CN115345401A - Six-dimensional analysis method for finding enterprise financial risk - Google Patents
Six-dimensional analysis method for finding enterprise financial risk Download PDFInfo
- Publication number
- CN115345401A CN115345401A CN202110527091.4A CN202110527091A CN115345401A CN 115345401 A CN115345401 A CN 115345401A CN 202110527091 A CN202110527091 A CN 202110527091A CN 115345401 A CN115345401 A CN 115345401A
- Authority
- CN
- China
- Prior art keywords
- risk
- enterprise
- dimensional
- information
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/211—Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
Abstract
The invention discloses a six-dimensional analysis method for finding enterprise financial risks, which comprises the following steps: s1, acquiring basic information of a monitored enterprise; s2, acquiring related information of an enterprise in a data acquisition or interface mode; s3, scoring the five-dimensional risk; s4, service experts score service risks; and S5, scoring the six-dimensional risk. Has the advantages that: the invention can analyze the financial risk of the enterprise from six dimensions, is beneficial to the state to discover the financial risk of the enterprise as early as possible, and adopts different risk prevention and treatment measures according to different risk types.
Description
Technical Field
The invention relates to the field of enterprise financial risk supervision, in particular to a six-dimensional analysis method for finding enterprise financial risks.
Background
In a future period of time, the work of supervision informatization, science and technology, financial risk quantification and the like becomes the main task of financial supervision work, and is also one of important scenes for supervision science and technology falling to the ground.
Due to a series of problems of difference of financial state of each region, diversity of illegal financial activity means, asymmetric information among supervision authorities and unshared data resources, the manpower, technical and professional abilities of supervision departments in each region hardly meet the supervision requirements in a short time.
An effective solution to the problems in the related art has not been proposed yet.
Disclosure of Invention
Technical problem to be solved
Aiming at the defects of the prior art, the invention provides a six-dimensional analysis method for finding financial risks of enterprises, which has the advantages of pertinence prevention and treatment, so that the aim of effectively resolving the financial risks is fulfilled, and the problem that the requirements of supervision are difficult to meet in a short time by the manpower, technical and professional abilities of supervision departments is solved.
(II) technical scheme
In order to realize the purpose of the targeted prevention and treatment and effectively solving the financial risk, the invention adopts the following specific technical scheme:
a six-dimensional analysis method for discovering enterprise financial risks comprises the following steps:
s1, acquiring basic information of a monitored enterprise;
s2, acquiring related information of an enterprise in a data acquisition or interface mode;
s3, scoring the five-dimensional risk;
s4, service experts score service risks;
and S5, scoring the six-dimensional risk.
Further, the step S1 of acquiring the basic information of the monitored enterprise includes the following steps:
s11, constructing a monitored enterprise library;
and S12, importing the basic information of the monitored enterprise into the system through an excel template.
Further, in step S12, basic information of the monitored enterprise is imported into the system through an excel template, where the basic information of the enterprise includes: enterprise name, enterprise abbreviation, affiliated industry, industrial and commercial registration place and actual operation place.
Further, the step S2 of acquiring the related information of the enterprise through a data acquisition or interface mode includes the following steps:
s21, acquiring relevant industrial and commercial data, recruitment data, enterprise APP information, website access amount, ICP filing information and three-party platform information of an enterprise;
and S22, collecting public opinion information through a data collection technology.
Further, in the step S22, public opinion information is collected through a data collection technology, the public opinion information includes news, forum, APP news, weChat, post bar, blog, and information published by a microblog platform.
Further, the step S3 of five-dimensional risk scoring includes the following steps:
s31, judging scale risk;
s32, public opinion risk judgment;
s33, judging associated risks;
s34, operation risk judgment;
s35, judging judicial risks;
and S36, carrying out five-dimensional risk scoring aiming at the steps S31-S35.
Further, the step S31 of scale risk determination includes the steps of:
s311, acquiring the transaction scale, the number of investors, the number of borrowers, the APP download amount of the enterprise, the access amount of an enterprise website and the personnel scale of the enterprise by monitoring the enterprise;
s312, establishing a rule-based scale risk judgment model for the scale related data, and setting different threshold intervals for each scale data;
and S313, judging by using the scale data of the monitored enterprise and a scale risk judgment model, and finding the risk of the overlarge enterprise scale.
