A method for extracting standardized product units based on the commodity details data in invoices

Abstract

The present invention discloses a method for extracting standardized product units based on the details data of commodities in invoices, belonging to the field of big data processing. Specifically, it includes: First, define the industrial chain and determine the scope of the industrial chain; and initially establish a knowledge-based industrial chain structure. Then, determine the basic SPU for each link in the upstream, midstream, and downstream of the industrial chain, that is, the common name of the commodity link. Next, write the policy statements for the commodity / service link and determine the invoice code. In the superset internal system platform, use MySQL statements to search for the invoice details data related to the commodity / service link, extract the standardized product units of the invoice details data after removing the impurity invoice data, and finally, label the SPU tags for the invoice details data. The present invention has the advantages of fast extraction and batch acquisition, as well as the significance of improving the data analysis efficiency and increasing the data application fields.

Description

Technical Field

[0001] The present invention belongs to the field of data processing in business or finance, and relates to big data processing. Specifically, it is a method for extracting standardized product units based on the commodity detail data in invoices. Background Art

[0002] Invoices issued between enterprises serve as a type of third-party evidence for verifying the transaction relationship between enterprises, known as transaction data. In addition to invoice data, customs data, contract information, winning bid information, listed company announcements, and e-commerce platform data can also reflect the trade between enterprises and the traded commodities involved in the trade. Especially in the field of production, manufacturing, and circulation, invoice data issued between enterprises can better restore the real trade situation, with advantages such as wide coverage and complete commodity details.

[0003] Invoice data used to be mostly archived in paper form. With the implementation of the Golden Tax Project in recent years, the relevant data has just been digitized. The commodity detail data in invoices is a natural language description of the smallest stock keeping unit (SKU), including the brand, color, specifications, uses, models, and standardized product units (SPUs) of the commodities. Since this description is filled in manually during the operation of enterprises, the standardization degree is poor. At the same time, due to the extremely wide range of fields involved in enterprise operations and the extremely complex variety of commodities, there are strong professional barriers.

[0004] Invoice data is mainly applied in the commercial fields of the industrial chain and financial institutions. In the past, enterprises on the industrial chain were mounted through well-known enterprises and publicly available information of listed companies. The number of mounted enterprises at each link was in the tens, lacking big data support and the representation of small, medium, and micro enterprises. There is an urgent need for a technology that can accurately find the trading parties using invoice data, with the number of mounted enterprises in the tens of thousands, to mount more and more real industrial members for the industrial chain, covering large, medium, small, and micro enterprises of all types.

[0005] Since the digitization of invoice data is relatively late, the extraction of commodity detail data in invoices through big data technology has not been popularized. If commodity recognition is directly carried out through big model-related technologies, the deviation of the results is relatively large and cannot be used in application scenarios. Summary of the Invention

[0006] In view of the above problems, the present invention proposes a method for extracting standardized product units based on the commodity detail data in invoices. By adopting big data processing technical means, through data analysis, data sorting, and data extraction, the non-standardized invoice detail data is transformed into standardized product units (SPUs), which has the advantages of fast extraction and batch acquisition, as well as the significance of improving data analysis efficiency and increasing data application fields.

[0007] The method for extracting standardized product units based on commodity details data in the invoice has the following specific steps:

[0008] Step 1: Determine and build the industrial chain based on big data models, expert experience and machine learning;

[0009] An industrial chain refers to a structure with a chain-like relationship that includes upstream, midstream and downstream, with core commodities or services as the midstream, extending upward as the upstream, and extending downward as the downstream.

[0010] The upstream and downstream extensions of goods / services are determined through transactions between enterprises, thus forming the upstream and downstream system of the industrial chain, and the goods / services traded between enterprises are defined as SPU.

[0011] Step 2: Obtain synonyms for each product / service link in the upstream, midstream and downstream of the industrial chain through semantic recognition and knowledge learning. Each group of synonyms constitutes a group of SPUs.

