Industrial chain identification method and device, terminal equipment and storage medium

Abstract

The invention is suitable for the technical field of data analysis, and provides an industrial chain identification method and device, terminal equipment and a storage medium, and the method comprises the steps: firstly determining an enterprise operation keyword corresponding to each to-be-identified enterprise according to the operation scope fields of a plurality of to-be-identified enterprises; then, performing similarity calculation on the enterprise operation keyword of each to-be-identified enterprise and the industrial feature word and sentence of the corresponding industrial chain position to obtain the industrial chain position similarity of each to-be-identified enterprise, and then according to the industrial chain position of each to-be-identified enterprise and the industrial chain position similarity, identifying the to-be-identified enterprise according to the industrial chain position of each to-be-identified enterprise and the industrial chain position similarity. And determining at least two target enterprises from the plurality of to-be-identified enterprises, and finally determining upstream and downstream relationships among the target enterprises. Therefore, whether each to-be-identified enterprise is the enterprise of the corresponding industrial chain position is judged by determining the industrial chain position similarity, so that the target enterprises are accurately screened out, the upstream and downstream relationship between the target enterprises is determined, and a more accurate industrial chain is obtained.

Description

Technical Field

[0001] The present application belongs to the field of data analysis technology, and in particular, relates to an industrial chain identification method, device, terminal equipment and storage medium. Background Art

[0002] Industrial chain identification can help all parties understand key links such as the supply chain, distribution chain, value chain, etc. in a specific industry, so as to make more informed decisions and help with product sales and promotion.

[0003] In related technologies, due to the influence of industry characteristics, market changes, corporate strategies, etc., the industrial chain division of some industries is relatively vague. Sometimes, some companies that do not belong to the industrial chain are mistakenly classified into the industrial chain, resulting in low accuracy of the ultimately identified industrial chain. Summary of the invention

[0004] The embodiments of the present application provide an industrial chain identification method, apparatus, terminal device, and storage medium, which can solve the problem of low accuracy of traditional industrial chain identification solutions.

[0005] A first aspect of an embodiment of the present application provides an industrial chain identification method, including:

[0006] Based on the business scope fields of multiple enterprises to be identified, the business operation keywords corresponding to each enterprise to be identified are determined, wherein the multiple enterprises to be identified are all enterprises in the target industry; the similarity between the business operation keywords of each enterprise to be identified and the industrial characteristic words and phrases of the corresponding industrial chain position is calculated to obtain the industrial chain position similarity of each enterprise to be identified, wherein the industrial chain position is determined based on the historical waybill data of the enterprise to be identified; based on the industrial chain position of each enterprise to be identified and the industrial chain position similarity, at least two target enterprises are determined from the multiple enterprises to be identified; and the upstream and downstream relationships between the target enterprises are determined.

[0007] In a possible implementation of the first aspect, before determining the business operation keyword corresponding to each to-be-identified enterprise according to the business scope fields of the multiple to-be-identified enterprises, the method further includes:

[0008] Obtaining historical waybill data of multiple initial enterprises, wherein the historical waybill data includes delivery and collection information;

[0009] Based on the mailing and receiving information and target industries corresponding to each initial enterprise, multiple initial enterprises are screened to determine multiple enterprises to be identified.

[0010] Optionally, in another possible implementation of the first aspect, determining the business operation keywords corresponding to each to-be-identified enterprise according to the business scope fields of the multiple to-be-identified enterprises includes:

[0011] Perform data cleaning on the business scope field of each enterprise to be identified to determine the business scope cleaning text;

[0012] Keyword extraction is performed on the cleaned text of the business scope of the enterprise to be identified to determine the business operation keywords corresponding to the enterprise to be identified.

[0013] Optionally, in yet another possible implementation of the first aspect, the keyword extraction is performed on the business scope cleansing text of the enterprise to be identified to determine the business keywords corresponding to the enterprise to be identified, including:

[0014] Perform word segmentation on the business scope cleansing text of the enterprise to be identified to obtain multiple short texts of the business scope;

[0015] Perform word frequency statistics on multiple business scope short texts to build a bag-of-words model;

[0016] The preset latent Dirichlet distribution topic model is called to process the bag-of-words model to determine the weight corresponding to each short text of the business scope;

[0017] Based on the weight corresponding to each business scope short text, the business operation keywords corresponding to the enterprise to be identified are determined.

[0018] Optionally, in another possible implementation of the first aspect, the similarity between the business operation keywords of each enterprise to be identified and the industry characteristic words and phrases at the corresponding industrial chain position is calculated to obtain the industrial chain position similarity of each enterprise to be identified, including:

[0019] The preset text similarity matching model is called to calculate the similarity between the business operation keywords of each enterprise to be identified and the industrial characteristic words and phrases of the corresponding industrial chain position, so as to obtain the industrial chain position similarity of each enterprise to be identified.