The step S32 and the public opinion risk judgment comprise the following steps:
s321, extracting keywords in the text description through a textrank algorithm, and judging whether the text belongs to the company entity name to be judged;
s322, constructing a training set, a test set and a verification set for emotion positive and negative judgment;
s323, building an RNN neural model by using a tenserflow neural network framework;
s324, constructing judgment of positive and negative emotions of the public sentiment through a bidirectional RNN neural network model, and obtaining positive and negative information of the public sentiment article;
and S325, if the public sentiment is judged to be negative public sentiment through sentiment analysis, the enterprise is judged to have public sentiment risk.
The step S33 of determining the associated risk includes the steps of:
s331, firstly establishing a problem enterprise library, wherein data sources are public opinion, a three-party platform and industrial and commercial data, and if the states of the enterprise are found to be sales loss, case setting, loss of connection and logout, adding the enterprise into the problem enterprise library;
s332, associating the monitoring enterprise with the problem enterprise library, wherein the association path comprises: stock right association, high management association, legal person association, division association, telephone association and mailbox association;
and S333, if the enterprise with the problem can be associated, judging that the enterprise has associated risks.
The step S34 and the operation risk determination include the following steps:
s341, discovering and monitoring the number of branches or branch companies set by the enterprise through the industrial and commercial data, and if the number exceeds a threshold value, judging that excessive risks of the branches exist;
s342, counting the number of the recruitment positions comprising a financial advisor, an investment advisor and a financial planner based on the recruitment data, and judging that financial risks exist if the number of the recruitment positions exceeds a threshold value;
s343, searching for the industrial and commercial data, and judging that the industrial and commercial data abnormal risk exists if the registered place is inconsistent with the operating place, the annual newspaper is not disclosed, the industrial and commercial information is frequently changed, and the tax payment information is abnormal;
s344, ICP record information of the enterprise is searched through the station leader tool data, and therefore ICP non-record, ICP record number false and a plurality of domain name risks are found and judged;
and S345, detecting the enterprise website or the app, judging the enterprise website or the app to be normal if the enterprise website or the app can be requested normally, and otherwise judging that the risk that the website cannot be accessed and the app cannot be accessed exists.
The step S35 and the judicial risk assessment include the steps of:
and searching information related to judicial affairs, tax and administrative punishment of the monitored enterprise through the industrial and commercial data, and if the information exists, judging whether the enterprise has risks of complaints, distrusted persons, distrusted executives, abnormal operation, abnormal tax payment, case setting, serious law violation and administrative punishment according to the condition.
The step S36 of five-dimensional risk scoring for the steps S31 to S35 includes the steps of:
s361, establishing a five-dimensional risk scoring model aiming at scale risk, public opinion risk, associated risk, operation risk and judicial risk, and setting different risk scores for different risk items;
and S362, setting different weights for the five dimensions, taking the risk data analyzed in the previous step as input parameters of a five-dimensional risk scoring model, and automatically calculating the five-dimensional risk score of the enterprise by the system.
Further, the step S4 of service risk scoring by the service expert includes the following steps:
aiming at enterprises with five-dimensional risk scores of more than 20 points, the system automatically generates a business risk auditing task through a background batch running program and assigns the business risk auditing task to related business risk auditors, and the auditors mark business risks of the enterprises according to business experiences of the auditors after receiving the task.
Further, the step S5 of six-dimensional risk scoring includes the following steps:
s51, firstly, establishing a six-dimensional risk scoring model aiming at scale risk, public opinion risk, associated risk, operation risk, judicial risk and business risk, and setting different risk scores for different risk items;
s52, setting different weights for the six dimensions, taking the risk data analyzed in the previous step as input parameters of a six-dimensional risk scoring model, and automatically calculating the six-dimensional risk score of the enterprise by the system.
In addition, according to another aspect of the present invention, there is provided a six-dimensional analysis system for finding financial risks of an enterprise, the system comprising an enterprise basic information acquisition module, a data acquisition module, a five-dimensional risk judgment/scoring module, a business risk scoring module, and a six-dimensional risk scoring module;
wherein the enterprise basic information module: for building a monitored enterprise repository;
a data acquisition module: the system is used for acquiring relevant industrial and commercial data, recruitment data, enterprise APP information, website access amount, ICP filing information and three-party platform information of an enterprise, and acquiring public opinion information through a data acquisition technology;
a five-dimensional risk judgment/scoring module: the method is used for scale risk judgment, public opinion risk judgment, association risk judgment, operation risk judgment, judicial risk judgment and five-dimensional risk scoring;
a business risk scoring module: and marking the business risk of the enterprise with the five-dimensional risk score larger than 20 points.