[0012] Step 3: Determine the most simplified name of the goods / services in each group of SPUs as the basic SPU of the group;

[0013] The method to determine the base SPU is as follows:

[0014] (1) Use the results of the SPU strategy test to find the extended basic SPU, and group the extended basic SPU into multiple basic SPUs corresponding to the SPU;

[0015] (2) Using synonyms of goods / services as basic SPU:

[0016] Step 4: Write a policy statement based on the product details of the invoice;

[0017] The invoice includes: the name of the selling company, the name of the purchasing company, the invoice classification code and the product details;

[0018] The specific process is:

[0019] First, write basic policy statements based on product names or basic SPUs;

[0020] It is divided into basic SPU form and modifier form;

[0021] (1) When the basic SPU is not limited, the policy statement directly adopts the basic SPU form:

[0022] (2) When the SPU is in the form of "modifier" + "basic SPU", the policy statement adopts the modifier form;

[0023] Then, the commodity classification code is obtained through the SPU and the corresponding basic SPU, and the basic policy statement is limited according to the commodity classification code to which the SPU belongs.

[0024] Finally, for the goods on the industrial chain, the applicable industrial fields of the goods are locked through the penetration strategy, and the final strategy statements are compiled in a penetrating form:

[0025] 1) Compile strategy statements based on the name of the enterprise that purchases or sells the SPU;

[0026] 2) Compile strategy statements based on the goods purchased by the enterprise selling the SPU;

[0027] 3) Compile strategy statements based on the goods sold by the enterprise purchasing the SPU;

[0028] Step Five: For a single invoice, screen it through the strategy statements to obtain the SPU specifically representing the sales of the invoice, that is, extract the standardized product unit of the commodity details data in the invoice.

[0029] Similarly, for a large number of invoices, based on the strategy statements, screen out the SPU represented by each invoice, so that the invoices containing the SPU can be aggregated, and then information such as the enterprises that sell / purchase the SPU can be statistically analyzed.

[0030] The advantages of the present invention are as follows:

[0031] 1) A method for extracting standardized product units based on the commodity details data in invoices, which constructs a commercialized industrial chain structure and makes the understanding of the industrial chain more concrete and suitable for commercial applications;

[0032] 2) A method for extracting standardized product units based on the commodity details data in invoices, which shows the actual transaction parties of a certain commodity according to the invoice data and provides a more accurate list of enterprise members;

[0033] 3) A method for extracting standardized product units based on the commodity details data in invoices, which can quickly and batch complete the standardized processing of invoice data. Description of the Drawings

[0034] Figure 1 It is a flowchart of a method for extracting standardized product units based on the commodity details data in invoices of the present invention. Detailed Embodiments

[0035] In order to facilitate the understanding and implementation of the present invention by those of ordinary skill in the art, the present invention will be further described in detail below with reference to the drawings and embodiments. Obviously, the described embodiments are only 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 scope of protection of the present invention.

[0036] A method for extracting standardized product units based on the commodity details data in invoices of the present invention adopts big data processing technology. Through data analysis, data sorting, and data extraction, the non-standardized invoice details data is transformed into standardized product units (SPUs), which has the advantages of rapid extraction and batch acquisition, improves the data analysis efficiency, and expands the data application fields.

[0037] The method for extracting standardized product units based on the commodity details data in invoices is as Figure 1 shown, and the specific steps are as follows:

[0038] Step 1: Determine and build an industrial chain based on big data models, expert experience, and machine learning;

[0039] Different institutions and customer groups have different definitions of the industrial chain. Determine the scope of the industrial chain based on big data models and machine learning to define the scope for the subsequent structure building process.

[0040] Summarize the initial structure of the industrial chain using big data, expert experience, and knowledge learning. The industrial chain refers to a chain-like associated relationship structure that includes upstream, midstream, and downstream, with the core commodity or service as the midstream, extending upward as the upstream, and extending downward as the downstream.

[0041] The upstream and downstream extensions of commodities / services are determined through transactions between enterprises, thereby constituting the upstream and downstream systems of the industrial chain. Define the commodities / services traded between enterprises as SPUs.

[0042] According to different types of industrial chains, it diverges upward into raw materials, production equipment, components, or production processes, etc., and extends downward into midstream classification, scenario applications, and terminal links, etc. By constructing an industrial chain map based on the characteristics of different industries for the upstream, midstream, and downstream, the production process of the chemical industry, the composition of automotive parts, or the application of lithium batteries can be clearly seen.