[0020] Optionally, in another possible implementation of the first aspect, the above-mentioned industrial chain position includes upstream, midstream and downstream, and the above-mentioned similarity calculation of the business operation keywords of each enterprise to be identified and the industrial characteristic words and phrases of the corresponding industrial chain position to obtain the industrial chain position similarity of each enterprise to be identified includes:

[0021] For the enterprises to be identified whose industrial chain position is upstream, the similarity between the enterprise business keywords and the corresponding industrial characteristic words and phrases of the upstream is calculated to obtain the industrial chain position similarity of each upstream enterprise to be identified, as well as,

[0022] For the enterprises to be identified in the middle of the industrial chain, the similarity between the enterprise operation keywords and the industrial characteristic words and phrases corresponding to the middle of the industrial chain is calculated to obtain the industrial chain position similarity of each enterprise to be identified in the middle of the industrial chain, as well as,

[0023] For the enterprises to be identified that are in the downstream position of the industrial chain, the similarity between the enterprise business keywords and the corresponding downstream industrial characteristic words and phrases is calculated to obtain the industrial chain position similarity of each downstream enterprise to be identified.

[0024] Optionally, in yet another possible implementation of the first aspect, after determining the upstream and downstream relationships between the target enterprises, the method further includes:

[0025] Based on the historical waybill data, corporate information and bidding information of each target enterprise, the importance of the enterprises to be identified in each industrial chain position is ranked respectively.

[0026] A second aspect of an embodiment of the present application provides an industrial chain identification device, including:

[0027] A keyword determination module, used to determine the business keywords corresponding to each enterprise to be identified based on the business scope fields of multiple enterprises to be identified, wherein the multiple enterprises to be identified are all enterprises in the target industry;

[0028] A similarity determination module is used to calculate the similarity between the business operation keywords of each enterprise to be identified and the industry characteristic words and phrases of the corresponding industrial chain position to obtain the similarity of the industrial chain position of each enterprise to be identified, wherein the industrial chain position is determined based on the historical waybill data of the enterprise to be identified;

[0029] An enterprise determination module, used to determine at least two target enterprises from multiple enterprises to be identified according to the industrial chain position and industrial chain position similarity of each enterprise to be identified;

[0030] The industrial chain identification module is used to determine the upstream and downstream relationships between target enterprises.

[0031] A third aspect of an embodiment of the present application provides a terminal device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the industrial chain identification method of the first aspect when executing the computer program.

[0032] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the industrial chain identification method of the first aspect is implemented.

[0033] A fifth aspect of the embodiments of the present application provides a computer program product. When the computer program product runs on a terminal device, the terminal device executes the industrial chain identification method of the first aspect.

[0034] Compared with the prior art, the embodiments of the present application have the following beneficial effects: the embodiments of the present application provide an industrial chain identification method, device, terminal device and storage medium, wherein the method first determines the business operation keywords corresponding to each enterprise to be identified based on the business scope fields of multiple enterprises to be identified, and then calculates the similarity between the business operation keywords of each enterprise to be identified and the industrial characteristic words and phrases of the corresponding industrial chain position to obtain the industrial chain position similarity of each enterprise to be identified, and then determines at least two target enterprises from multiple enterprises to be identified based on the industrial chain position and industrial chain position similarity of each enterprise to be identified, and finally determines the upstream and downstream relationship between each target enterprise. Thus, by determining the industrial chain position similarity, it is judged whether each enterprise to be identified is an enterprise at the corresponding industrial chain position, thereby accurately screening out the target enterprises, and then determining the upstream and downstream relationship between each target enterprise, and obtaining a more accurate industrial chain. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0036] Figure 1 This is a flow chart of an industrial chain identification method provided in Example 1 of the present application;

[0037] Figure 2 It is a schematic diagram of a process for calculating the similarity of a to-be-identified enterprise provided in an embodiment of the present application;

[0038] Figure 3 This is a flow chart of an industrial chain identification method provided in Example 2 of the present application;

[0039] Figure 4 This is a schematic diagram of the structure of an industrial chain identification device provided in Example 3 of the present application;

[0040] Figure 5 It is a structural diagram of a terminal device provided in Example 4 of the present application. DETAILED DESCRIPTION

[0041] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0042] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0043] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0044] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0045] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0046] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0047] It should be understood that the size of the serial numbers of the steps in this embodiment does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application.

[0048] In related technologies, due to the influence of industry characteristics, market changes, corporate strategies, etc., the industrial chain division of some industries is relatively vague. Sometimes, some companies that do not belong to the industrial chain are mistakenly classified into the industrial chain, resulting in low accuracy of the ultimately identified industrial chain.

[0049] In view of this, the embodiments of the present application provide an industrial chain identification method, device, terminal equipment and storage medium. By determining the similarity of the industrial chain position, it is judged whether each enterprise to be identified is an enterprise at the corresponding industrial chain position, so as to accurately screen out the target enterprises, and then determine the upstream and downstream relationships between each target enterprise to obtain a more accurate industrial chain.