A six-dimensional risk scoring module: for a six-dimensional risk score for scale risk, public opinion risk, association risk, operational risk, judicial risk, business risk.
(III) advantageous effects
Compared with the prior art, the invention provides a six-dimensional analysis method for discovering enterprise financial risks, which has the following beneficial effects: the invention is the best way to realize supervision informatization and science and technology as soon as possible and put into application by means of the strength of third-party science and technology companies. The six-dimensional analysis method for finding the financial risk of the enterprise is provided by combining data acquisition, natural language processing and machine learning technologies, a set of more comprehensive analysis method is provided in the aspects of monitoring and early warning of illegal funding or financial risk, a client is helped to find the financial risk of the enterprise more easily, and targeted prevention and treatment are achieved, so that the financial risk is effectively solved.
The invention can analyze the financial risk of the enterprise from six dimensions, is beneficial to the state to discover the financial risk of the enterprise as early as possible, and adopts different risk prevention and treatment measures according to different risk types.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
FIG. 1 is a flow diagram of a six-dimensional analysis method for discovering enterprise financial risk in accordance with an embodiment of the present invention;
FIG. 2 is a flow chart of a large-scale risk determination in a six-dimensional analysis method for discovering enterprise financial risks according to an embodiment of the present invention;
FIG. 3 is a flow chart of public opinion risk determination in a six-dimensional analysis method for discovering enterprise financial risk according to an embodiment of the present invention;
FIG. 4 is a flow chart of an associative risk determination in a six-dimensional analysis method for discovering enterprise financial risk according to an embodiment of the present invention;
FIG. 5 is a system block diagram of a six-dimensional analytics system for discovering enterprise financial risks in accordance with an embodiment of the present invention.
Detailed Description
For further explanation of the various embodiments, the drawings which form a part of the disclosure and which are incorporated in and constitute a part of this specification, illustrate embodiments and, together with the description, serve to explain the principles of operation of the embodiments, and to enable one skilled in the art to understand the embodiments and advantages of the disclosure for reference and without scale, wherein elements are not shown in the drawings and like reference numerals are used to refer to like elements generally.
Before describing a six-dimensional analysis method for finding financial risk of an enterprise provided by the invention, the terms mentioned in the invention are briefly explained:
financial risk
Financial risk refers to risk caused by monitoring illegal/non-compliant business activities, transaction activities, business behaviors, etc. of an enterprise subject in a specific time and network space, that is, risk caused by monitoring illegal/active financial activities of the enterprise, which is collectively referred to as financial risk.
Public opinion risk
Public opinion risk refers to the expression of media, news websites and netizens (usually investors or borrowers) on monitoring business activities or business behaviors of enterprises/main bodies, and such expressions usually relate to the problems of monitoring business, products, compliance and cash of enterprises and are the main complaint reporting channels of investors.
Associating risks
The associated risk refers to that in the enterprise picture of the monitored enterprise/main body, the characteristics of the enterprise group are obvious, and the associated network is huge; there may also be risks due to associative relationships between transaction principals, with high-risk businesses, or associated transactions.
Operational risk
The operation risk refers to the risk that monitoring enterprises/bodies have huge scales in the range of business activities, branch organizations, related publicity channels, the number of customers and the like, and a group event is caused once a problem occurs.
Judicial risks
The judicial risks mainly refer to risks generated by monitoring subjects due to problems of judicial affairs, tax affairs, administrative punishment and the like.
Risk of scale
The scale risk refers to a series of risk indexes related to people-related problems due to transaction scale, personnel scale and the like of the monitoring subject, the indexes reveal the business scale of the monitoring subject, and once a problem occurs, the monitoring subject relates to a plurality of people, and mass events are easily caused.
Business risk
The business risk refers to a comprehensive risk prompt given by experts engaged in the field of enterprise financial risk monitoring for a long time through personal experience from aspects of national policies, enterprise-related data, industry development conditions, financial activities engaged in by enterprises and the like, and is generally called business risk.
According to an embodiment of the invention, a six-dimensional analysis method for discovering enterprise financial risk is provided.