[0043] For example: An enterprise selling complete vehicles purchases automotive engines, and the automotive engine is the upstream link of the complete vehicle;

[0044] Step 2: Through semantic recognition and knowledge learning, obtain synonyms for each commodity / service link in the upstream, midstream, and downstream of the industrial chain, and each group of synonyms forms a group of SPUs;

[0045] Through semantic recognition, obtain synonyms of commodities. For example, synonyms of water cups include cups. When issuing invoices, due to manual writing, there will be multiple synonyms. By obtaining synonyms, lock different names of the same commodity and obtain as many invoice-related details as possible. For example, chip, integrated circuit, and IC are synonyms and jointly form a group of SPUs;

[0046] Step 3: Determine the most simplified name of the product / service in each group of SPU as the basic SPU of this group;

[0047] The basic SPU is the common name of this product link. For example, a chip can also be called an integrated circuit, a microcircuit, or an IC. Therefore, "chip", "integrated circuit", "microcircuit", and "IC" can all be used as the basic SPU. For example, for a blue glass, "glass" is the basic SPU; for SPU - automotive engine, the basic SPU is "engine";

[0048] The method for determining the basic SPU is as follows:

[0049] (1) Use the results of the SPU strategy test to search for extended basic SPU, and classify the extended basic SPU into multiple basic SPU corresponding to this SPU;

[0050] For example: The basic SPU - engine can find the extended basic SPU as engine assembly or engine group, etc.;

[0051] (2) Use synonyms in the product / service link as the basic SPU:

[0052] Step 4: Write strategy statements based on the product details of the invoice;

[0053] The invoice includes: the name of the selling company, the name of the purchasing company, the invoice classification code, and the product details;

[0054] The strategy anchors the transaction relationship between enterprises at this product level. The strategy writing establishes a correlation between the product name and the invoice details through SQL statements. The strategy includes two situations: "no need to limit enterprises" and "need to limit enterprises". Among them, "no need to limit enterprises" includes three forms: basic SPU form, modifier form, noun form, and coding form, and "need to limit enterprises" includes five forms: private sea form, enterprise form, specific form, dealer form, and subquery form.

[0055] The specific process is as follows:

[0056] First, write the basic strategy statements based on the product name or the basic SPU;

[0057] It is divided into the basic SPU form and the modifier form;

[0058] (1) When not limiting the basic SPU, the strategy statement directly adopts the basic SPU form:

[0059] For example: The SPU is "engine", and the corresponding basic SPU is: engine, engine group, engine assembly, which are used as the strategy statements;

[0060] (2) When the SPU is in the form of "modifier" + "basic SPU", the strategy statement adopts the modifier form;

[0061] For example, if the SPU is an automotive engine, the corresponding basic SPU is engine, engine group, and engine assembly, and the policy statements are "automobile + engine" or "automobile + engine group" or "automobile + engine assembly".

[0062] Then, obtain the commodity classification code through the SPU and the corresponding basic SPU, and limit the basic policy statement according to the commodity classification code to which the SPU belongs.

[0063] The commodity classification code uniformly compiled by the tax authority according to the commodity type; for example: automotive wheels can correspond to both the commodity classification code for automotive parts and the commodity classification code for metal products;

[0064] Finally, for the commodities on the industrial chain, lock the industrial field applicable to the commodity through the penetration strategy, and implement the final policy statement written in the penetration form:

[0065] When the SPU is a general part or raw material, it is necessary to accurately locate the general part or raw material on a specific industrial chain through the penetration strategy;

[0066] 1) Write the policy statement based on the name of the enterprise that purchases or sells the SPU; for example: the SPU on the automotive industrial chain is an engine. To determine that the enterprise selling the engine meets the requirements of the automotive industrial chain, limit the name of the enterprise selling or purchasing the engine to contain "automobile";

[0067] 2) Write the policy statement based on the commodities purchased by the enterprise selling the SPU; for example: the SPU on the bio-based materials industrial chain is a plastic bag. To determine that the enterprise selling the plastic bag meets the requirements of the bio-based materials industrial chain, limit the enterprise selling the plastic bag to have purchased bio-based raw materials such as straw, cassava, and corn;

[0068] 3) Write the policy statement based on the commodities sold by the enterprise purchasing the SPU; for example: the SPU on the automotive industrial chain is an engine. To determine that the enterprise selling the engine meets the requirements of the automotive industrial chain, limit the enterprise purchasing the engine sold by the above enterprise to have sold automobiles;

[0069] Step Five: For a single invoice, screen it through the policy statement to obtain the specific SPU representing the sales of the invoice, that is, extract the standardized product unit of the commodity details in the invoice.

