Industrial chain identification method and system

Abstract

The present invention provides an industrial chain identification method and system, which obtains transaction flow data, pre-processes the transaction flow data, obtains operating transaction flow data and key information for constructing a capital map; constructs a capital map based on the operating transaction flow data and key information; based on the capital map, uses an association mining algorithm to mine industry sequences that meet a support threshold in the capital map to obtain an industry frequent sequence set; uses the k industry sequences with the highest industry association as chain sequences to obtain and display an industrial chain map. In this solution, a capital map is constructed using transaction flow data, and an association mining algorithm is used to mine industry sequences that meet a support threshold in the capital map, and the k industry sequences with the highest industry association are used as chain sequences to obtain and display an industrial chain map, thereby being able to identify the degree of association between industries, improve the integrity and adequacy of identification, reduce identification time, and improve versatility.

Description

Technical Field

[0001] The present invention relates to the technical field of transaction data processing, and in particular to an industrial chain identification method and system. Background Art

[0002] The industrial chain is a relatively macro concept that describes the technical and economic relationship between industrial sectors. There are a large number of upstream and downstream relationships and value exchanges in the industrial chain. In order to understand the technical and economic relationship between industrial sectors, it is usually necessary to outline the relationship between industries and form an industrial chain through industry research reports, media information, and corporate visits.

[0003] However, this method of industrial chain identification relies heavily on manpower and research, and has strong subjective judgments, resulting in incomplete and insufficient identification of the industrial chain. In addition, it takes a long time to research and sort out the industrial chain, and it is difficult to obtain industrial transformation data in a timely manner. In addition, the industrial chain constructed through industry research lacks quantification of the relationship between the strengths and weaknesses of the industry, and the degree of connection between industries cannot be identified. Different institutions use different industry segmentation standards to construct industrial chains, and the industry systems of the constructed industrial chains are also different. When it comes to specific applications, different industry systems cannot fully correspond to the national standard industry divisions, and have poor versatility.

[0004] It can be seen from this that the use of existing industrial chain identification methods for industrial chain identification has problems such as incomplete and insufficient identification, a long time, inability to identify the degree of correlation between industries, and poor versatility. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides an industrial chain identification method and system to solve the problems of incomplete and insufficient identification, long time, inability to identify the degree of correlation between industries and poor versatility in existing industrial chain identification methods.

[0006] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0007] A first aspect of an embodiment of the present invention discloses a method for identifying an industrial chain, the method comprising:

[0008] Acquire transaction flow data, pre-process the transaction flow data, obtain operating transaction flow data and key information for constructing a capital map, wherein the key information at least includes the enterprises involved in the transaction, the transaction direction, and the industry category to which the enterprises involved in the transaction belong;

[0009] Constructing a capital map based on the operating transaction flow data and the key information;

[0010] Based on the capital graph, the industry sequence that meets the support threshold in the capital graph is mined using the association mining algorithm PrefixSpan to obtain an industry frequent sequence set, wherein the industry sequence has an industry correlation, and the industry correlation is the weight of the industry sequence in the industrial chain;

[0011] The k industry sequences with the highest industrial correlation are taken as chain sequences to obtain and display the industrial chain map, where k is a positive integer.

[0012] Optionally, the acquiring of transaction flow data and preprocessing of the transaction flow data to obtain operating transaction flow data and key information for constructing a capital map include:

[0013] Acquire transaction flow data, perform word segmentation processing on the transaction flow data, and obtain word segmentation results;

[0014] Based on the word segmentation result, determining a word representing a purpose in the word segmentation result;

[0015] Acquire words representing purposes in different types of transaction flow data to obtain a word set representing purposes;

[0016] Based on the words representing the purpose in the word segmentation results and the word set representing the purpose, the key information for operating transaction flow data and constructing a capital map is obtained.

[0017] Optionally, constructing a capital map based on the operating transaction flow data and the key information includes:

[0018] Obtaining the enterprises involved in the transaction and the capital outflow relationship in the operating transaction flow data;

[0019] A directed capital graph is constructed with the enterprise as a node, the capital outflow relationship as a relationship edge, and the number of transactions and the amount as attributes of the relationship edge.

[0020] Optionally, based on the capital graph, the association mining algorithm PrefixSpan is used to mine industry sequences that meet the support threshold in the capital graph to obtain an industry frequent sequence set, including:

[0021] According to the capital map, determine the industrial chain core and industry sequence;

[0022] Taking each of the nodes on the capital graph as a root node, a full path search is performed in the capital graph to obtain the one-way capital flow relationship edges occurring at the root node within a preset number of capital transfers within a preset period, and obtain an industry path;

[0023] The minimum value of each actual capital flow on the industry path is used as the weight on the industry path, and the minimum edge weight on the industry path is obtained according to the weight on the industry path;

[0024] The minimum edge weight on the industry path is used as the weight multiple of the industry sequence to obtain an industry sequence with industrial correlation;

[0025] Based on a preset support threshold and an association mining algorithm PrefixSpan, mining the industry sequence that meets the support threshold in the capital graph from the industry sequence with industry association to obtain a frequent industry sequence;

[0026] If the frequent industry sequence meets the requirements for displaying the industrial chain map, output all the frequent industry sequences that meet the requirements for displaying the industrial chain map to obtain an industry frequent sequence set;

[0027] If the frequent industry sequence does not meet the display requirements of the industrial chain map, the industry sequence that meets the support threshold in the capital map is mined from the industry sequence with industrial correlation based on the set support threshold and the associated mining algorithm PrefixSpan, until the frequent industry sequence meets the display requirements of the industrial chain map.

[0028] Optionally, based on a preset support threshold and an association mining algorithm PrefixSpan, mining industry sequences that meet the support threshold in the capital map from the industry sequences with industry association to obtain frequent industry sequences includes:

[0029] According to the weight multiples of the industry sequence, the industry sequence with industry correlation is sorted by using the association mining algorithm PrefixSpan, and the prefix sequence and the suffix sequence in the industry sequence with industry correlation are determined, each of the industry sequences with industry correlation contains multiple item sets, and each of the item sets contains one or more elements;

[0030] The suffix sequence is converted into the prefix sequence based on a recursive algorithm until the industry sequence with industrial relevance does not meet the support threshold, thereby obtaining a frequent industry sequence.