[0050] The following is an example of the application scenario of the industrial chain identification method provided in the embodiment of the present application. The present application can be applied to large-scale and complex industrial chain identification such as the home appliance industry. The general process is as follows: first, the industrial chain position (upstream, midstream and downstream) of the home appliance industry is defined, wherein the upstream is mainly raw material manufacturers and parts manufacturing suppliers; the midstream is mainly home appliance manufacturing; the downstream is mainly focused on home appliance sales, mainly e-commerce platforms, supermarket retail, and brand self-operation. Then, by determining the similarity between multiple home appliance companies and the corresponding industrial chain positions, target companies that meet the industrial chain of the home appliance industry are selected from multiple home appliance companies. Finally, the target companies finally selected are used to obtain an accurate industrial chain of the home appliance industry.

[0051] In order to illustrate the technical solution of the present application, specific embodiments are provided below.

[0052] Reference Figure 1 , shows a flow chart of an industrial chain identification method provided in Example 1 of the present application. Figure 1 As shown, the industrial chain identification method may include the following steps:

[0053] Step 101, determining the business keywords corresponding to each enterprise to be identified based on the business scope fields of multiple enterprises to be identified.

[0054] Among them, the multiple enterprises to be identified are all enterprises in the target industry. It should be understood that the target industry can be the home appliance industry, the automobile industry, the electronics industry, the pharmaceutical and biological industry, etc., and the embodiment of the present application does not limit this.

[0055] It should be noted that the business scope field of the enterprise to be identified can be obtained by querying through the following channels. Among them, since the enterprise needs to submit relevant business scope information when conducting industrial and commercial registration, the business registration information of the enterprise can be queried through the industrial and commercial administrative department, the enterprise credit information disclosure system, etc., so as to obtain the business scope field. In addition, many companies publicly list their business scope and business areas on their official websites, so the business scope field can also be obtained by visiting the company's official website. In addition to the above methods, the business scope field of the enterprise to be identified can also be obtained through commercial databases, such as Qichacha, Tianyancha, etc. These commercial databases usually provide a large amount of basic information and business scope information of enterprises, which can be searched and queried through these commercial databases to obtain the business scope field of the relevant enterprise.

[0056] In the embodiment of the present application, the latent Dirichlet Allocation (LDA) topic model, TF-IDF (Term Frequency-Inverse Document Frequency) algorithm, TextRank (text sorting) algorithm, machine learning algorithm, etc. can be used to extract keywords from the business scope field of each enterprise to be identified, so as to obtain the business keywords corresponding to each enterprise to be identified. Among them, the LDA topic model is a technology for topic modeling of text. The LDA topic model can identify the topic in the text and extract keywords related to the topic; the TF-IDF algorithm determines the importance of keywords by calculating the frequency (TF) of a word in the text and the inverse document frequency (IDF) in the entire corpus. The higher the TF-IDF score of a word, the more important the word is in the text, thereby realizing keyword extraction; the TextRank algorithm is a sorting algorithm based on a graph. The TextRank algorithm constructs a graph through the co-occurrence relationship between words and uses the connection strength between words to determine the importance of keywords; in addition, a machine learning algorithm (such as classification and clustering) can be used to train the model, so as to use the trained model to extract keywords from the text field.

[0057] Furthermore, the enterprises to be identified can be determined through the historical waybill data of the enterprises. The historical waybill data includes the information of receiving and sending items, which can often reflect the industry and industrial chain position of the enterprise. Taking the screening of multiple initial enterprises in the home appliance industry as an example, for upstream enterprises, since the consignments of parts manufacturing enterprises and industrial raw material enterprises are special, generally mainly industrial materials, etc., they can use the classification of "other hardware tools | instruments and equipment" as a classification indicator to screen out enterprises that have sent the above-mentioned categories of items, so as to determine the upstream enterprises to be identified; and for mid- and downstream enterprises, these enterprises basically use a complete set of furniture or electrical appliances as their main consignments, so they can use the classification of "furniture" or "electrical appliances" as a classification indicator to screen out enterprises that have sent items of this category, so as to determine the mid- and downstream enterprises. Through the above method, multiple enterprises to be identified can be screened out from multiple initial enterprises. That is, as a possible implementation method of the embodiment of the present application, before the above step 101, it includes: obtaining historical waybill data of multiple initial enterprises, wherein the historical waybill data includes mail collection and delivery information; screening multiple initial enterprises according to the mail collection and delivery information and target industry corresponding to each initial enterprise to determine multiple enterprises to be identified.

[0058] It should be noted that the initial enterprises can be preliminarily screened out for enterprises related to the target industry using existing industry classifications, company names, etc. Taking the home appliance industry as an example, the enterprise whose company name includes the field "home appliances" can be directly determined as the initial enterprise. It is also possible to use industry classification information, for example, to preliminarily screen out the initial upstream parts manufacturers using the classification field containing "industrial raw materials"; to preliminarily screen out the initial upstream raw material production enterprises using the classification field containing "electronic components | industrial manufacturing"; and to more accurately screen out the midstream and downstream initial enterprises related to the home appliance industry using the classification field containing "home appliances". It should be understood that it is difficult to directly determine whether an enterprise is a manufacturer or seller of home appliances through a wide range of industry classifications, so it is impossible to accurately distinguish whether the initial enterprise's industrial chain position is midstream or downstream.