[0070] Similarly, for a large number of invoices, screen through the policy statement to obtain the SPU represented by each invoice, so that invoices containing the SPU can be aggregated, and then information such as the enterprises selling / purchasing the SPU can be statistically analyzed.

[0071] Example:

[0072] Step 1: Build the industrial chain;

[0073] Among them, (1) Composition of the upstream structure

[0074] 1. According to the first-level division of the upstream

[0075] (1) Division by composition:

[0076] If the name of the industrial chain (generally consistent with the name of the midstream link) is a large physical object, the upstream is generally split according to the components of the large physical object;

[0077] ① For example: the drone industrial chain (drone flight control system, drone transmission system, drone power system and other parts)

[0078] ② For example: the automobile industrial chain (automobile engine, automobile chassis, automobile body)

[0079] ③ For example: the photovoltaic power generation system industrial chain (photovoltaic modules, photovoltaic brackets, photovoltaic inverters, busbar boxes, photovoltaic batteries, photovoltaic cables, MPPT)

[0080] (2) Division by classification:

[0081] If the name of the industrial chain is a general term for a certain type of physical object, the upstream is generally divided according to the classification focus of the general term of the physical object;

[0082] ① For example: Hydrogen energy focuses on classification by production method (the first-level of the upstream of the hydrogen energy industrial chain is divided into hydrogen production from chemical by-products, hydrogen production from methanol reforming, hydrogen production from fossil fuels, and hydrogen production from electrolyzed water)

[0083] ② For example: Superhard materials focus on classification by upstream product types (the first-level of the upstream of the superhard material industrial chain is divided into natural diamond, synthetic diamond, cubic boron nitride, and carbon nitride)

[0084] ③ For example: Refractory materials focus on classification by production materials and equipment (the first-level of the upstream of the refractory material industrial chain is divided into refractory materials, auxiliary materials, and refractory material processing equipment)

[0085] (3) Division by process

[0086] If the name of the industrial chain is a product / service provided according to a certain production / service process, the upstream is generally divided by process;

[0087] ① For example: the IC industrial chain (IC design, IC manufacturing, IC packaging and testing);

[0088] ② The green and low-carbon steel industrial chain (ironmaking, steelmaking, continuous casting, rolling)

[0089] (4)Division by combination of software and hardware

[0090] If the name of the industrial chain is a relatively broad concept, it is generally considered from two aspects: soft (technology / process / software) and hard (composition). Judge the importance of the "soft" aspect. If the soft degree is relatively low, it is classified according to the composition; if the soft degree is relatively high, then the combination of soft (technology / process / software) and hard (composition) is used as the first-level classification.

[0091] ① The high-speed railway industrial chain (rails, railway signals, train sets) and the UHV industrial chain (power control terminal, power transmission) are classified according to the composition.

[0092] ② The VR industrial chain (key technologies, hardware) is classified according to technology and composition.

[0093] ③ The 5G industrial chain (network planning and design, main equipment, base station system, transmission equipment, tower construction, supporting equipment, network optimization) is classified according to process and composition.

[0094] ④ The commercial aerospace industrial chain (satellite manufacturing, satellite launch, ground station, satellite application ground equipment, satellite management application software) is classified according to process, software and composition.

[0095] (2) Composition of the downstream structure

[0096] 1. First-level classification

[0097] The downstream structure is generally classified according to downstream products, industries, classifications (uses, forms, channels, application scenarios, terminals), etc.

[0098] (1) Classification by product

[0099] When the downstream products can be exhausted, it is classified by product: for example, the pure electric vehicle industrial chain (downstream is divided into pure electric transport vehicles, pure electric passenger vehicles, pure electric road sweepers, pure electric garbage trucks, pure electric engineering vehicles, pure electric buses, pure electric trucks, pure electric service vehicles); such as the copper-based material industrial chain (downstream is divided into copper pipes, copper rods, copper wires, copper plates, copper strips, etc.); such as the semiconductor industrial chain (downstream is divided into integrated circuits, discrete devices, optoelectronic devices, sensors, photovoltaic cells).

[0100] (2) Classification by industry

[0101] When the downstream products cannot be exhausted, but the application industries can be exhausted, it is classified by industry: for example, the sugar substitute industrial chain (downstream is divided into food and beverage, medicine, cosmetics); such as the soda ash industrial chain (downstream is divided into glass industry, chemical industry, metallurgical industry, ceramic industry, food industry, pharmaceutical industry, printing and dyeing industry, analysis and monitoring).