Referring to the drawings and the detailed description, the invention will be further described, as shown in fig. 1, a six-dimensional analysis method for discovering enterprise financial risk according to an embodiment of the invention includes the following steps:
s1, acquiring basic information of a monitored enterprise;
this module is mainly to collect the enterprise list that will monitor, constructs the enterprise storehouse that is monitored, imports the basic information of enterprise into the system through excel template, and the basic information of enterprise contains: enterprise name, enterprise abbreviation, affiliated industry, industrial and commercial registration place, actual business place and the like.
S2, acquiring related information of an enterprise in a data acquisition or interface mode;
firstly, a data acquisition system is constructed, the following information is acquired by the acquisition system, and a database is stored.
Public opinion information: news, forums, APP news, weChat, bar, blog, microblog, etc.;
three-party platform information: such as the world Wide Web loan, the family of the world Wide Web loan, the society of Chinese securities investment fund industry, etc.;
APP information: APP malls such as IOS, huashi, millet, 360, tengxue and the like;
recruitment information: recruitment of intelligent associations, 51 jobs, hook pulling, boss direct hiring and the like;
other information: official website and station keeper of enterprise
And inquiring the industrial and commercial data of the enterprise in an interface mode and storing a database.
Step S3, five-dimensional risk scoring (scale risk, judicial risk, associated risk, operation risk and public opinion risk);
as shown in fig. 2, S31, scale risk determination;
the S31 scale risk judgment specifically comprises the following steps:
step S311, acquiring the transaction scale, the number of investors and the number of borrowers of the enterprise by monitoring the information leaked by the official website of the enterprise or the information of the three-party platform;
obtaining APP download amount of a monitored enterprise through APP malls such as IOS, huacheng, millet, 360, tengxue and the like;
acquiring the access amount of the monitored enterprise website through a station leader;
and obtaining the personnel scale of the monitored enterprise through data such as intelligent joining recruitment, 51 jobs, hook pulling, boss direct recruitment and the like.
In step S312, a scale risk determination model based on the rule is established for the scale-related data, and a different threshold is set for each scale data.
Step S313, using the scale data of the monitored enterprise, using the scale risk determination model to perform determination, and finding the risk of the enterprise with an excessively large scale.
As shown in fig. 3, S32, public opinion risk determination;
the S32 public opinion risk judgment specifically comprises the following steps:
s321, extracting keywords in the text description through a textrank algorithm, and judging whether the text belongs to the company entity name to be judged;
s322, constructing a training set, a test set and a verification set for emotion positive and negative judgment;
s323, building an RNN neural model by using a tenserflow neural network framework;
s324, constructing judgment of positive and negative emotions of the public sentiment through a bidirectional RNN neural network model, and obtaining positive and negative information of the public sentiment article;
and S325, if the public sentiment is judged to be negative public sentiment through sentiment analysis, the enterprise is judged to have public sentiment risk.
Wherein the principle of the Textrank algorithm is as follows:
the core formula of the TextRank algorithm is as follows, wherein the core formula is used for representing that the edge connection between two nodes has different importance degrees:
as shown in fig. 4, S33, association risk determination;
the S33 associated risk determination specifically includes the following steps:
s331, firstly establishing a problem enterprise library, wherein data sources are public opinion, a three-party platform and industrial and commercial data, and if the states of the enterprise are found to be sales loss, case setting, loss of connection and logout, adding the enterprise into the problem enterprise library;
s332, associating the monitoring enterprise with the problem enterprise library, wherein the association path comprises: stock right association, high management association, legal person association, division association, telephone association and mailbox association;
s333, if the enterprise with the problems can be associated, judging that the enterprise has the associated risk.
S34, operation risk judgment;
the S34 operation risk determination specifically includes the following steps:
s341, discovering and monitoring the number of branches or branch companies set by the enterprise through the industrial and commercial data, and if the number exceeds a threshold value, judging that excessive risks of the branches exist;
s342, counting the number of the recruitment positions comprising a financial advisor, an investment advisor and a financial planner based on the recruitment data, and judging that financial risks exist if the number of the recruitment positions exceeds a threshold value;
s343, searching for industrial and commercial data, and if the registration place is inconsistent with the operation place, the annual report is not disclosed, the industrial and commercial information is frequently changed, and the tax payment information is abnormal, judging that the industrial and commercial data abnormal risk exists;
s344, ICP record information of the enterprise is searched through the station leader tool data, and therefore ICP non-record, ICP record number false and a plurality of domain name risks are found and judged;
and S345, detecting the enterprise website or the app, judging the enterprise website or the app to be normal if the enterprise website or the app can be requested normally, and otherwise judging that the risk that the website cannot be accessed and the app cannot be accessed exists.