[0059] In one possible implementation, since the historical waybill data contains some noise and irrelevant information in addition to key waybill data such as waybill address information, name of the sending and receiving company, and sending and receiving area, the historical waybill data can be cleaned in advance using regular rules to remove non-Chinese and non-numeric symbols in the historical waybill data, and different deduplication operations can be performed for each waybill. For waybill address information, only Chinese and numeric characters can be retained, and Chinese numbers can be replaced with Arabic numbers. In addition, the waybill address may contain repeated written parts such as provinces, cities, and districts, so deduplication operations are required. For company names, replace the Arabic numbers filled in the sending company with Chinese numbers, and remove illegal characters such as spaces. The above steps complete the cleaning and deduplication of historical waybill data.

[0060] In one possible implementation, since the main recipients of midstream enterprises may be customers of B-end enterprises, and the downstream sales channels are mainly C-end recipients, for the midstream and downstream enterprises to be identified, it is possible to further determine whether the enterprise to be identified is a midstream enterprise or a downstream enterprise through the sending and receiving enterprise information contained in the sending and receiving information.

[0061] Step 102 , calculate the similarity between the business operation keywords of each enterprise to be identified and the industry characteristic words and phrases at the corresponding industrial chain position to obtain the industrial chain position similarity of each enterprise to be identified.

[0062] Among them, the position in the industrial chain is determined based on the historical waybill data of the enterprise to be identified.

[0063] It should be noted that the industry characteristic phrases of the industrial chain position are pre-set based on the target industry. Taking the home appliance industry as an example, the industry characteristic phrases corresponding to the upstream enterprises can be set to "industrial parts | valve parts | electrical appliances | metal materials", the industry characteristic phrases corresponding to the midstream enterprises can be set to "furniture manufacturing | home appliance manufacturing", and the industry characteristic phrases corresponding to the downstream enterprises can be set to "sales | distribution". It should be understood that the industry characteristic phrases corresponding to the industrial chain position can be pre-set in combination with the actual application scenario, and the embodiments of this application do not limit this.

[0064] In a possible implementation of the present application, a text similarity matching model can be built, such as a Sentence-BERT model, and then a preset text similarity matching model can be called to calculate the similarity between the business keywords of each enterprise to be identified and the industry characteristic words and phrases of the corresponding industrial chain position, so as to obtain the industrial chain position similarity of each enterprise to be identified. Among them, the Sentence-BERT model can fully mine the semantic information of the question and show better performance in the question similarity measurement.

[0065] Furthermore, since the industrial characteristic words and phrases corresponding to different industrial chain positions are different, it is necessary to perform corresponding similarity calculations for the upstream, midstream and downstream enterprises to be identified respectively. That is, as a possible implementation method of the embodiment of the present application, the above-mentioned step 102 may include: for the enterprises to be identified whose industrial chain position is upstream, the similarity between the enterprise business keywords and the industrial characteristic words and phrases corresponding to the upstream is calculated to obtain the industrial chain position similarity of each upstream enterprise to be identified; and, for the enterprises to be identified whose industrial chain position is midstream, the similarity between the enterprise business keywords and the industrial characteristic words and phrases corresponding to the midstream is calculated to obtain the industrial chain position similarity of each midstream enterprise to be identified; and, for the enterprises to be identified whose industrial chain position is downstream, the similarity between the enterprise business keywords and the industrial characteristic words and phrases corresponding to the downstream is calculated to obtain the industrial chain position similarity of each downstream enterprise to be identified.

[0066] See also Figure 2 , shows a schematic diagram of a process for calculating the similarity of a to-be-identified enterprise provided in an embodiment of the present application. Figure 2 As shown, first build a twin Sentence-BERT model, that is, a BERT model with two parameters bound, which are used to receive industry characteristic words and business operation keywords extracted in step 101 respectively. Then build a pooling layer to perform pooling operations on industry characteristic words and business operation keywords respectively, generate sentence vectors u and sentence vectors v of fixed dimensions, and ensure that the vector lengths of the next similarity calculation are consistent. Finally, use the following formula to calculate cosine similarity:

[0067]

[0068] Among them, i represents the index of the sentence vector; n represents the dimension of the sentence vector.

[0069] In a possible implementation, if the industrial chain position of a certain enterprise to be identified is unknown, the business operation keywords of the enterprise to be identified can be respectively calculated with the industrial characteristic words and phrases corresponding to each industrial chain position, and the one with the higher similarity in each calculation is taken as the similarity of the industrial chain position corresponding to the enterprise to be identified. For example, if the industrial chain position of an enterprise to be identified is midstream or downstream, it is necessary to calculate the similarity of the business operation keywords of the enterprise to be identified with the industrial characteristic words and phrases corresponding to the midstream, assuming that the calculation result is 0.9, and calculate the similarity with the industrial characteristic words and phrases corresponding to the downstream, assuming that the calculation result is 0.2, then 0.9 is taken as the similarity of the industrial chain position of the enterprise to be identified.