[0102] (3) Classification by classification

[0103] When the industries cannot be exhausted, it is classified according to classifications (uses, forms, channels, application scenarios, terminals, etc.).

[0104] For example, the drone industry chain is divided by use (downstream is divided into civilian drones and military drones); for example, the photovoltaic power generation system industry chain is divided by form (downstream is divided into distributed photovoltaic power stations and centralized photovoltaic power stations); for example, the traditional Chinese medicine decoction pieces industry chain is divided by channel (downstream is divided into hospitals, pharmacies, and clinics); for example, the artificial intelligence industry chain is divided by application scenario (downstream is unmanned driving, new retail, Industry 4.0, intelligent assistants, intelligent security, intelligent finance, and intelligent education); for example, the industrial Internet industry chain is divided by terminal equipment (downstream is controlled construction machinery and controlled unmanned fixed equipment); for example, the 5G industry chain is divided by application and terminal (downstream is divided into applications (Internet, enhanced mobile broadband, others), terminals (mobile phones, cars, smart home appliances, smart wearable devices, electronic industrial equipment)).

[0105] Step 2: Strategy writing for the commodity / service link

[0106] Self-develop and build a database. The underlying data is invoice detail data. Retrieve, classify, count, and test invoice details in the database through MySQL statements.

[0107] MySQL statements can formulate strategies through product names, basic SPU, invoice codes, trading parties, etc. The strategic goal is to anchor the establishment of relevance between enterprises at the product layer. Through the strategy, find the production, sales, and purchasing enterprises in the commodity link. For example, if the commodity link is "refrigerator", through SQL statements, data containing refrigerators in the invoice can be found. Although the data all contains refrigerators, there are impurity data in the data, and more statements are needed to find more accurate trading parties.

[0108] 1. Determine the basic SPU;

[0109] The basic SPU is the simplest name of the commodity / service. For example, if the commodity name is double-door refrigerator, the basic SPU is refrigerator.

[0110] The method for determining the basic SPU is as follows:

[0111] (1)Use the results of the strategy test to view the basic SPU. For example, for the above-mentioned "automobile window glass lifter", find the automobile window glass lifter through the statement. After querying, it is found that there are impurities such as the window glass lifter motor, the right rear window regulator of the window glass lifter, the window glass lifter switch, and the window glass lifter control button. Through all the data related to the automobile window glass lifter, perform conditional screening. For example, sort by transaction amount to find several names with larger transaction amounts, such as the front window glass lifter and the total assembly of the electric window glass lifter. Summarize the commonalities of the names. Therefore, the common names can be used as the basic SPU. The basic SPU of the automobile window glass lifter can be "lifter" and "total assembly of lifter";

[0112] (2)Adopt the general names and their aliases or synonyms in the commodity / service link as the basic SPU: For example, a chip can also be called an integrated circuit, a microcircuit, or an IC. Therefore, "chip", "integrated circuit", "microcircuit", and "IC" can all be used as the basic SPU;

[0113] (3)Identify well-known enterprises and listed companies through the knowledge learned in this commodity / service link, and query the transaction records of the company through statements to obtain the names of the corresponding commodities / services to determine the basic SPU.

[0114] Due to the irregularity of the invoice details, determining the basic SPU first is to have more combination forms to match the invoice details when writing strategy statements.

[0115] 2. Try to write strategy statements according to the names in the commodity / service link

[0116] (1)When the commodity / service does not need to limit a certain type of enterprise, the statements adopt the basic SPU form, the modifier form, and the noun form:

[0117] 1)When there are multiple common sayings in the commodity / service link, the basic SPU form can be adopted. There are the following three situations:

[0118] A. If the name of the commodity / service link is a fixed term, such as "chip", then directly use the SPU strategy statement. However, since a chip is also called an integrated circuit, a microcircuit, or an IC, "chip", "integrated circuit", "microcircuit", and "IC" can all be used as the basic SPU;

[0119] B. If the name of the commodity / service link is relatively long, it can generally be split into several parts. For example, for "automobile window glass lifter", it can be split into multiple words "automobile", "window", "glass", and "lifter". However, "window" already covers the meanings of automobile and glass, and "window" is a modifier of "lifter". Therefore, the strategy statement is written as a combination of the modifier and the basic SPU;