S35, judging judicial risks;
the S35 judicial risk assessment specifically comprises the following steps:
and searching information related to judicial affairs, tax, administrative punishment and the like of the monitored enterprise through the industrial and commercial data, and if the information exists, judging whether the enterprise has risks of complaints, distrusted persons, distrusted executives, abnormal operation, abnormal tax payment, case setting, serious illegal activities and administrative punishment according to the conditions.
S36, five-dimensional risk scoring is carried out aiming at the steps S31-S35
The step S36 of five-dimensional risk scoring for the steps S31 to S35 includes the steps of:
s361, establishing a five-dimensional risk scoring model aiming at scale risk, public opinion risk, associated risk, operation risk and judicial risk, and setting different risk scores for different risk items;
and S362, setting different weights for the five dimensions, taking the risk data analyzed in the previous step as input parameters of a five-dimensional risk scoring model, and automatically calculating the five-dimensional risk score of the enterprise by the system.
Step S4, business risk is carried out by business experts
Aiming at enterprises with five-dimensional risk scores of more than 20 points, the system automatically generates a business risk auditing task through a background batch running program and assigns the task to related business risk auditors, and the auditors mark business risks of the enterprises according to business experiences of the auditors after receiving the task.
Step S5 six-dimensional risk scoring
S51, firstly, establishing a six-dimensional risk scoring model aiming at scale risk, public opinion risk, associated risk, operation risk, judicial risk and business risk, and setting different risk scores for different risk items;
s52, setting different weights for the six dimensions, taking the risk data analyzed in the previous step as input parameters of a six-dimensional risk scoring model, and automatically calculating the six-dimensional risk score of the enterprise by the system.
In addition, in one implementation, as shown in fig. 5, a six-dimensional analysis system for finding financial risk of an enterprise is provided, which includes an enterprise basic information acquisition module, a data acquisition module, a five-dimensional risk judgment/scoring module, a business risk scoring module, and a six-dimensional risk scoring module;
wherein the enterprise basic information module: for constructing a monitored enterprise repository;
a data acquisition module: the system is used for acquiring relevant industrial and commercial data, recruitment data, enterprise APP information, website access amount, ICP filing information and three-party platform information of an enterprise, and acquiring public opinion information through a data acquisition technology;
a five-dimensional risk judgment/scoring module: the method is used for scale risk judgment, public opinion risk judgment, association risk judgment, operation risk judgment, judicial risk judgment and five-dimensional risk scoring;
a business risk scoring module: and marking the business risk of the enterprise with the five-dimensional risk score larger than 20 points.
A six-dimensional risk scoring module: for a six-dimensional risk score for scale risk, public opinion risk, association risk, operational risk, judicial risk, business risk.
In specific implementation, for convenience of understanding the above technical solution of the present invention, specific steps for extracting keywords and keyword groups by using the TextRank algorithm are given as follows:
step S3231, the given text is divided according to the whole sentence, i.e.
T＝[S 1 ，S 2 ，...，S m ]；
Step S3232, for each sentence S i ∈TPerforming word segmentation and part-of-speech tagging, then eliminating stop words, and only keeping words with specified part-of-speech, such as noun, verb, adjective, and the like, namely S i ＝[t i，1 ，t i，2 ，...，t i，n ]Wherein t is i，j A word reserved in the sentence i;
step S3233, constructing a word graph G = (V, E), where V is a node set and is composed of the words generated in the above steps, and then constructing an edge between any two nodes by using a co-occurrence relationship: edges exist between two nodes, and K represents the size of a window only when the corresponding words co-occur in the window with the length of K, namely the maximum number of co-occurring K words is 2 in general K;
step S3234, according to the formula, the weight of each node is calculated in an iterative manner until convergence;
step S3235, carrying out reverse sequencing on the weights of the nodes, and obtaining the most important t words as top-t key words;
and step S3236, marking the obtained top-t key words in the original text, and extracting the key words as key words if adjacent word groups are formed between the top-t key words and the original text.