[0070] Step 103: Determine at least two target enterprises from the multiple enterprises to be identified based on the industrial chain position and industrial chain position similarity of each enterprise to be identified.

[0071] In the embodiment of the present application, the result of the cosine similarity calculation is between [-1, 1]. The larger the result, the more similar the texts are. By setting a similarity threshold, if the similarity of the industrial chain position corresponding to the enterprise to be identified is greater than the set similarity threshold, it means that the correlation of the enterprise to be identified is high, and the enterprise to be identified is determined as the target enterprise.

[0072] It should be noted that the similarity thresholds corresponding to different industrial chain positions may be the same, or may be adaptively adjusted based on the needs of different industrial chain positions, and the embodiments of the present application do not impose any restrictions on this.

[0073] As a possible implementation method, after determining the target enterprise, the accuracy of the target enterprise can be further verified by the proportion of consignment types in the historical waybill data corresponding to the target enterprise. Taking the home appliance industry as an example, assuming that in the consignment data of a downstream target enterprise, 90% of the consignments are documents, and only 10% of the consignments are "furniture | tables and chairs | air conditioners | TVs", it means that the target enterprise does not meet the general regulations of downstream enterprises in the industrial chain and needs to be deleted.

[0074] Step 104, determining the upstream and downstream relationships between the target enterprises.

[0075] In the embodiment of the present application, based on the industrial chain position corresponding to each target enterprise, each target enterprise is filled into the upstream, midstream and downstream position corresponding to the industrial chain to obtain the upstream and downstream relationship of the target industry.

[0076] As a possible implementation method of the embodiment of the present application, in order to better understand and evaluate the position and role of each target enterprise in the entire industrial chain, thereby helping decision makers, investors or researchers to better grasp the importance of industrial development and related enterprises, the importance of the enterprises to be identified at each industrial chain position can be ranked based on the historical waybill data, corporate information and bidding information of each target enterprise.

[0077] Specifically, the importance can be judged by assigning a score to the node enterprise. For historical waybill data, the number of waybills and freight can be aggregated for the target enterprise according to the consignment dimension, and multiple levels can be set according to the number of enterprise tickets and freight. The higher the level, the higher the corresponding waybill score. For enterprise information, the public staff size and registered capital of the target enterprise are used to judge the importance of the enterprise. Generally speaking, the larger the scale of enterprise staff and the higher the registered capital, the more important the position of the enterprise in the industrial chain, and the higher the enterprise score. For bidding information, the number and amount of bidding of the target enterprise in recent years are used to judge the importance. The larger the number and amount of enterprise bidding, the higher the enterprise bidding score. Finally, the final importance score of each target node is calculated by combining the waybill score, enterprise score and bidding score. Further, the Analytic Hierarchy Process (AHP) can be used to assign corresponding weights to the above three scores. For example, the weight corresponding to the waybill score is 0.43, the weight corresponding to the enterprise score is 0.14, and the weight corresponding to the bidding score is 0.43. It should be understood that the specific weight corresponding to each type of score can be determined in combination with the actual application scenario and requirements, and the embodiments of the present application do not limit this.

[0078] The industrial chain identification method disclosed in the above embodiment of the present application first determines the business keywords corresponding to each enterprise to be identified based on the business scope fields of multiple enterprises to be identified, and then calculates the similarity between the business keywords of each enterprise to be identified and the industrial characteristic words and phrases of the corresponding industrial chain position to obtain the industrial chain position similarity of each enterprise to be identified, and then determines at least two target enterprises from multiple enterprises to be identified based on the industrial chain position and industrial chain position similarity of each enterprise to be identified, and finally determines the upstream and downstream relationship between each target enterprise. Therefore, by determining the industrial chain position similarity, it is judged whether each enterprise to be identified is an enterprise at the corresponding industrial chain position, so as to accurately screen out the target enterprises, and then determine the upstream and downstream relationship between each target enterprise, and obtain a more accurate industrial chain.

[0079] See also Figure 3 , shows a flow chart of an industrial chain identification method provided in Example 2 of the present application. Figure 3 As shown, the industrial chain identification method may include the following steps:

[0080] Step 301 , performing data cleaning on the business scope field of each enterprise to be identified to determine a business scope cleansing text.

[0081] Among them, since the business scope of an enterprise is generally a long text, there may be redundant and repeated content, so it needs to be cleaned first. This can improve the accuracy and reliability of subsequent keyword extraction and avoid interference from irrelevant factors during the keyword extraction process.

[0082] Specifically, interference factors such as punctuation marks, spaces, Arabic numerals, and stop words in the business scope text of the enterprise can be removed to improve the accuracy of keyword extraction.

[0083] Step 302 , extract keywords from the cleaned text of the business scope of the enterprise to be identified to determine the business keywords corresponding to the enterprise to be identified.

[0084] Specifically, taking the LDA topic model as an example, a solution for extracting business keywords is specifically described. First, the business scope cleaning text of the enterprise to be identified is segmented to obtain multiple business scope short texts. For example, the word segmentation function of the jieba library can be used to segment the text into words, and the segmented text is saved in the form of a list. Then, the word frequency statistics of multiple business scope short texts are performed to build a bag of words model, wherein the bag of words model is a common text representation method. The bag of words model can represent the text as a frequency vector of words. Based on the text list after word segmentation, the corpora method of the gensim library can be used to perform word frequency statistics and sparsely vectorize the words. Then, the preset latent Dirichlet distribution LDA topic model is called to process the bag of words model to determine the weight corresponding to each business scope short text, wherein the LDA topic model is constructed based on the sparsely vectorized sentence samples in advance and the corresponding number of topics is set, and then the LDA topic model is used to analyze the weight in each topic. Finally, based on the weight corresponding to each business scope short text, the business operation keywords corresponding to the enterprise to be identified are determined. Among them, the short text with the largest weight in each topic can be used as the keyword of the enterprise's business scope, and the extracted keywords are combined into keyword sentences to obtain the enterprise business keywords.

[0085] Step 303 , calculate the similarity between the business operation keywords of each enterprise to be identified and the industry characteristic words and phrases at the corresponding industrial chain position to obtain the industrial chain position similarity of each enterprise to be identified.

[0086] Among them, the position in the industrial chain is determined based on the historical waybill data of the enterprise to be identified.

[0087] Step 304: Determine at least two target enterprises from the multiple enterprises to be identified based on the industrial chain position and industrial chain position similarity of each enterprise to be identified.

[0088] Step 305, determining the upstream and downstream relationships between the target enterprises.

[0089] The specific implementation process and principle of the above steps 303-305 can refer to the detailed description of the above embodiment, which will not be repeated here.

[0090] The industrial chain identification method disclosed in the above embodiment of the present application first performs data cleaning on the business scope field of each enterprise to be identified to determine the business scope cleaning text, then performs keyword extraction on the business scope cleaning text of the enterprise to be identified to determine the business operation keywords corresponding to the enterprise to be identified, and then calculates the industrial chain position similarity based on the business operation keywords, further screens the target enterprises and determines the upstream and downstream relationships between the target enterprises. Thus, through data cleaning and keyword extraction, the business operation keywords corresponding to the enterprise to be identified are accurately obtained.

[0091] See also Figure 4 , shows a structural schematic diagram of an industrial chain identification device provided in Example 3 of the present application. For the sake of convenience of explanation, only the parts related to the example of the present application are shown.

[0092] The industrial chain identification device may specifically include the following modules:

[0093] The keyword determination module 401 is used to determine the business keywords corresponding to each enterprise to be identified according to the business scope fields of multiple enterprises to be identified, wherein the multiple enterprises to be identified are all enterprises in the target industry.

[0094] The similarity determination module 402 is used to calculate the similarity between the business operation keywords of each enterprise to be identified and the industrial characteristic words and phrases of the corresponding industrial chain position to obtain the similarity of the industrial chain position of each enterprise to be identified, wherein the industrial chain position is determined based on the historical waybill data of the enterprise to be identified.

[0095] The enterprise determination module 403 is used to determine at least two target enterprises from multiple enterprises to be identified according to the industrial chain position and industrial chain position similarity of each enterprise to be identified.

[0096] The industrial chain identification module 404 is used to determine the upstream and downstream relationships between various target enterprises.

[0097] The industrial chain identification device disclosed in the above embodiment of the present application first determines the business keywords corresponding to each enterprise to be identified based on the business scope fields of multiple enterprises to be identified, and then calculates the similarity between the business keywords of each enterprise to be identified and the industrial characteristic words and phrases of the corresponding industrial chain position to obtain the industrial chain position similarity of each enterprise to be identified, and then determines at least two target enterprises from multiple enterprises to be identified based on the industrial chain position and industrial chain position similarity of each enterprise to be identified, and finally determines the upstream and downstream relationship between each target enterprise. Therefore, by determining the industrial chain position similarity, it is judged whether each enterprise to be identified is an enterprise at the corresponding industrial chain position, so as to accurately screen out the target enterprises, and then determine the upstream and downstream relationship between each target enterprise, and obtain a more accurate industrial chain.

[0098] In a possible implementation of the third embodiment of the present application, the industrial chain identification device may further include the following modules:

[0099] The first acquisition module is used to acquire historical waybill data of multiple initial enterprises, wherein the historical waybill data includes delivery and collection information.

[0100] The first screening module is used to screen multiple initial enterprises according to the mail collection and delivery information and target industries corresponding to each initial enterprise to determine multiple enterprises to be identified.

[0101] Furthermore, in another possible implementation of the third embodiment of the present application, the keyword determination module 401 may specifically include the following submodules:

[0102] The first determination submodule is used to perform data cleaning on the business scope field of each enterprise to be identified to determine the business scope cleaning text.