[0120] C. For some product / service segments, corresponding products cannot be found using the above method based on the name, or the search results are very limited. In such cases, try to write using other alternative expressions. For example, for "virtual manufacturing verification", since the test data is scarce, first use statements to test and determine that this product belongs to the category of virtual manufacturing verification, with the basic SPU being "system" and "software". Write the statements in the form of nouns and basic SPU;

[0121] 2) When the product / service segment has a fixed name and few impurities, it is not necessary to write the statements in the form of basic SPU, but can be written in the form of nouns: for example, for consulting services, the statements can be directly written as nouns.

[0122] 3) When the product / service segment consists of a general "modifier" + "basic SPU", and the result range of the statements written in the form of basic SPU is too large and most do not meet the requirements, but the results of using "modifier" statements are more accurate, it is not necessary to write the statements in the form of basic SPU, but can be written in the form of modifiers:

[0123] For example, no corresponding results can be found for sonar receivers using the basic SPU receiver. However, through statement testing, it is found that the results of "sonar" are more satisfactory.

[0124] 3. Usage of invoice coding statements

[0125] The invoice code is a string of 19 digits. The first digit is a number from 1 to 6, and each number represents a major category, as shown in Table 1: Classification of Invoice Coding Major Categories. The following every 2 digits form a group, representing the classification of minor categories, as shown in Table 2: Classification of Invoice Coding Minor Categories (Partial).

[0126] Table 1

[0127]

[0128] Table 2

[0129]

[0130] 1) The distinction of product categories can be made by adding screening conditions for invoice codes. For example, distinguish the trading parties of household appliances refrigerators and toy refrigerators.

[0131] For example, find "toy refrigerators" through the invoice code, add a precise classification with the invoice code being "toy" in the SQL statement, and combine it with the product name "refrigerator" to obtain the result;

[0132] 2) Through the invoice code, products can also be accurately locked. After distinguishing by the invoice code of the major category, conduct precise product classification within the same major category. For example, for refrigerators and refrigerator accessories, distinguish them through different product codes;

[0133] 3) For some commodity / service segments that are major categories and have exclusive invoice codes, instead of using commodity names, the invoice code form is adopted to form the policy statement.

[0134] 4. Write statements in a penetrating form

[0135] For general commodities such as engines, there are automotive engines and aviation engines. Taking the engine as the basic SPU, in addition to writing policies by combining the basic SPU with adjectives such as automotive or aviation, it can also be judged by the buyer or seller on the invoice. For example, when looking for automotive engines, it can be distinguished that the purchased engine is an automotive engine by the seller or buyer being an automotive-related enterprise.

[0136] 5. Utilize the logical relationships in the statements

[0137] In the policy statement, the relationship between commodities and invoices is expressed through combinatorial logics such as "yes", "no", "or", or through logical statements such as fuzzy queries and multi-condition queries for matching.

[0138] Among the intricate invoices, relevant invoices can be quickly found through policy statements. Through the basic SPU, invoice codes, penetration of both trading parties, and logical relationships, impurities are removed to accurately match the standardized product unit.

[0139] Step 3: Label the SPU label for the invoice details data

[0140] Each invoice obtains the SPU label through the process of determining the standardized product unit.

[0141] Obtain the distribution of invoice codes through the commodity name and the relevant basic SPU. For example, when determining the invoice code through "automobile", there will be two major categories, one is transportation equipment and the other is toys. At this time, when applied in the automotive industry chain, the transportation equipment category is selected.

[0142] In the supersetn internal system platform, use MySQL statements to search for invoice details data related to the commodity / service segment. Among the intricate invoices, relevant invoices can be quickly found through policy statements. Although the invoices found at this time are relevant, there are also impurity invoices.

[0143] Preprocess and remove the impurity invoice data through the basic SPU, invoice codes, both trading parties, and statement logical relationships. Since the invoice details are non-standardized statements, invoices with stronger relevance can be screened through logical operations such as inclusion or exclusion, and impurities in the invoice details can be removed at the same time.

[0144] Determine the scope of the standardized product unit through dependency analysis and through the basic SPU. Due to the irregularity of invoice details, find relevant invoices through elements such as the basic SPU and invoice code, continue to determine the relevance of the invoices by removing impurities, and then form the invoice standardized product unit through dependency analysis.