[0031] Optionally, the k industry sequences with the highest industry correlation are used as chain sequences to obtain and display the industrial chain map, including:

[0032] Obtain k industry sequences with the highest industry correlation in the industry frequent sequence set, and use the k industry sequences with the highest industry correlation as chain sequences;

[0033] According to the industrial relevance of each of the frequent industry sequences in the frequent industry sequence set, the frequent industry sequences are sorted in reverse order to obtain a reverse order result;

[0034] Taking the chain sequence as the backbone of the industrial chain map, the frequent industry sequences are added to the industrial chain map in sequence according to the reverse sorting results to obtain and display the industrial chain map.

[0035] A second aspect of an embodiment of the present invention discloses an industrial chain identification system, the system comprising:

[0036] A preprocessing module is used to obtain transaction flow data, preprocess the transaction flow data, obtain operating transaction flow data and key information for constructing a capital map, wherein the key information at least includes the enterprises involved in the transaction, the transaction direction, and the industry category to which the enterprises involved in the transaction belong;

[0037] A construction module, used to construct a capital map based on the operating transaction flow data and the key information;

[0038] A mining module is used to mine industry sequences that meet a support threshold in the capital map using an association mining algorithm PrefixSpan based on the capital map to obtain a frequent industry sequence set, wherein the industry sequence has an industry association degree, and the industry association degree is a weight of the industry sequence in the industrial chain;

[0039] The module is used to take the k industry sequences with the highest industrial correlation as chain sequences, obtain and display the industrial chain map, and k is a positive integer.

[0040] Optionally, the preprocessing module includes:

[0041] A word segmentation processing unit, used to obtain transaction flow data, perform word segmentation processing on the transaction flow data, and obtain a word segmentation result;

[0042] A determination unit, configured to determine, based on the word segmentation result, a word representing a purpose in the word segmentation result;

[0043] An acquisition unit, used to acquire words representing purposes in different types of transaction flow data to obtain a word set representing purposes;

[0044] The obtaining unit is used to obtain the key information of the operating transaction flow data and the construction of the capital map based on the words representing the purpose and the word set representing the purpose in the word segmentation result.

[0045] Optionally, the building blocks include:

[0046] An acquisition unit, used to acquire the enterprises involved in the transaction and the capital outflow relationship in the operating transaction flow data;

[0047] A construction unit is used to construct a directed capital graph with the enterprise as a node, the capital outflow relationship as a relationship edge, and the number of transactions and the amount as attributes of the relationship edge.

[0048] Optionally, the mining module includes:

[0049] A determination unit, used to determine the industrial chain core and industry sequence according to the capital map;

[0050] A search and acquisition unit, configured to take each node on the capital graph as a root node, perform a full path search in the capital graph, obtain the one-way capital flow relationship edge occurring at the root node within a preset capital transfer number within a preset period, and obtain an industry path;

[0051] The obtaining unit is used to use the minimum value of each actual capital flow on the industry path as the weight on the industry path, and obtain the minimum edge weight on the industry path according to the weight on the industry path; use the minimum edge weight on the industry path as the weight multiple of the industry sequence to obtain the industry sequence with industry correlation;

[0052] A mining unit, configured to mine industry sequences that meet the support threshold in the capital graph from the industry sequences with industry associations based on a preset support threshold and an association mining algorithm PrefixSpan, to obtain frequent industry sequences;

[0053] The first mining processing unit is used for outputting all the frequent industry sequences that meet the requirements of the industrial chain map display if the frequent industry sequences meet the requirements of the industrial chain map display, so as to obtain an industry frequent sequence set;

[0054] The second mining processing unit is used to mine the industry sequences that meet the support threshold in the capital map based on the set support threshold and the associated mining algorithm PrefixSpan if the frequent industry sequence does not meet the display requirements of the industrial chain map, until the frequent industry sequence meets the display requirements of the industrial chain map.

[0055] Based on the above-mentioned embodiment of the present invention, a method and system for identifying an industrial chain is provided, the method includes: obtaining transaction flow data, preprocessing the transaction flow data, obtaining operating transaction flow data and key information for constructing a capital map, the key information at least including the enterprises involved in the transaction, the transaction direction and the industry category to which the enterprises involved in the transaction belong; constructing a capital map based on the operating transaction flow data and the key information; based on the capital map, using the association mining algorithm PrefixSpan to mine the industry sequence that meets the support threshold in the capital map, and obtaining an industry frequent sequence set, the industry sequence has an industry correlation, and the industry correlation is the weight of the industry sequence in the industrial chain; taking the k industry sequences with the highest industry correlation as chain sequences, obtaining an industrial chain map and displaying it. In this scheme, the capital map is constructed using transaction flow data, and the association mining algorithm PrefixSpan is used to mine the industry sequence that meets the support threshold in the capital map, and the k industry sequences with the highest industry correlation are taken as chain sequences, and the industrial chain map is obtained and displayed, so that the degree of correlation between industries can be identified, the integrity and sufficiency of identification can be improved, the identification time can be reduced, and the versatility can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0057] Figure 1 A schematic diagram of a process flow of an industrial chain identification method provided by an embodiment of the present invention;

[0058] Figure 2 A schematic diagram of a preprocessing process provided by an embodiment of the present invention;

[0059] Figure 3 A schematic diagram of a process for constructing a capital map provided by an embodiment of the present invention;

[0060] Figure 4 An application scenario diagram for constructing a capital map provided by an embodiment of the present invention;

[0061] Figure 5 A schematic diagram of a process for mining industry sequences that meet a support threshold in a capital graph provided by an embodiment of the present invention;

[0062] Figure 6 A schematic diagram of a process for constructing a frequent industry sequence provided by an embodiment of the present invention;

[0063] Figure 7 A schematic diagram of a process for obtaining an industrial chain map provided by an embodiment of the present invention;

[0064] Figure 8 A schematic diagram of the structure of an industrial chain identification system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0065] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] In this application, the terms "comprises", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element.

[0067] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein.

[0068] In order to facilitate understanding of the technical solution of the present invention, the technical terms appearing in the present invention are explained:

[0069] Chain core: the core industry in the industrial chain.