[0103] The second determination submodule is used to extract keywords from the business scope cleansing text of the enterprise to be identified, so as to determine the business operation keywords corresponding to the enterprise to be identified.

[0104] Furthermore, in another possible implementation of the third embodiment of the present application, the second determining submodule may specifically include the following units:

[0105] The first determination unit is used to segment the business scope cleansing text of the enterprise to be identified to obtain multiple business scope short texts.

[0106] The first statistical unit is used to perform word frequency statistics on multiple business scope short texts to build a bag-of-words model.

[0107] The second determination unit is used to call a preset latent Dirichlet distribution topic model to process the bag-of-words model to determine the weight corresponding to each business scope short text.

[0108] The third determination unit is used to determine the business operation keywords corresponding to the enterprise to be identified based on the weight corresponding to each business scope short text.

[0109] Furthermore, in another possible implementation of the third embodiment of the present application, the similarity determination module 402 may specifically include the following submodules:

[0110] The third determination submodule is used to call a preset text similarity matching model to calculate the similarity between the business operation keywords of each enterprise to be identified and the industrial characteristic words and phrases of the corresponding industrial chain position, so as to obtain the industrial chain position similarity of each enterprise to be identified.

[0111] Furthermore, in another possible implementation of the third embodiment of the present application, the above-mentioned industrial chain position includes upstream, midstream and downstream, and the above-mentioned similarity determination module 402 may further include the following submodules:

[0112] The fourth determination submodule is used to calculate the similarity between the enterprise business keywords and the corresponding industry characteristic words and phrases in the upstream for the enterprise to be identified whose industrial chain position is upstream, so as to obtain the industrial chain position similarity of each upstream enterprise to be identified; and, for the enterprise to be identified whose industrial chain position is midstream, calculate the similarity between the enterprise business keywords and the corresponding industry characteristic words and phrases in the midstream, so as to obtain the industrial chain position similarity of each midstream enterprise to be identified; and, for the enterprise to be identified whose industrial chain position is downstream, calculate the similarity between the enterprise business keywords and the corresponding industry characteristic words and phrases in the downstream, so as to obtain the industrial chain position similarity of each downstream enterprise to be identified.

[0113] Furthermore, in another possible implementation of the third embodiment of the present application, the industrial chain identification device may further include the following modules:

[0114] The importance ranking module is used to rank the importance of the enterprises to be identified at each position in the industrial chain based on the historical waybill data, enterprise information and bidding information of each target enterprise.

[0115] The industrial chain identification device disclosed in the above embodiment of the present application first determines the business keywords corresponding to each enterprise to be identified based on the business scope fields of multiple enterprises to be identified, and then calculates the similarity between the business keywords of each enterprise to be identified and the industrial characteristic words and phrases of the corresponding industrial chain position to obtain the industrial chain position similarity of each enterprise to be identified, and then determines at least two target enterprises from multiple enterprises to be identified based on the industrial chain position and industrial chain position similarity of each enterprise to be identified, and finally determines the upstream and downstream relationship between each target enterprise. Therefore, by determining the industrial chain position similarity, it is judged whether each enterprise to be identified is an enterprise at the corresponding industrial chain position, so as to accurately screen out the target enterprises, and then determine the upstream and downstream relationship between each target enterprise, and obtain a more accurate industrial chain.

[0116] The industrial chain identification device provided in the embodiment of the present application can be applied in the aforementioned method embodiment. For details, please refer to the description of the aforementioned method embodiment, which will not be repeated here.

[0117] Figure 5 Schematic diagram of the structure of the terminal device provided in the fourth embodiment of the present application. Figure 5 As shown, the terminal device 500 of this embodiment includes: at least one processor 510 ( Figure 5 Only one is shown in the figure) a processor, a memory 520, and a computer program 521 stored in the memory 520 and executable on the at least one processor 510, and when the processor 510 executes the computer program 521, the steps in the above-mentioned industrial chain identification method embodiment are implemented.

[0118] The terminal device 500 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal device may include, but is not limited to, a processor 510 and a memory 520. Those skilled in the art will appreciate that Figure 5 It is only an example of the terminal device 500 and does not constitute a limitation on the terminal device 500. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components, for example, it may also include input and output devices, network access devices, etc.

[0119] The processor 510 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0120] In some embodiments, the memory 520 may be an internal storage unit of the terminal device 500, such as a hard disk or memory of the terminal device 500. In other embodiments, the memory 520 may also be an external storage device of the terminal device 500, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the terminal device 500. Further, the memory 520 may also include both an internal storage unit of the terminal device 500 and an external storage device. The memory 520 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program, etc. The memory 520 may also be used to temporarily store data that has been output or is to be output.

[0121] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0122] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0123] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0124] In the embodiments provided in the present application, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0125] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0126] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0127] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0128] The present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed through a computer program product. When the computer program product runs on a terminal device, the terminal device can implement the steps in the above-mentioned method embodiments when executing.