[0145] Each invoice obtains an SPU label through the process of determining the standardized product unit. Subsequent industrial chain construction, standards for the main products of the enterprise, etc. can be quickly obtained through the SPU label.

[0146] Step 4: Mount the main products through data dimensions such as transaction data

[0147] Summarize the invoice details of the enterprise, and determine the basic SPU classification of the enterprise through the invoice details. For example, the enterprise's invoicing includes refrigerators, accommodation, catering, etc. Among them, judge the main products of the enterprise through the invoice transaction data. For example, if the invoice transaction amount of the refrigerator exceeds a certain threshold, it can be judged that the refrigerator is the main product of the enterprise (the setting of the threshold is based on the industry where the enterprise is located, the region where the enterprise is located, etc.).

[0148] Extracting the standardized product unit based on invoice details can quickly build the multi-layer structure of the industrial chain, specifically form the products in each link, and accurately mount the enterprises in each product link. At present, Daokou Jinke Company has built 163 industrial chains based on extracting the standardized product unit from invoice details, covering multiple industries such as new generation information technology, metallurgical industry, chemical industry, machinery and equipment industry, biological industry, high-end equipment manufacturing industry, service industry, energy conservation and environmental protection industry, food processing industry, textile and garment industry, real estate construction industry, new materials industry, and agriculture, forestry, animal husbandry and fishery; a total of more than 6,500 kinds of products have been built, and the number of enterprises exceeds 20 million;

[0149] For example, Daokou Jinke Company has formed a total of 7 layers, 160 product links, and 718,636 enterprises in the upstream, middle and downstream of the automobile industrial chain based on extracting the standardized product unit from invoice details. Compared with the construction of other automobile industrial chains, there are 22 product links and more than 120 industrial chain members involved. The number of product links and the number of enterprises are far ahead.

Claims

1. A method for extracting standardized product units based on commodity details data in an invoice, characterized in that: The specific steps are as follows: Step 1: Determine and build the industrial chain based on big data models, expert experience and machine learning; The industrial chain refers to a structure that includes upstream, midstream and downstream, with core products or services as the midstream, upstream extending upwards and downstream extending downwards in a chain-like relationship form; The upstream and downstream extension of goods / services is determined through transactions between enterprises, thus forming the upstream and downstream system of the industrial chain, and the goods / services traded between enterprises are defined as SPU; Step 2: Obtain synonyms for each product / service link in the upstream, midstream and downstream of the industrial chain through semantic recognition and knowledge learning. Each group of synonyms constitutes a group of SPUs. Step 3: Determine the most simplified name of the goods / services in each group of SPUs as the basic SPU of the group; Step 4: Write a policy statement based on the product details of the invoice; The specific process is: First, write basic policy statements based on product names or basic SPUs; It is divided into basic SPU form and modifier form; (1) When the basic SPU is not limited, the policy statement directly adopts the basic SPU form: (2) When the SPU is in the form of "modifier" + "basic SPU", the policy statement adopts the modifier form; Then, the commodity classification code is obtained through the SPU and the corresponding basic SPU, and the basic policy statement is limited according to the commodity classification code to which the SPU belongs; Finally, for the products in the industrial chain, the penetration strategy is used to lock the industrial field to which the product is applicable, and the final strategy statement is written in a penetration form: Step 5: For a single invoice, filter it through the policy statement to obtain the SPU that specifically represents the sales of the invoice, that is, extract the standardized product unit of the commodity detail data in the invoice.

2. The method according to claim 1, characterized in that In step 3, the method for determining the basic SPU is as follows: (1) Use the results of the SPU strategy test to find the extended basic SPU, and group the extended basic SPU into multiple basic SPUs corresponding to the SPU; (2) Use synonyms of goods / services as basic SPU.

3. The method according to claim 1, characterized in that In step 4, the final policy statement includes: (1) Write a policy statement based on the name of the business that buys or sells the SPU; (2) Write policy statements based on the products purchased by the SPU enterprise; (3) Write policy statements based on the products sold by the purchased SPU companies.

4. The method according to claim 1, characterized in that: In step 5, for the massive invoices, the SPU represented by each invoice is obtained by screening based on the policy statement, so as to aggregate the invoices containing the SPU, and then count the enterprise information that sells / purchases the SPU.

Images