[0070] Industrial chain: Industrial chain is a concept in industrial economics. It is a chain-like relationship between various industrial sectors that is objectively formed based on certain technical and economic connections and according to specific logical relationships and time-space layout relationships.

[0071] PrefixSpan (Prefix-Projected Pattern Growth) algorithm: can mine frequent sequences that meet the minimum support.

[0072] Support: refers to the proportion of the number of times several related data appear in the data set to the total data set.

[0073] As can be seen from the background technology, the existing industrial chain identification methods are used for industrial chain identification, but there are problems such as incomplete and insufficient identification, long time, inability to identify the degree of correlation between industries, and poor versatility.

[0074] Therefore, an embodiment of the present invention provides an industrial chain identification method and system. In this solution, transaction flow data is used to construct a capital map, and the association mining algorithm PrefixSpan is used to mine the industry sequences that meet the support threshold in the capital map. The industry sequence with the highest industry association is used as the chain sequence to obtain and display the industrial chain map, so as to identify the degree of association between industries, improve the completeness and adequacy of identification, reduce identification time, and improve versatility.

[0075] like Figure 1 FIG. 1 is a flow chart of an industrial chain identification method provided by an embodiment of the present invention, and the method mainly includes the following steps:

[0076] Step S101: Acquire transaction flow data, pre-process the transaction flow data, and obtain operating transaction flow data and key information for constructing a capital map.

[0077] In step S101, the key information at least includes the enterprises participating in the transaction, the transaction direction, and the industry category to which the enterprises participating in the transaction belong.

[0078] Among them, the industry categories of the enterprises participating in the transaction use the "National Economic Industry Classification".

[0079] In the process of implementing step S101, when financial transactions are conducted between enterprises, a large amount of transaction flow data will be generated. These transaction flow data are obtained and integrated, and pre-processed to obtain key information for constructing a financial map, such as operating transaction flow data and the enterprises involved in the transaction, transaction direction, and industry categories of the enterprises involved in the transaction.

[0080] Step S102: Construct a capital map based on operating transaction flow data and key information.

[0081] In the specific implementation of step S102, a capital map is constructed based on the operating transaction flow data and key information such as the enterprises participating in the transaction, the transaction direction, and the industry category to which the enterprises participating in the transaction belong.

[0082] Step S103: Based on the capital graph, the association mining algorithm PrefixSpan is used to mine the industry sequences that meet the support threshold in the capital graph to obtain the industry frequent sequence set.

[0083] In step S103, the industry sequence has an industry correlation, and the industry correlation is the weight of the industry sequence in the industrial chain.

[0084] The support threshold may be set to 0.001 or 0.002, which is not limited in the present invention.

[0085] In the process of implementing step S103, the target industry to be mined is determined according to the capital map, and the industrial chain core and industry sequence of the target industry are determined according to the target industry, and a support threshold is set. Based on the set support threshold, the association mining algorithm PrefixSpan is used to mine the industry sequence in the capital map that meets the set support threshold to obtain the industry frequent sequence set.

[0086] Step S104: Take the k industry sequences with the highest industry correlation as chain sequences, obtain the industrial chain map and display it.

[0087] Wherein, k is a positive integer.

[0088] In step S104, NetworkX can be used to display the industrial chain map.

[0089] In the specific implementation of step S104, the k industry sequences with the highest industry correlation in the above-mentioned industry frequent sequence set are obtained, and the k industry sequences with the highest industry correlation are obtained. The k industry sequences with the highest industry correlation are used as chain sequences to obtain the industrial chain map and input it into NetworkX for display.

[0090] For example, the top 100 industry sequences with the highest industry correlation in the above-mentioned industry frequent sequence set are obtained, and the top 100 industry sequences with the highest industry correlation are obtained. The top 100 industry sequences with the highest industry correlation are used as chain sequences to obtain the industrial chain map and input it into NetworkX for display.

[0091] Based on an industrial chain identification method provided by an embodiment of the present invention, by obtaining transaction flow data, pre-processing the transaction flow data, obtaining operating transaction flow data and key information for constructing a capital map, the key information at least includes the enterprises involved in the transaction, the transaction direction and the industry category to which the enterprises involved in the transaction belong; constructing a capital map based on the operating transaction flow data and key information; based on the capital map, using the association mining algorithm PrefixSpan to mine the industry sequence that meets the support threshold in the capital map, obtaining an industry frequent sequence set, the industry sequence has an industry correlation, and the industry correlation is the weight of the industry sequence in the industrial chain; taking the k industry sequences with the highest industry correlation as chain sequences, obtaining an industrial chain map and displaying it. In this scheme, the capital map is constructed using transaction flow data, and the association mining algorithm PrefixSpan is used to mine the industry sequence that meets the support threshold in the capital map, and the k industry sequences with the highest industry correlation are taken as chain sequences, obtaining an industrial chain map and displaying it, so that the degree of correlation between industries can be identified, the integrity and sufficiency of identification can be improved, the identification time can be reduced, and the versatility can be improved.

[0092] Based on the industrial chain identification method provided by the above embodiment of the present invention, step S101 is executed to obtain transaction flow data, pre-process the transaction flow data, and obtain operating transaction flow data and key information for constructing a capital map. Figure 2 As shown, a schematic diagram of a preprocessing process provided by an embodiment of the present invention mainly includes the following steps:

[0093] Step S201: Acquire transaction flow data, perform word segmentation processing on the transaction flow data, and obtain word segmentation results.

[0094] In step S201, Jieba word segmentation may be used to perform word segmentation processing on the transaction flow data, or other word segmentation tools may be used to perform word segmentation processing on the transaction flow data, such as SnowNLP and THULAC, etc., which are not limited in the present invention.

[0095] It should be noted that Jieba word segmentation is a Ptyhon Chinese word component that can perform operations such as word segmentation, part-of-speech tagging and keyword extraction on Chinese text, and supports custom dictionaries.

[0096] In the process of implementing step S201, when financial transactions are conducted between enterprises, a large amount of transaction flow data will be generated. These transaction flow data are obtained, and Jieba word segmentation is used to parse the transaction information such as transaction purpose and transaction notes in the transaction flow data. The unstructured fields in the transaction flow data are segmented, and the unstructured data is converted into structured data to obtain the word segmentation results.