When extracting key sentences from given texts, respectively regarding each sentence in the texts as a node, if two sentences have similarity, considering that an undirected weighted edge exists between the nodes corresponding to the two sentences, and the formula for measuring the similarity between the sentences is as follows:
S i 、S j : two sentences
w k : words in sentences
The numerator part means the number of the same word appearing in two sentences at the same time, and the denominator is the sum of the logarithms of the numbers of the words in the sentences, so that the design can inhibit the advantage of a longer sentence in similarity calculation.
And circularly calculating the similarity between any two nodes according to the similarity calculation formula, setting a threshold to remove edge connection with lower similarity between the two nodes, constructing a node connection graph, then iteratively calculating the TextRank value of each node, and selecting sentences corresponding to the nodes with the highest TextRank values as key sentences after sorting.
And searching for the name and abbreviation of a company or a company platform which needs risk judgment for the obtained top-t keyword, and if the name and abbreviation exist, judging that the main entity of the article is the company.
In summary, with the above technical solutions of the present invention, the present invention enables the monitoring informatization and the science and technology to be realized as soon as possible, and is put into an optimal way of application. The six-dimensional analysis method for finding the financial risks of the enterprises is provided by combining the data acquisition technology, the natural language processing technology and the machine learning technology, a set of more comprehensive analysis method is provided in the aspects of financial risk monitoring and early warning, the clients are helped to find the financial risks of the enterprises more easily, and targeted prevention and treatment are achieved, so that the financial risks are effectively solved. The invention can analyze the financial risk of the enterprise from six dimensions, is beneficial to the state to discover the financial risk of the enterprise as early as possible, and adopts different risk prevention and treatment measures according to different risk types.
In the present invention, unless otherwise expressly specified or limited, the terms "mounted," "disposed," "connected," "secured," "screwed" and the like are to be construed broadly, e.g., as meaning fixedly connected, detachably connected, or integrally formed; can be mechanically or electrically connected; the terms may be directly connected or indirectly connected through an intermediate, and may be communication between two elements or interaction relationship between two elements, unless otherwise specifically limited, and the specific meaning of the terms in the present invention will be understood by those skilled in the art according to specific situations.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (10)
1. A six-dimensional analysis method for discovering enterprise financial risk is characterized by comprising the following steps:
s1, acquiring basic information of a monitored enterprise;
s2, acquiring related information of an enterprise in a data acquisition or interface mode;
s3, scoring the five-dimensional risk;
s4, service experts score service risks;
and S5, scoring the six-dimensional risk.
2. The six-dimensional analysis method for discovering financial risk of enterprise according to claim 1, wherein said step S1 of obtaining basic information of monitored enterprise comprises the following steps:
s11, constructing a monitored enterprise library;
and S12, importing the basic information of the monitored enterprise into the system through an excel template.
3. The six-dimensional analysis method for discovering enterprise financial risk according to claim 2, wherein the step S12 imports the basic information of the monitored enterprise into the system through an excel template, wherein the basic information of the enterprise comprises: enterprise name, enterprise abbreviation, industry to which the company belongs, industrial and commercial registration place and actual business place.
4. The six-dimensional analysis method for discovering enterprise financial risk according to claim 1, wherein the step S2 of obtaining the related information of the enterprise through data collection or interface comprises the following steps:
s21, acquiring relevant industrial and commercial data, recruitment data, enterprise APP information, website access amount, ICP filing information and three-party platform information of an enterprise;
and S22, collecting public opinion information through a data collection technology.
5. The six-dimensional analysis method for finding financial risks of enterprises according to claim 4, wherein in the step S22, public opinion information is collected through a data collection technology, and the public opinion information includes information published by news, forum, APP news, weChat, post, blog and microblog platform.
6. The six-dimensional analysis method for finding financial risks of enterprises according to claim 1, wherein the step S3 of five-dimensional risk scoring comprises the following steps:
s31, scale risk judgment;
s32, public opinion risk judgment;
s33, judging associated risks;
s34, operation risk judgment;
s35, judging judicial risks;
and S36, carrying out five-dimensional risk scoring aiming at the steps S31-S35.