[0129] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application is described in detail with reference to the above-mentioned embodiments, a person skilled in the art should understand that the technical solutions described in the above-mentioned embodiments can still be modified, or some of the technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for identifying an industrial chain, characterized in that: include: Determine the business keywords corresponding to each of the enterprises to be identified according to the business scope fields of the enterprises to be identified, wherein the enterprises to be identified are all enterprises in the target industry; Calculate the similarity between the business operation keywords of each of the to-be-identified enterprises and the industry characteristic words and phrases at the corresponding industrial chain position to obtain the industrial chain position similarity of each of the to-be-identified enterprises, wherein the industrial chain position is determined based on the historical waybill data of the to-be-identified enterprises; Determining at least two target enterprises from the multiple enterprises to be identified according to the industrial chain position of each enterprise to be identified and the similarity of the industrial chain position; Determine the upstream and downstream relationships between each of the target enterprises.

2. The industrial chain identification method according to claim 1, characterized in that: Before determining the business keywords corresponding to each of the enterprises to be identified based on the business scope fields of the enterprises to be identified, the method further includes: Acquire historical waybill data of multiple initial enterprises, wherein the historical waybill data includes delivery and collection information; The multiple initial enterprises are screened according to the mail collection and delivery information corresponding to each of the initial enterprises and the target industry to determine the multiple enterprises to be identified.

3. The industrial chain identification method according to claim 1, characterized in that: The step of determining the business operation keywords corresponding to each of the enterprises to be identified based on the business scope fields of the enterprises to be identified includes: Performing data cleaning on the business scope field of each of the to-be-identified enterprises to determine a business scope cleaning text; Keyword extraction is performed on the cleaned text of the business scope of the enterprise to be identified to determine the business operation keywords corresponding to the enterprise to be identified.

4. The industrial chain identification method according to claim 3, characterized in that: The keyword extraction of the business scope cleansing text of the enterprise to be identified to determine the business operation keywords corresponding to the enterprise to be identified includes: Segment the business scope cleansing text of the enterprise to be identified to obtain multiple business scope short texts; Performing word frequency statistics on the multiple business scope short texts to construct a bag-of-words model; Calling a preset latent Dirichlet distribution topic model to process the bag-of-words model, and determining a weight corresponding to each of the business scope short texts; Based on the weight corresponding to each of the business scope short texts, the business operation keywords corresponding to the to-be-identified enterprise are determined.

5. The industrial chain identification method according to claim 1, characterized in that: The similarity calculation of the business operation keywords of each of the to-be-identified enterprises and the industry characteristic words and phrases at the corresponding industrial chain position to obtain the industrial chain position similarity of each of the to-be-identified enterprises includes: The preset text similarity matching model is called to calculate the similarity between the business operation keywords of each enterprise to be identified and the industry characteristic words and phrases at the corresponding industrial chain position, so as to obtain the industrial chain position similarity of each enterprise to be identified.

6. The industrial chain identification method according to claim 1, characterized in that: The industrial chain position includes upstream, midstream and downstream. The similarity calculation of the business operation keywords of each enterprise to be identified and the industrial characteristic words and phrases of the corresponding industrial chain position is performed to obtain the industrial chain position similarity of each enterprise to be identified, including: For the enterprises to be identified whose industrial chain position is upstream, the similarity between the enterprise business keywords and the corresponding industrial characteristic words and phrases of the upstream is calculated to obtain the industrial chain position similarity of each upstream enterprise to be identified, and, For the enterprises to be identified whose industrial chain position is in the middle, the similarity between the enterprise operation keywords and the industrial characteristic words and phrases corresponding to the middle is calculated to obtain the industrial chain position similarity of each enterprise to be identified in the middle, and, For the enterprises to be identified that are located downstream in the industrial chain, similarity is calculated between the enterprise business keywords and the corresponding downstream industrial characteristic words and phrases to obtain the industrial chain position similarity of each downstream enterprise to be identified.

7. The industrial chain identification method according to claim 1, characterized in that: After determining the upstream and downstream relationships between the target enterprises, the method further includes: Based on the historical waybill data, enterprise information and bidding information of each target enterprise, the enterprises to be identified at each position in the industrial chain are ranked in importance.

8. An industrial chain identification device, characterized in that: include: A keyword determination module, used to determine the business keywords corresponding to each of the enterprises to be identified according to the business scope fields of the enterprises to be identified, wherein the enterprises to be identified are all enterprises in the target industry; A similarity determination module is used to calculate the similarity between the business operation keywords of each of the to-be-identified enterprises and the industry characteristic words and phrases of the corresponding industrial chain position, so as to obtain the similarity of the industrial chain position of each of the to-be-identified enterprises, wherein the industrial chain position is determined based on the historical waybill data of the to-be-identified enterprises; An enterprise determination module, configured to determine at least two target enterprises from the plurality of enterprises to be identified according to the industrial chain position of each enterprise to be identified and the similarity of the industrial chain position; The industrial chain identification module is used to determine the upstream and downstream relationships between the target enterprises.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images