[0097] Step S202: Based on the word segmentation results, determine the words representing the usage in the word segmentation results.

[0098] In the specific implementation process of step S202, based on the word segmentation results, the words representing the purpose in the word segmentation results are determined, and a fund purpose rule base is constructed based on the words representing the purpose, and the fund purpose rule base is used to identify and judge the transaction purpose and the transaction counterparty.

[0099] Step S203: Acquire words representing purposes in different types of transaction flow data to obtain a word set representing purposes.

[0100] In step S203, the types of transaction flow data include operating transaction flow data, financing transaction flow data, investment flow data and account adjustment transaction flow data, but are not limited thereto.

[0101] In the specific implementation of step S203, the words representing the purpose in different types of transaction flow data are stored in a set, and these words representing the purpose are obtained to obtain a word set representing the purpose.

[0102] Step S204: Based on the words representing the purpose and the word sets representing the purpose in the word segmentation results, the key information for constructing the operating transaction flow data and the capital map is obtained.

[0103] In the process of implementing step S204, based on the words representing the purpose and the word set representing the purpose in the word segmentation results, the operating transaction flow data, financing transaction flow data, investment transaction flow data, account adjustment transaction flow data and key information for constructing the capital map are obtained, and the supply and demand relationship between industries is identified according to the capital transaction network. The operating transaction flow data is used as the data basis for constructing the capital map, because the financing transaction flow data and investment flow data reflect the investment and financing activities of the enterprise rather than the actual operating activities, which will interfere with the accuracy of industrial chain identification.

[0104] An industrial chain identification method provided based on an embodiment of the present invention performs word segmentation on transaction flow data, and uses the obtained word segmentation results to distinguish the types of transaction flow data, and then determines the type of transaction flow data, so as to identify the degree of correlation between industries, improve the completeness and adequacy of identification, reduce identification time, and improve versatility.

[0105] Based on the industrial chain identification method provided by the above embodiment of the present invention, step S102 is executed to construct a capital map based on the operating transaction flow data and key information. Figure 3 As shown, a schematic diagram of a process of constructing a capital map provided by an embodiment of the present invention mainly includes the following steps:

[0106] Step S301: Obtain the enterprises involved in the transaction and the capital outflow relationship in the operating transaction flow data.

[0107] In the specific implementation of step S301, the enterprises participating in the transaction and the capital outflow relationship when the capital transaction is carried out between the enterprises are obtained from the obtained operating transaction flow data.

[0108] Step S302: construct a directed capital graph with the enterprise as the node, the capital outflow relationship as the relationship edge, and the number of transactions and the amount as the attributes of the relationship edge.

[0109] It should be noted that the capital map is directional. For example, there are two companies A and B. Each transaction between the two companies is in the direction of A->B or B->A. Whether the transaction between the two companies is bidirectional can be determined based on the actual data.

[0110] The Funding Map contains available funding payment information for all industries.

[0111] In the specific implementation of step S302, the enterprises participating in the transaction are taken as nodes of the capital graph, the capital outflow relationship is taken as the relationship edge of the capital graph, and the number and amount of transactions generated when capital transactions are conducted between enterprises are taken as attributes of the relationship edge of the capital graph to construct a directed capital graph.

[0112] For example, Figure 4 As shown, it is an application scenario diagram for constructing a capital map provided by an embodiment of the present invention.

[0113] exist Figure 4 In the figure, A, B, C, D and E are nodes of the capital graph, that is, the enterprises involved in the transaction. The arrows represent the relationship edges of the capital graph, that is, the capital outflow relationship between the enterprises involved in the transaction. For example, A->B represents the capital outflow relationship when enterprises A and B conduct capital transactions, specifically, the capital flows from enterprise A to enterprise B. The arrows are marked with "(number of transactions, amount)", which represents the attributes of the relationship edges of the capital graph. Therefore, based on the above content, a directed capital graph can be constructed.

[0114] An industrial chain identification method provided based on an embodiment of the present invention forms a directed capital map by constructing nodes, relationship edges and attributes of relationship edges of a capital map, thereby providing a guarantee for the subsequent identification of the degree of correlation between industries, avoiding low accuracy when calculating the weight of a single transaction, improving the completeness and sufficiency of identification, reducing identification time, and improving versatility.

[0115] Based on the industrial chain identification method provided by the above embodiment of the present invention, step S103 is performed to mine industry sequences that meet the support threshold in the capital map using the association mining algorithm PrefixSpan based on the capital map to obtain the industry frequent sequence set. Figure 5 As shown, a schematic diagram of a process of mining industry sequences that meet a support threshold in a capital graph provided by an embodiment of the present invention mainly includes the following steps:

[0116] Step S501: Determine the industrial chain core and industry sequence based on the capital map.

[0117] In the specific implementation of step S501, the target industry to be explored is determined based on the capital map, and the industrial chain core and industry sequence of the target industry are determined based on the target industry.

[0118] Step S502: Taking each node on the capital graph as a root node, a full path search is performed in the capital graph to obtain the one-way capital flow relationship edges that occur at the root node within a preset number of capital transfers within a preset period, and obtain the industry path.

[0119] In step S502, the preset period is an observation period, which may be one year or half a year, and the present invention does not limit this.

[0120] The number of fund transfers refers to how many times the funds are transferred, that is, the predetermined number of lots.

[0121] The number of fund transfers may be 1-N lots or 2-N lots, which is not limited in the present invention.

[0122] In the process of implementing step S502, each node on the capital map is taken as the root node, a period is set, and a global full-path search is performed in the capital map to obtain the one-way capital flow relationship edge that occurs at the root node within the preset number of capital transfers within the set period, and obtain the industry path.

[0123] For example, Figure 4 As shown, with node B as the root node, the preset period is one year, and the preset number of fund transfers is 2 lots, a global full-path search is performed in the fund map to obtain the one-way fund flow relationship edges that occur within 2 lots of the root node within one year, and obtain the one-way fund flow relationship edges BD and DE, and obtain the industry path based on the one-way fund flow relationship edges.

[0124] Step S503: The minimum value of each actual capital flow on the industry path is used as the weight on the industry path, and the minimum edge weight on the industry path is obtained according to the weight on the industry path.

[0125] In step S503, fund flow = number of transactions * amount, which is Figure 4The attributes of the relationship edge.

[0126] In the process of specifically implementing step S503, each actual capital flow on the industry path is obtained, each actual capital flow is compared to obtain the minimum actual capital flow, that is, the minimum value of the actual capital flow, and the minimum value of each actual capital flow on the industry path is used as the weight on the industry path. According to the weight on the industry path, each actual capital flow on the industry path is divided by the minimum value of each actual capital flow to obtain the minimum edge weight on the industry path.

[0127] Step S504: taking the minimum edge weight on the industry path as the weight multiple of the industry sequence to obtain an industry sequence with industrial correlation.

[0128] In the specific implementation of step S504, the minimum edge weight on the obtained industry path is used as the weight multiple of the industry sequence to obtain an industry sequence with industrial correlation.

[0129] Step S505: Based on the preset support threshold and the association mining algorithm PrefixSpan, industry sequences that meet the support threshold in the capital map are mined from industry sequences with industrial associations to obtain frequent industry sequences.

[0130] In the specific implementation process of step S505, a support threshold is preset, and based on the set support threshold, the association mining algorithm PrefixSpan is used to mine industry sequences that meet the set support threshold in the capital map to obtain frequent industry sequences.

[0131] Step S506: Determine whether the frequent industry sequence meets the requirements for displaying the industrial chain map. If so, execute step S507; if not, execute step S508.

[0132] In the process of implementing step S506, it is determined whether the frequent industry sequence meets the display requirements of the industrial chain map. If so, it means that there is no need to mine the frequent industry sequence again, and step S507 is executed. If not, it means that there is a need to mine the frequent industry sequence again, and step S508 is executed.

[0133] Step S507: Output all frequent industry sequences that meet the requirements of industrial chain graph display to obtain an industry frequent sequence set.

[0134] In the process of specifically implementing step S507, it is determined that the obtained frequent industry sequence meets the display requirements of the industrial chain map, and the frequent industry sequences that meet the display requirements of the industrial chain map are obtained, and all frequent industry sequences that meet the display requirements of the industrial chain map are output to obtain an industry frequent sequence set.

[0135] Step S508: Based on the set support threshold and the associated mining algorithm PrefixSpan, industry sequences that meet the support threshold in the capital map are mined from industry sequences with industrial associations until the frequent industry sequences meet the requirements for displaying the industrial chain map.

[0136] In the process of implementing step S508, it is determined that the mined frequent industry sequence does not meet the display requirements of the industrial chain map, and a new support threshold is reset. Based on the set new support threshold, the association mining algorithm PrefixSpan is reused to mine the industry sequence in the capital map that meets the set new support threshold to obtain the frequent industry sequence.

[0137] If the frequent industry sequence obtained at this time meets the requirements of the industrial chain map display, proceed to the next step.

[0138] If the frequent industry sequence obtained at this time does not meet the display requirements of the industrial chain map, continue to execute the operation of re-based on the set support threshold and association mining algorithm PrefixSpan, and mine the industry sequence that meets the support threshold in the capital map from the industry sequence with industrial correlation, until the obtained frequent industry sequence meets the display requirements of the industrial chain map.

[0139] For example, the current frequent industry sequence is <d> <db> <dc> <dcb>, determine the frequent industry sequence <d> <db> <dc> <dcb>If it meets the needs of displaying the industrial chain map, the frequent industry sequence <d> <db> <dc> <dcb>Output and obtain the industry frequent sequence set.

[0140] For example, the current frequent industry sequence is <bc> <bd> <bdc>, determine the frequent industry sequence <bc> <bd> <bdc>If it does not meet the requirements of the industrial chain map display, continue to execute the operation of re-based on the set support threshold and the associated mining algorithm PrefixSpan, and mine the industry sequence that meets the support threshold in the capital map from the industry sequence with industrial correlation, until the obtained frequent industry sequence meets the requirements of the industrial chain map display.

[0141] An industrial chain identification method provided based on an embodiment of the present invention uses the capital map constructed as above and based on a set support threshold to utilize the association mining algorithm PrefixSpan to mine industry sequences that meet the support threshold in the capital map, thereby obtaining an industry frequent sequence set, in preparation for the subsequent industrial chain map, thereby being able to identify the degree of association between industries, improve the completeness and adequacy of identification, reduce identification time, and improve versatility.

[0142] Based on the industrial chain identification method provided by the above embodiment of the present invention, step S505 is executed to mine the industry sequence that meets the support threshold in the capital map from the industry sequence with industrial association based on the preset support threshold and the association mining algorithm PrefixSpan to obtain the frequent industry sequence. Figure 6 As shown, a flow chart of constructing a frequent industry sequence provided by an embodiment of the present invention mainly includes the following steps:

[0143] Step S601: according to the weight multiples of the industry sequence, the industry sequence with industry correlation is sorted using the association mining algorithm PrefixSpan, and the prefix sequence and suffix sequence in the industry sequence with industry correlation are determined.

[0144] In step S601 , each industry sequence with industry correlation includes multiple item sets, and each item set includes one or more elements.

[0145] In the process of specifically implementing step S601, the industry sequence with industry correlation is sorted according to the weight multiples of the industry sequence obtained above using the association mining algorithm PrefixSpan to obtain the sorting result, and the prefix sequence and suffix sequence in the industry sequence with industry correlation are determined according to the sorting result.

[0146] Step S602: Convert the suffix sequence into a prefix sequence based on a recursive algorithm until the industry sequence with industry association does not meet the support threshold, thereby obtaining a frequent industry sequence.

[0147] In the process of specifically implementing step S602, the suffix sequence is converted into a prefix sequence. During the conversion process, a first item set is extracted from the suffix sequence and added to the prefix sequence. The prefix sequence is scanned from left to right to determine the second item set corresponding to the elements in the first item set in the prefix sequence. The second item set is changed and the recursive processing is continued using a recursive algorithm according to the conversion rules in the conversion process until the industry sequence with industrial correlation does not meet the support threshold and a frequent industry sequence is obtained.

[0148] An industrial chain identification method provided based on an embodiment of the present invention determines the prefix sequence and the suffix sequence in the industry sequence with industrial correlation, and converts the suffix sequence into a prefix sequence to obtain a frequent industry sequence, thereby preparing for the subsequent industrial chain map, thereby being able to identify the degree of correlation between industries, improve the completeness and adequacy of identification, reduce identification time, and improve versatility.

[0149] Based on the industrial chain identification method provided by the above embodiment of the present invention, step S104 is performed to use the k industry sequences with the highest industrial correlation as the chain sequence to obtain and display the industrial chain map. Figure 7 As shown, a schematic diagram of a process for obtaining an industrial chain map provided by an embodiment of the present invention mainly includes the following steps:

[0150] Step S701: Obtain k industry sequences with the highest industry correlation in the industry frequent sequence set, and use the k industry sequences with the highest industry correlation as chain sequences.

[0151] In the specific implementation of step S701, the k industry sequences with the highest industry correlation in the above-obtained industry frequent sequence set are obtained, the k industry sequences with the highest industry correlation are obtained, and the k industry sequences with the highest industry correlation are used as chain sequences.

[0152] Step S702: Reverse sorting the frequent industry sequences according to the industry relevance of each frequent industry sequence in the frequent industry sequence set to obtain a reverse sorting result.

[0153] In the specific implementation of step S702, the frequent industry sequences to be displayed in the industrial chain map are selected, and the frequent industry sequences are reversed according to the industrial relevance of each frequent industry sequence in the frequent industry sequence set to obtain the reversed sorting result.

[0154] Step S703: Taking the chain sequence as the backbone of the industrial chain map, according to the reverse sorting results, the frequent industry sequences are added to the industrial chain map in sequence to obtain and display the industrial chain map.

[0155] In step S703, NetworkX can be used to display the industrial chain map.

[0156] In the specific implementation of step S703, the chain sequence is used as the backbone of the industrial chain map. According to the reverse sorting results, the frequent industry sequences with high industrial correlation are first added to the industrial chain map in sequence to form a long chain, and then the frequent industry sequences with low industrial correlation are added to the industrial chain map in sequence to form a short chain. It is determined whether the newly added short chain is included in the long chain. If so, the support threshold of the long chain is reduced, and the support threshold of the short chain is updated to obtain the industrial chain map and input it into NetworkX for display.

[0157] It should be noted that, for example, under the set support threshold, there are two industry sequences A and B.<A,B> It is frequent.<B,A> It is also frequent to set corresponding parameters in the construction function of displaying the industrial chain map to support the determination of upstream and downstream directions based on the size of capital flow.

[0158] An industrial chain identification method provided based on an embodiment of the present invention uses the k industry sequences with the highest industrial correlation as the chain sequence to avoid processing "ring-type" links in the industrial chain, thereby being able to identify the degree of correlation between industries, improve the completeness and adequacy of identification, reduce identification time, and improve versatility.

[0159] Corresponding to the industrial chain identification method shown in the above embodiment of the present invention, the embodiment of the present invention also provides an industrial chain identification system, such as Figure 8 As shown, the industrial chain identification system includes: a preprocessing module 81, a construction module 82, a mining module 83 and an acquisition module 84.

[0160] The preprocessing module 81 is used to obtain transaction flow data, preprocess the transaction flow data, obtain operating transaction flow data and key information for constructing a capital map, and the key information at least includes the enterprises participating in the transaction, the transaction direction, and the industry category to which the enterprises participating in the transaction belong.

[0161] The construction module 82 is used to construct a capital map based on operating transaction flow data and key information.

[0162] The mining module 83 is used to mine the industry sequence that meets the support threshold in the capital map based on the capital map using the association mining algorithm PrefixSpan to obtain a set of frequent industry sequences. The industry sequence has an industry correlation, and the industry correlation is the weight of the industry sequence in the industrial chain.

[0163] Obtain module 84, which is used to take the k industry sequences with the highest industrial correlation as chain sequences, obtain and display the industrial chain map, and k is a positive integer.

[0164] It should be noted that the specific principles and execution processes of each module or unit in the industrial chain identification system disclosed in the above embodiment of the present invention are the same as those of the industrial chain identification method implemented in the above embodiment of the present invention. Please refer to the corresponding parts of the industrial chain identification method disclosed in the above embodiment of the present invention, and will not be repeated here.

[0165] Based on an industrial chain identification system provided by an embodiment of the present invention, by acquiring transaction flow data, pre-processing the transaction flow data, obtaining operating transaction flow data and key information for constructing a capital map, the key information at least includes the enterprises involved in the transaction, the transaction direction and the industry category to which the enterprises involved in the transaction belong; constructing a capital map based on the operating transaction flow data and key information; based on the capital map, using the association mining algorithm PrefixSpan to mine the industry sequence that meets the support threshold in the capital map, obtaining an industry frequent sequence set, the industry sequence has an industry correlation, and the industry correlation is the weight of the industry sequence in the industrial chain; taking the k industry sequences with the highest industry correlation as chain sequences, obtaining an industrial chain map and displaying it. In this scheme, the capital map is constructed using transaction flow data, and the association mining algorithm PrefixSpan is used to mine the industry sequence that meets the support threshold in the capital map, and the k industry sequences with the highest industry correlation are taken as chain sequences, obtaining an industrial chain map and displaying it, so that the degree of correlation between industries can be identified, the integrity and sufficiency of identification can be improved, the identification time can be reduced, and the versatility can be improved.

[0166] Optional, based on the above Figure 8 The preprocessing module 81 shown further includes: a word segmentation processing unit, a determination unit, an acquisition unit and a obtaining unit.

[0167] The word segmentation processing unit is used to obtain transaction flow data, perform word segmentation processing on the transaction flow data, and obtain word segmentation results.

[0168] The determination unit is used to determine the words representing the usage in the word segmentation results based on the word segmentation results.

[0169] The acquisition unit is used to acquire words representing purposes in different types of transaction flow data to obtain a word set representing purposes.

[0170] The obtaining unit is used to obtain the key information of the operating transaction flow data and the construction of the capital map based on the words representing the purpose and the word sets representing the purpose in the word segmentation results.

[0171] An industrial chain identification system provided based on an embodiment of the present invention performs word segmentation on transaction flow data, and uses the obtained word segmentation results to distinguish the types of transaction flow data, and then determines the type of transaction flow data, so as to identify the degree of correlation between industries, improve the completeness and adequacy of identification, reduce identification time, and improve versatility.

[0172] Optional, based on the above Figure 8 The construction module 82 shown further includes: an acquisition unit and a construction unit.

[0173] The acquisition unit is used to obtain the enterprises involved in the transaction and the capital outflow relationship in the operating transaction flow data.

[0174] The construction unit is used to construct a directed capital graph with enterprises as nodes, capital outflow relationships as relationship edges, and transaction numbers and amounts as attributes of the relationship edges.

[0175] An industrial chain identification system provided based on an embodiment of the present invention forms a directed capital map by constructing nodes, relationship edges and attributes of relationship edges of the capital map, thereby providing a guarantee for the subsequent identification of the degree of correlation between industries, avoiding low accuracy when calculating the weight of a single transaction, improving the integrity and adequacy of the identification, reducing the identification time, and improving the versatility.

[0176] Optional, based on the above Figure 8 The mining module 83 shown in the figure further includes: a determination unit, a search and acquisition unit, a obtaining unit, a mining unit, a first mining processing unit and a second mining processing unit.

[0177] The determination unit is used to determine the industrial chain core and industry sequence based on the capital map.

[0178] The search and acquisition unit is used to take each node on the capital map as the root node, perform a full-path search in the capital map, obtain the one-way capital flow relationship edge that occurs at the root node within a preset number of capital transfers within a preset period, and obtain the industry path.

[0179] A unit is obtained, which is used to take the minimum value of each actual capital flow on the industry path as the weight on the industry path, and obtain the minimum edge weight on the industry path according to the weight on the industry path; the minimum edge weight on the industry path is used as the weight multiple of the industry sequence to obtain an industry sequence with industrial correlation.

[0180] The mining unit is used to mine industry sequences that meet the support threshold in the capital map in the industry sequence with industrial association based on a preset support threshold and the association mining algorithm PrefixSpan to obtain frequent industry sequences.

[0181] The first mining processing unit is used to output all frequent industry sequences that meet the requirements of industrial chain map display if the frequent industry sequences meet the requirements of industrial chain map display, and obtain an industry frequent sequence set.

[0182] The second mining processing unit is used to mine industry sequences that meet the support threshold in the capital map from industry sequences with industrial correlation if the frequent industry sequences do not meet the display requirements of the industrial chain map, based on the set support threshold and the associated mining algorithm PrefixSpan, until the frequent industry sequences meet the display requirements of the industrial chain map.

[0183] An industrial chain identification system provided based on an embodiment of the present invention uses the capital map constructed as above and based on a set support threshold to mine industry sequences that meet the support threshold in the capital map using the association mining algorithm PrefixSpan to obtain an industry frequent sequence set, thereby preparing for the subsequent industrial chain map, thereby being able to identify the degree of association between industries, improve the completeness and adequacy of identification, reduce identification time, and improve versatility.

[0184] Optional, based on the above Figure 8 The mining module 83 shown is specifically used for:

[0185] According to the weight multiples of the industry sequence, the association mining algorithm PrefixSpan is used to sort the industry sequences with industry correlation, and the prefix sequence and suffix sequence in the industry sequence with industry correlation are determined. Each industry sequence with industry correlation contains multiple item sets, and each item set contains one or more elements; based on the recursive algorithm, the suffix sequence is converted into a prefix sequence until the industry sequence with industry correlation does not meet the support threshold, and the frequent industry sequence is obtained.

[0186] An industrial chain identification system provided based on an embodiment of the present invention determines the prefix sequence and suffix sequence in an industry sequence with industrial correlation, and converts the suffix sequence into a prefix sequence to obtain a frequent industry sequence, thereby preparing for the subsequent industrial chain map, thereby being able to identify the degree of correlation between industries, improve the completeness and adequacy of identification, reduce identification time, and improve versatility.

[0187] Optional, based on the above Figure 8 The obtaining module 84 is shown, and the obtaining module 84 is specifically used for:

[0188] Get the k industry sequences with the highest industrial correlation in the industry frequent sequence set, and use the k industry sequences with the highest industrial correlation as the chain sequence; sort the frequent industry sequences in reverse order according to the industrial correlation of each frequent industry sequence in the frequent industry sequence set to obtain the reverse sorting results; take the chain sequence as the backbone of the industrial chain map, and according to the reverse sorting results, add the frequent industry sequences to the industrial chain map in turn to obtain and display the industrial chain map.

[0189] An industrial chain identification system provided based on an embodiment of the present invention uses the k industry sequences with the highest industrial correlation as the chain sequence to avoid the appearance of "ring-shaped" links in the industrial chain, thereby being able to identify the degree of correlation between industries, improve the completeness and adequacy of identification, reduce identification time, and improve versatility.

[0190] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can refer to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiment. The system and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without creative work.

[0191] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may 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 the present invention.

[0192] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.< / bdc> < / bd> < / bc> < / bdc> < / bd> < / bc> < / dcb> < / dc> < / db> < / d> < / dcb> < / dc> < / db> < / d> < / dcb> < / dc> < / db> < / d>

Claims

1. A method for identifying an industrial chain, characterized in that: The method comprises: Acquire transaction flow data, pre-process the transaction flow data, and obtain operating transaction flow data and key information for constructing a capital map, including: acquire transaction flow data, perform word segmentation processing on the transaction flow data, and obtain word segmentation results; based on the word segmentation results, determine the words representing the purpose in the word segmentation results; acquire the words representing the purpose in different types of the transaction flow data to obtain a word set representing the purpose; based on the words representing the purpose in the word segmentation results and the word set representing the purpose, obtain operating transaction flow data and key information for constructing a capital map, the key information at least including the enterprises participating in the transaction, the transaction direction, and the industry category to which the enterprises participating in the transaction belong; Constructing a capital map based on the operating transaction flow data and the key information; Based on the capital graph, the industry sequence that meets the support threshold in the capital graph is mined using the association mining algorithm PrefixSpan to obtain an industry frequent sequence set, wherein the industry sequence has an industry correlation, and the industry correlation is the weight of the industry sequence in the industrial chain; Take the k industry sequences with the highest industry correlation as the chain sequence, obtain the industrial chain map and display it, k is a positive integer; Wherein, based on the capital graph, the association mining algorithm PrefixSpan is used to mine the industry sequences that meet the support threshold in the capital graph to obtain the industry frequent sequence set, including: According to the capital map, determine the industrial chain core and industry sequence; Taking each node on the capital graph as a root node, a full path search is performed in the capital graph to obtain the one-way capital flow relationship edge occurring at the root node within a preset capital transfer number within a preset period, and obtain an industry path; The minimum value of each actual capital flow on the industry path is used as the weight on the industry path, and the minimum edge weight on the industry path is obtained according to the weight on the industry path; The minimum edge weight on the industry path is used as the weight multiple of the industry sequence to obtain an industry sequence with industrial correlation; Based on a preset support threshold and an association mining algorithm PrefixSpan, mining the industry sequence that meets the support threshold in the capital graph from the industry sequence with industry association to obtain a frequent industry sequence; If the frequent industry sequence meets the requirements of the industrial chain graph display, output all the frequent industry sequences that meet the requirements of the industrial chain graph display to obtain an industry frequent sequence set; If the frequent industry sequence does not meet the display requirements of the industrial chain map, the industry sequence that meets the support threshold in the capital map is mined from the industry sequence with industrial correlation based on the set support threshold and the associated mining algorithm PrefixSpan, until the frequent industry sequence meets the display requirements of the industrial chain map.

2. The method according to claim 1, characterized in that The constructing of a capital map based on the operating transaction flow data and the key information includes: Obtaining the enterprises involved in the transaction and the capital outflow relationship in the operating transaction flow data; A directed capital graph is constructed with the enterprise as a node, the capital outflow relationship as a relationship edge, and the number of transactions and the amount as attributes of the relationship edge.

3. The method according to claim 1, characterized in that Based on the preset support threshold and the association mining algorithm PrefixSpan, the industry sequence that meets the support threshold in the capital map is mined in the industry sequence with industrial association to obtain a frequent industry sequence, including: According to the weight multiples of the industry sequence, the industry sequence with industry correlation is sorted by using the association mining algorithm PrefixSpan, and the prefix sequence and the suffix sequence in the industry sequence with industry correlation are determined, each of the industry sequences with industry correlation contains multiple item sets, and each of the item sets contains one or more elements; The suffix sequence is converted into the prefix sequence based on a recursive algorithm until the industry sequence with industrial relevance does not meet the support threshold, thereby obtaining a frequent industry sequence.

4. The method according to claim 1, characterized in that The k industry sequences with the highest industry correlation are used as chain sequences to obtain and display the industrial chain map, including: Obtain k industry sequences with the highest industry correlation in the industry frequent sequence set, and use the k industry sequences with the highest industry correlation as chain sequences; According to the industrial relevance of each of the frequent industry sequences in the frequent industry sequence set, the frequent industry sequences are sorted in reverse order to obtain a reverse order result; Taking the chain sequence as the backbone of the industrial chain map, the frequent industry sequences are added to the industrial chain map in sequence according to the reverse sorting results to obtain and display the industrial chain map.

5. An industrial chain identification system, characterized in that: The system comprises: A preprocessing module is used to obtain transaction flow data, preprocess the transaction flow data, obtain operating transaction flow data and key information for constructing a capital map, wherein the key information at least includes the enterprises involved in the transaction, the transaction direction, and the industry category to which the enterprises involved in the transaction belong; A construction module, used to construct a capital map based on the operating transaction flow data and the key information; A mining module is used to mine industry sequences that meet a support threshold in the capital map using an association mining algorithm PrefixSpan based on the capital map to obtain a frequent industry sequence set, wherein the industry sequence has an industry association degree, and the industry association degree is a weight of the industry sequence in the industrial chain; The module is used to take the k industry sequences with the highest industry correlation as the chain sequence, obtain the industrial chain map and display it, where k is a positive integer; Wherein, the preprocessing module includes: A word segmentation processing unit, used to obtain transaction flow data, perform word segmentation processing on the transaction flow data, and obtain a word segmentation result; A determination unit, configured to determine, based on the word segmentation result, a word representing a purpose in the word segmentation result; An acquisition unit, used to acquire words representing purposes in different types of transaction flow data to obtain a word set representing purposes; An obtaining unit, used to obtain the key information of the business transaction flow data and the capital map construction based on the words representing the purpose and the word set representing the purpose in the word segmentation result; Wherein, the mining module includes: A determination unit, used to determine the industrial chain core and industry sequence according to the capital map; A search and acquisition unit, configured to take each node on the capital graph as a root node, perform a full path search in the capital graph, obtain the one-way capital flow relationship edge occurring at the root node within a preset capital transfer number within a preset period, and obtain an industry path; The obtaining unit is used to use the minimum value of each actual capital flow on the industry path as the weight on the industry path, and obtain the minimum edge weight on the industry path according to the weight on the industry path; use the minimum edge weight on the industry path as the weight multiple of the industry sequence to obtain the industry sequence with industry correlation; A mining unit, configured to mine industry sequences that meet the support threshold in the capital graph from the industry sequences with industry associations based on a preset support threshold and an association mining algorithm PrefixSpan, to obtain frequent industry sequences; The first mining processing unit is used for outputting all the frequent industry sequences that meet the requirements of the industrial chain map display if the frequent industry sequences meet the requirements of the industrial chain map display, so as to obtain an industry frequent sequence set; The second mining processing unit is used to mine the industry sequences that meet the support threshold in the capital map based on the set support threshold and the associated mining algorithm PrefixSpan if the frequent industry sequence does not meet the display requirements of the industrial chain map, until the frequent industry sequence meets the display requirements of the industrial chain map.

6. The system according to claim 5, characterized in that The building blocks include: An acquisition unit, used to acquire the enterprises involved in the transaction and the capital outflow relationship in the operating transaction flow data; A construction unit is used to construct a directed capital graph with the enterprise as a node, the capital outflow relationship as a relationship edge, and the number of transactions and the amount as attributes of the relationship edge.

Images