise purchasing power levels; Based on the enterprise purchasing power and productivity levels, combined with the product information of the enterprise purchase invoices and sales invoices, conduct secondary screening on the upstream or downstream files of enterprises in basic raw material industries, manufacturing industries, and commercial service industries, calculate the correlation score in combination with the enterprise transaction frequency, and arrange the enterprises retained in the upstream or downstream files in the enterprise files in descending order of the correlation score.

2. The industrial brain data analysis platform based on the industrial knowledge graph according to claim 1, wherein: Combining the bill data extracted from different types of enterprises within the platform to generate enterprise labels and establish an industrial map, the specific steps are as follows: According to the known commodity names, create a keyword list for each category, and through the fuzzy matching method, use the Jaccard similarity to match the data extracted from the e-bills of different types of enterprises with the keyword list, where: For basic raw material industries, divide the product type labels according to the commodity names in the sales invoices, including metal and ore categories, non-metal ore categories, chemical raw material categories, building material categories, agricultural raw material categories, energy raw material categories, rare and precious raw material categories, and other basic raw material categories. Based on the commodity names in the enterprise sales invoices, record the product type labels in the enterprise files of basic raw material industries; For manufacturing industries, divide the raw material type labels according to the commodity names in the purchase invoices, including metal and ore categories, non-metal ore categories, chemical raw material categories, building material categories, agricultural raw material categories, energy raw material categories, rare and precious raw material categories, and other basic raw material categories. At the same time, divide the product type labels according to the commodity names in the sales invoices, including mechanical equipment categories, electronic and electrical equipment categories, automotive and parts categories, chemical product categories, metal product categories, building materials and household categories, textile and clothing categories, food and packaging categories, medical device categories, and other manufacturing categories. Based on the commodity names in the enterprise purchase invoices and sales invoices, record the raw material type labels and product type labels in the enterprise files of manufacturing industries; For commercial service industries, divide the product type labels according to the commodity names in the purchase invoices, including mechanical equipment categories, electronic and electrical equipment categories, automotive and parts categories, chemical product categories, metal product categories, building materials and household categories, textile and clothing categories, food and packaging categories, medical device categories, and other manufacturing categories, and record the product type labels in the enterprise files of commercial service industries.

3. The industrial brain data analysis platform based on the industrial knowledge graph according to claim 1, characterized in that, Conduct credit rating division for the current enterprise, where the bill data status includes overdue and normal, and the specific steps are as follows: If the bill data status of an enterprise is not overdue within n years and the bill data status from year n + 1 to year n + 4 is not overdue, record the enterprise label as "excellent credit" in the enterprise file; If the bill data status of an enterprise is not overdue within n years and the bill data status from year n + 1 to year n + 4 is overdue, record the enterprise label as "average credit" in the enterprise file; If the bill data status of an enterprise is overdue within n years, record the enterprise label as "poor credit" in the enterprise file.

4. The industrial brain data analysis platform based on the industrial knowledge graph according to claim 1, characterized in that: For enterprises in the basic raw material industry, manufacturing industry, financial service industry, and commercial service industry, combining the enterprise's electronic bill data information, purchasing power, productivity, and credit rating, conduct upstream and downstream association matching analysis of enterprises within the platform based on the enterprise type, and generate upstream and downstream files for different enterprises. The specific steps are as follows: For the financial service industry, count the bill and check amount data of financial service industry enterprises within the platform in the past 2 years, respectively obtain the maximum bill amount, minimum bill amount, maximum check amount, and minimum check amount, and calculate the maximum and minimum current enterprise transactions: Transaction maximum = max{maximum bill amount, maximum check amount}; Transaction minimum = min{minimum bill amount, minimum check amount}; Obtain the enterprise customer range by combining the maximum and minimum transaction values of the current enterprise Statistically analyze the purchase invoice amounts of the basic raw material industry, manufacturing industry, and commercial service industry within two years, respectively obtain the maximum purchase amount and the minimum purchase amount, and obtain the enterprise transaction amount range δ = [minimum purchase amount, maximum purchase amount]. According to the current financial service industry enterprise's Establish the enterprise's downstream archives, and incorporate the enterprise archives into the downstream archives of the current financial service industry enterprise; For the manufacturing industry, establish downstream and upstream files in the enterprise file respectively. Based on the raw material type label in the enterprise file, divide the enterprise files with the same product type label as the raw material type label into the upstream file of the current enterprise. Based on the product type label in the enterprise file, divide the enterprise files with the same raw material type label as the product type label into the downstream file of the current enterprise; For the basic raw material industry, establish a downstream file in the enterprise file. Based on the product type label in the enterprise file, divide the enterprise files with the same raw material type label as the product type label into the downstream file of the current enterprise; For the commercial service industry, establish an upstream file in the enterprise file. Based on the raw material type label in the enterprise file, divide the enterprise files with the same product type label as the raw material type label into the upstream file of the current enterprise; Based on the current enterprise's creditworthiness, further screen the upstream and downstream files of the enterprise. The specific steps are as follows: When the credit in the current enterprise file label is excellent, retain all enterprise files in the upstream or downstream file of the current enterprise; When the credit in the current enterprise file label is average, retain the enterprise files with the label of average credit or poor credit in the upstream or downstream file of the current enterprise; When the credit in the current enterprise file label is poor, retain the enterprise files with the label of poor credit in the upstream or downstream file of the current enterprise.

5. The industrial brain data analysis platform based on the industrial knowledge graph according to claim 1, wherein: The priority recommendation determination module combines the recent electronic bill information of enterprises in the upstream and downstream files, calculates the enterprise's purchasing power and productivity levels, and combines the monthly order quantity of enterprises in the file to sort the enterprises in the upstream and downstream files. The specific steps are as follows: Determine the purchasing power and productivity of enterprises based on the amounts in the purchase invoices and sales invoices of enterprises in the internal basic raw material industry, manufacturing industry, and commercial service industry. The specific steps are as follows: For the basic raw material industry and manufacturing industry, classify the productivity levels of enterprises. The specific algorithm formula is: For enterprises under different product type labels, calculate the average annual sales amount K: Obtain the maximum annual average sales K based on the annual average sales of enterprises with the same product type labels within the platform max and the minimum annual average sales K min , and divide the interval width H: where D represents the number of segments, and based on H and K min , calculate the upper and lower limits of each productivity level, and its algorithm formula is: Lower limit of the i-th interval = K min +(i - 1)·H; Upper limit of the i-th interval = K min + i·H; Based on the average annual sales amounts of different enterprises, divide the enterprises into different intervals, and record the upper limit amount and lower limit amount of the intervals in the enterprise files; For the manufacturing industry and commercial service industry, classify the purchasing power levels of enterprises. The specific algorithm formula is: For enterprises under different product type labels, calculate the average annual purchase amount G: Based on the average annual purchase amount of enterprises with the same product type label within the platform, obtain the maximum value G of the average annual purchase amount max and the minimum value G of the average annual purchase amount min , and divide the interval width h: where d represents the number of segments, and based on h and G min , calculate the upper and lower limits of each purchasing power level. The algorithm formula is as follows: Lower limit of the I-th interval = G min +(I - 1)·h; Upper limit of the I-th interval = G min + I·h; Based on the average annual purchase amounts of different enterprises, divide the enterprises into different intervals, and record the upper limit amount and lower limit amount of the intervals in the enterprise files; Based on the purchasing power and productivity levels of enterprises, combined with the product information in the enterprise purchase invoices and sales invoices, conduct a secondary screening of the upstream files or downstream files of enterprises in the basic raw material industry, manufacturing industry, and commercial service industry. Combine the monthly order quantities of enterprises in the files to sort the enterprises in the upstream and downstream files.

6. The industrial brain data analysis platform based on the industrial knowledge graph according to claim 5, characterized in that: Arrange the enterprises retained in the upstream files or downstream files in the enterprise files in descending order of the correlation scores. The specific steps are as follows: For the upstream archives of manufacturing and commercial service enterprises, extract the commodity names in the current enterprise purchase invoice and the commodity names in the enterprise sales invoice in the upstream archives, and establish the enterprise purchase commodity set P = {p1, p2, ..., p n } and the collection of goods sold by each enterprise in the upstream archive E i ={e1,e2,…,e n }, where {p1,p2,…,p n } is the name of the product to be deduplicated in the current enterprise purchase invoice, {e1,e2,…,e n } is the deduplicated commodity name in the sales invoices of each enterprise in the upstream file of the current enterprise, i is the number of enterprises in the upstream file, when P∩E is not an empty set, the relevant enterprises are retained in the upstream file, when P∩E is an empty set, the relevant enterprises are removed from the upstream file; For the downstream files of basic raw material industries and manufacturing enterprises, extract the commodity names in the current enterprise's sales invoices and the commodity names in the enterprise procurement invoices in the downstream files, and establish the enterprise sales commodity set F = {f1, f2, …, f n} and the procurement commodity sets J i = {j1, j2, …, j n} of each enterprise in the downstream file, where {f1, f2, …, f n} is the distinct commodity names in the current enterprise's sales invoices, and {j1, j2, …, j n} is the distinct commodity names in the procurement invoices of each enterprise in the current enterprise's downstream file. i is the number of enterprises in the downstream file. When P ∩ E is not an empty set, retain the relevant enterprises in the downstream file; when P ∩ E is an empty set, eliminate the relevant enterprises in the downstream file; Respectively count the number of sales invoices of different retained enterprises in the upstream files within 2 years and the number of purchase invoices of different retained enterprises in the downstream files within 2 years, and calculate the average monthly order quantity. The algorithm formula is: For upstream files: For downstream files: For the enterprises in the upstream files, classify the different enterprises in the upstream files according to the productivity levels, and within each classification, arrange them in descending order based on the average monthly order quantities of different enterprises; For the enterprises in the downstream files, classify the different enterprises in the downstream files according to the purchasing power levels, and within each classification, arrange them in descending order based on the average monthly order quantities of different enterprises.

Images