7. The six-dimensional analysis method for discovering enterprise financial risk according to claim 6, wherein the step S31 of determining the risk scale comprises the steps of:
s311, acquiring the transaction scale, the number of investors, the number of borrowers, the APP download amount of the enterprise, the access amount of an enterprise website and the personnel scale of the enterprise by monitoring the enterprise;
s312, establishing a rule-based scale risk judgment model for the scale related data, and setting different threshold intervals for each scale data;
and S313, judging by using the scale data of the monitored enterprise and a scale risk judgment model, and finding the risk of the overlarge enterprise scale.
The step S32 and the public opinion risk judgment comprise the following steps:
s321, extracting keywords in the text description through a textrank algorithm, and judging whether the text belongs to the company entity name to be judged;
s322, constructing a training set, a test set and a verification set for emotion positive and negative judgment;
s323, building an RNN neural model by using a tenserflow neural network framework;
s324, the judgment of positive and negative emotions of the public opinion is constructed through a bidirectional RNN neural network model, and therefore the positive and negative information of the public opinion article is obtained;
and S325, if the public sentiment is judged to be negative public sentiment through sentiment analysis, the enterprise is judged to have public sentiment risk.
The step S33 of determining the associated risk includes the steps of:
s331, firstly establishing a problem enterprise library, wherein data sources are public opinion, a three-party platform and industrial and commercial data, and if the states of the enterprise are found to be sales loss, case setting, loss of connection and logout, adding the enterprise into the problem enterprise library;
s332, associating the monitoring enterprise with the problem enterprise library, wherein the association path comprises: stock right association, high management association, legal person association, division association, telephone association and mailbox association;
s333, if the enterprise with the problems can be associated, judging that the enterprise has the associated risk.
The step S34 and the operation risk determination include the steps of:
s341, discovering the number of branches or branch companies established by the monitoring enterprise through the industrial and commercial data, and if the number exceeds a threshold value, judging that excessive risks of the branches exist;
s342, counting the number of the recruitment positions comprising a financial advisor, an investment advisor and a financial planner based on the recruitment data, and judging that financial risks exist if the number of the recruitment positions exceeds a threshold value;
s343, searching for the industrial and commercial data, and judging that the industrial and commercial data abnormal risk exists if the registered place is inconsistent with the operating place, the annual newspaper is not disclosed, the industrial and commercial information is frequently changed, and the tax payment information is abnormal;
s344, ICP record information of the enterprise is searched through the station leader tool data, and therefore ICP non-record, ICP record number false and a plurality of domain name risks are found and judged;
and S345, detecting the enterprise website or the app, judging the enterprise website or the app to be normal if the enterprise website or the app can be requested normally, and otherwise judging that the risk that the website cannot be accessed and the app cannot be accessed exists.
The step S35 and the judicial risk assessment include the steps of:
and searching information related to judicial affairs, tax and administrative punishment of the monitored enterprise through the industrial and commercial data, and if the information exists, judging whether the enterprise has risks of complaints, distrusted persons, distrusted executives, abnormal operation, abnormal tax payment, case setting, serious law violation and administrative punishment according to the condition.
The step S36 of five-dimensional risk scoring for the steps S31 to S35 includes the steps of:
s361, establishing a five-dimensional risk scoring model aiming at scale risk, public opinion risk, associated risk, operation risk and judicial risk, and setting different risk scores for different risk items;
and S362, setting different weights for the five dimensions, taking the risk data analyzed in the previous step as input parameters of a five-dimensional risk scoring model, and automatically calculating the five-dimensional risk score of the enterprise by the system.
8. The six-dimensional analysis method for finding enterprise financial risk according to claim 1, wherein the step S4 of the business experts performing business risk scoring comprises the following steps:
aiming at enterprises with five-dimensional risk scores of more than 20 points, the system automatically generates a business risk auditing task through a background batch running program and assigns the task to related business risk auditors, and the auditors mark business risks of the enterprises according to business experiences of the auditors after receiving the task.
9. The six-dimensional analysis method for discovering enterprise financial risk according to claim 1, wherein the step S5 of six-dimensional risk scoring comprises the steps of:
s51, firstly, establishing a six-dimensional risk scoring model aiming at scale risk, public opinion risk, associated risk, operation risk, judicial risk and business risk, and setting different risk scores for different risk items;
s52, setting different weights for the six dimensions, taking the risk data analyzed in the previous step as input parameters of a six-dimensional risk scoring model, and automatically calculating the six-dimensional risk score of the enterprise by the system.
10. A six-dimensional analysis system for finding enterprise financial risk, which is used for implementing the six-dimensional analysis method for finding enterprise financial risk of claims 1-9, and is characterized in that the system comprises an enterprise basic information acquisition module, a data acquisition module, a five-dimensional risk judgment/scoring module, a business risk scoring module and a six-dimensional risk scoring module;
wherein the enterprise basic information module: for constructing a monitored enterprise repository;
a data acquisition module: the system is used for acquiring relevant industrial and commercial data, recruitment data, enterprise APP information, website access amount, ICP filing information and three-party platform information of an enterprise, and acquiring public opinion information through a data acquisition technology;
a five-dimensional risk judgment/scoring module: the method is used for scale risk judgment, public opinion risk judgment, association risk judgment, operation risk judgment, judicial risk judgment and five-dimensional risk scoring;
a business risk scoring module: and marking the business risk of the enterprise with the five-dimensional risk score larger than 20 points.
A six-dimensional risk scoring module: for a six-dimensional risk score for scale risk, public opinion risk, association risk, operational risk, judicial risk, business risk.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110527091.4A CN115345401A (en) | 2021-05-14 | 2021-05-14 | Six-dimensional analysis method for finding enterprise financial risk |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110527091.4A CN115345401A (en) | 2021-05-14 | 2021-05-14 | Six-dimensional analysis method for finding enterprise financial risk |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115345401A true CN115345401A (en) | 2022-11-15 |
Family
ID=83947374
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110527091.4A Pending CN115345401A (en) | 2021-05-14 | 2021-05-14 | Six-dimensional analysis method for finding enterprise financial risk |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115345401A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115907568A (en) * | 2023-02-27 | 2023-04-04 | 北京金信网银金融信息服务有限公司 | Illegal financial activity monitoring method and system based on smoking index |
-
2021
- 2021-05-14 CN CN202110527091.4A patent/CN115345401A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115907568A (en) * | 2023-02-27 | 2023-04-04 | 北京金信网银金融信息服务有限公司 | Illegal financial activity monitoring method and system based on smoking index |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11574204B2 (en) | Integrity evaluation of unstructured processes using artificial intelligence (AI) techniques | |
| CN110704572B (en) | Suspected illegal fundraising risk early warning method, device, equipment and storage medium | |
| West et al. | Author‐level Eigenfactor metrics: Evaluating the influence of authors, institutions, and countries within the social science research network community | |
| Clarkson et al. | A textual analysis of US corporate social responsibility reports | |
| Saumya et al. | Detection of spam reviews: a sentiment analysis approach | |
| Han et al. | Artificial intelligence for anti-money laundering: a review and extension | |
| US10678798B2 (en) | Method and system for scoring credibility of information sources | |
| Holton | Identifying disgruntled employee systems fraud risk through text mining: A simple solution for a multi-billion dollar problem | |
| CN110796470A (en) | Market subject supervision and service oriented data analysis system | |
| KR20170004165A (en) | Device and method for analyzing corporate reputation by data mining of news, recording medium for performing the method | |
| CN104137128A (en) | Methods and systems for generating corporate green score using social media sourced data and sentiment analysis | |
| Fariss et al. | Human rights texts: Converting human rights primary source documents into data | |
| CN110880075A (en) | Employee departure tendency detection method | |
| CN110633316A (en) | Multi-scene fusion double-random market supervision method | |
| CN109492097B (en) | Enterprise news data risk classification method | |
| Fan et al. | Assessing topic model relevance: Evaluation and informative priors | |
| CN114860882A (en) | Fair competition review auxiliary method based on text classification model | |
| CN114303140A (en) | Analysis of intellectual property data related to products and services | |
| CN115345401A (en) | Six-dimensional analysis method for finding enterprise financial risk | |
| CN110782163A (en) | Enterprise data processing method and device | |
| CN115952216A (en) | Aging insurance data mining method and device, storage medium and electronic equipment | |
| Thi et al. | A novel solution for anti-money laundering system | |
| CN110766091B (en) | Method and system for identifying trepanning loan group partner | |
| Lee et al. | Novel methods for resolving false positives during the detection of fraudulent activities on stock market financial discussion boards | |
| Jabbari et al. | Towards a knowledge base of financial relations: Overview and project description |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |