CN110704803A - Target object evaluation value calculation method and device, storage medium and electronic device - Google Patents
Target object evaluation value calculation method and device, storage medium and electronic device Download PDFInfo
- Publication number
- CN110704803A CN110704803A CN201910939902.4A CN201910939902A CN110704803A CN 110704803 A CN110704803 A CN 110704803A CN 201910939902 A CN201910939902 A CN 201910939902A CN 110704803 A CN110704803 A CN 110704803A
- Authority
- CN
- China
- Prior art keywords
- view data
- target object
- data matrix
- evaluation value
- learner
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- 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
- 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
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Human Resources & Organizations (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Evolutionary Biology (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Operations Research (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Software Systems (AREA)
- Marketing (AREA)
- Evolutionary Computation (AREA)
- Probability & Statistics with Applications (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention relates to a method and a device for calculating an evaluation value of a target object, a storage medium and electronic equipment, which relate to the technical field of big data processing, and the method comprises the following steps: acquiring data to be processed corresponding to a target object, and obtaining a plurality of view data matrixes according to the data to be processed; inputting each view data matrix into a learner corresponding to each view data matrix respectively to obtain a weight value corresponding to each view data matrix; and obtaining the evaluation value of the target object according to the probability value corresponding to each view data matrix. The embodiment of the invention improves the accuracy of the evaluation value of the target object.
Description
Technical Field
The embodiment of the invention relates to the technical field of big data processing, in particular to a method for calculating an evaluation value of a target object, a device for calculating the evaluation value of the target object, a computer-readable storage medium and an electronic device.
Background
With the development of economy in recent years, more and more people take a step of creating an industry, and meanwhile, the rapid development of the internet information technology enables enterprise-related data to be conveniently obtained and utilized. By using basic information of enterprises, enterprise operation conditions, real-time public sentiments of the Internet and other related data, the enterprises with illegal operation and serious abnormal risks can be early warned in advance.
In the method related to enterprise risk early warning at the present stage, the number of abnormal indexes of enterprises in different abnormal types is counted, and the enterprise risk is evaluated by utilizing the number of the abnormal indexes.
However, the above method has the following drawbacks: because the number of the enterprise abnormal indexes is related to factors such as the establishment duration of the enterprise, for example, two enterprises a and B with the same number of the abnormal indexes have a longer establishment duration than B, if the risk of a and B is obtained as high as the number of the abnormal indexes, the accuracy of the risk assessment result is low.
Therefore, it is desirable to provide a new method and apparatus for calculating an evaluation value of a target object.
It is to be noted that the information invented in the above background section is only for enhancing the understanding of the background of the present invention, and therefore, may include information that does not constitute prior art known to those of ordinary skill in the art.
Disclosure of Invention
An object of the present invention is to provide an evaluation value calculation method of a target object, an evaluation value calculation apparatus of a target object, a computer-readable storage medium, and an electronic device, thereby overcoming, at least to some extent, the problem of low accuracy of an evaluation value due to the limitations and drawbacks of the related art.
According to an aspect of the present disclosure, there is provided an evaluation value calculation method of a target object, including:
acquiring data to be processed corresponding to a target object, and obtaining a plurality of view data matrixes according to the data to be processed;
inputting each view data matrix into a learner corresponding to each view data matrix to obtain a probability value corresponding to each view data matrix;
and obtaining the evaluation value of the target object according to the probability value corresponding to each view data matrix.
In an exemplary embodiment of the present disclosure, the data to be processed includes a plurality of first category data, second category data, and third category data;
obtaining a plurality of view data matrices according to the data to be processed comprises:
respectively carrying out abnormal value cleaning and/or missing value completion on the first category data and the second category data;
normalizing the cleaned and/or completed first class data and the second class data to obtain a first view data matrix and a second view data matrix; and
and performing feature extraction on the third category data to obtain text features, and performing vectorization processing on the text features to obtain a third view data matrix.
In an exemplary embodiment of the present disclosure, inputting each of the view data matrices into a learner corresponding to each of the view data matrices, and obtaining a weight value and a probability value corresponding to each of the view data matrices includes:
performing characteristic combination on the first view data matrix, and performing standardization processing on the first view data matrix after the characteristic combination to obtain a standard data matrix;
inputting the standard data matrix and the class label corresponding to the standard data matrix into a first learner to obtain a first probability value corresponding to the first view data matrix; and
inputting the second view data matrix into a second learner to obtain a second probability value corresponding to the second view data matrix; and
and inputting the third view data matrix into a third learner to obtain a third probability value corresponding to the third view data matrix.
In an exemplary embodiment of the disclosure, inputting the standard data matrix and the category label corresponding to the standard data matrix into a first learner, obtaining a first probability value corresponding to the first view data matrix comprises:
according to the category label corresponding to the standard data matrix, carrying out weighted summation on the combination characteristics included in the standard data matrix and the weight corresponding to each combination characteristic to obtain a first operation result;
and mapping the first operation result by using an activation function to obtain the first probability value.
In an exemplary embodiment of the present disclosure, the first learner is a logistic regression model;
the second learner is a lightweight gradient lifting tree model;
the third learner is a naive bayes model.
In an exemplary embodiment of the present disclosure, obtaining the evaluation value of the target object according to the weight value and the probability value corresponding to each of the view data matrices includes:
configuring a first weight value corresponding to the first learner, a second probability value corresponding to the second learner, and a third probability value corresponding to the third learner;
weighting and summing the first weight value and the first probability value, the second weight value and the second probability value, and the third weight value and the third probability value to obtain an evaluation value of the target object;
wherein the sum of the first weight value, the second weight value and the third weight value is 1.
In an exemplary embodiment of the present disclosure, the evaluation value calculation method of the target object further includes:
sequencing the target objects according to the evaluation value of the target objects to obtain a sequencing result;
and identifying the target object with the evaluation value larger than a preset threshold value according to the sequencing result.
According to an aspect of the present disclosure, there is provided an evaluation value calculation apparatus of a target object, including:
the first processing module is used for acquiring data to be processed corresponding to a target object and obtaining a plurality of view data matrixes according to the data to be processed;
the second processing module is used for respectively inputting the view data matrixes into the learners corresponding to the view data matrixes to obtain probability values corresponding to the view data matrixes;
and the evaluation value calculation module is used for obtaining the evaluation value of the target object according to the probability value corresponding to each view data matrix.
According to an aspect of the present disclosure, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the evaluation value calculation method of the target object described in any one of the above.
According to an aspect of the present disclosure, there is provided an electronic device including:
a processor; and
a memory for storing executable instructions of the processor;
wherein the processor is configured to execute the evaluation value calculation method of the target object of any one of the above items via execution of the executable instructions.
On one hand, data to be processed corresponding to the target object are obtained, and a plurality of view data matrixes are obtained according to the data to be processed; respectively inputting the view data matrixes into a learner corresponding to the view data matrixes to obtain probability values corresponding to the view data matrixes; finally, the evaluation value of the target object is obtained according to the probability value corresponding to each view data matrix, the problem that in the prior art, the risk of an enterprise is evaluated only through the number of abnormal indexes, so that the accuracy of the evaluation result is low is solved, and the accuracy of the evaluation value of the target object is improved; on the other hand, the probability values corresponding to the view data matrixes are obtained by respectively inputting the view data matrixes into the learners corresponding to the view data matrixes; then, the evaluation value of the target object is obtained according to the probability value corresponding to each view data matrix, so that the problem that the evaluation efficiency of the target object is low due to the fact that the evaluation value of the target object needs to be obtained manually through a rule analysis method is solved, and the evaluation efficiency of the target object is improved; on the other hand, the problem that the evaluation value of the target object lacks stability due to the fact that the evaluation value of the target object needs to be obtained manually through a rule analysis method is solved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention, as claimed.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention. It is obvious that the drawings in the following description are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.
Fig. 1 schematically shows a flowchart of an evaluation value calculation method of a target object according to an exemplary embodiment of the present invention.
Fig. 2 schematically shows a flow chart of a method for deriving a plurality of view data matrices from data to be processed according to an exemplary embodiment of the present invention.
Fig. 3 is a flowchart schematically illustrating a method for inputting each of the view data matrices into a learner corresponding to each of the view data matrices to obtain a weight value and a probability value corresponding to each of the view data matrices according to an exemplary embodiment of the present invention.
Fig. 4 schematically shows a flowchart of another evaluation value calculation method of a target object according to an exemplary embodiment of the present invention.
Fig. 5 schematically shows a flow chart of a method of training a learner according to an exemplary embodiment of the present invention.
Fig. 6 schematically shows a flowchart of another evaluation value calculation method of a target object according to an exemplary embodiment of the present invention.
Fig. 7 schematically shows a block diagram of an evaluation value calculation apparatus of a target object according to an exemplary embodiment of the present invention.
Fig. 8 schematically illustrates an electronic device for implementing the above-described evaluation value calculation method of the target object according to an exemplary embodiment of the present invention.
Detailed Description
Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in many different forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to provide a thorough understanding of embodiments of the invention. One skilled in the relevant art will recognize, however, that the invention may be practiced without one or more of the specific details, or with other methods, components, devices, steps, and so forth. In other instances, well-known technical solutions have not been shown or described in detail to avoid obscuring aspects of the invention.
Furthermore, the drawings are merely schematic illustrations of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus their repetitive description will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different networks and/or processor devices and/or microcontroller devices.
In some methods for assessing risk of enterprises, two methods can be mainly included: one is that, enterprise risk early warning is completed by a method of counting abnormal indexes: the enterprise risk ranking is completed by counting the number of abnormal indexes of different abnormal types of enterprises and utilizing the number of the abnormal indexes; the other is that the enterprises with risks are discovered through a large number of manual participation modes based on enterprise data: and determining the industry class to which the enterprise belongs through the enterprise related data, and selecting a proper risk identification model according to the industry class.
However, the above method has the following drawbacks: in the method for counting the number of abnormal indexes: due to the influence of numerous random uncertain factors such as supervision and the like, enterprises with risks are difficult to objectively identify through single statistic of abnormal enterprise indexes. For example, the number of abnormal indexes of an enterprise is related to the time length of establishment of the enterprise, for example, two enterprises a and B with the same number of abnormal indexes, the establishment time of a is longer than that of B, and the risk of a and B cannot be obtained as high as if the number of abnormal indexes is equal. Therefore, the method cannot comprehensively and accurately identify the enterprise risk.
In selecting a recognition model method according to industry categories, the method has two difficulties: firstly, determining the industry type of an enterprise by using enterprise data; second, the selection aspect of the model. Specifically, determining the industry class to which the enterprise belongs according to enterprise data requires a large amount of manual work to set the industry class and the division rule; and part of the enterprise demarcation boundaries are not quite obvious, so it is difficult to determine the appropriate rule. If the existing industry classification to which the enterprise belongs is relied on, the number of the classification is too large, and correspondingly, a risk identification model is also needed to be too many, which is difficult to implement in the actual application scene. In addition, it is difficult for some enterprises with industry intersection to select a very suitable unique risk identification model, and the identification process of enterprises of the same industry category has great randomness, which will affect the final risk prediction effect.
The present exemplary embodiment first provides a method for calculating an evaluation value of a target object, which may be performed by a server, a server cluster, a cloud server, or the like, or may be performed by a terminal device; of course, those skilled in the art may also operate the method of the present invention on other platforms as needed, and this is not particularly limited in this exemplary embodiment. Referring to fig. 1, the evaluation value calculation method of the target object may include the steps of:
and S110, acquiring data to be processed corresponding to the target object, and obtaining a plurality of view data matrixes according to the data to be processed.
And S120, respectively inputting each view data matrix into a learner corresponding to each view data matrix to obtain a probability value corresponding to each view data matrix.
And S130, obtaining the evaluation value of the target object according to the probability value corresponding to each view data matrix.
In the evaluation value calculation method of the target object, on one hand, a plurality of view data matrixes are obtained by acquiring to-be-processed data corresponding to the target object and according to the to-be-processed data; respectively inputting the view data matrixes into a learner corresponding to the view data matrixes to obtain probability values corresponding to the view data matrixes; finally, the evaluation value of the target object is obtained according to the probability value corresponding to each view data matrix, the problem that in the prior art, the risk of an enterprise is evaluated only through the number of abnormal indexes, so that the accuracy of the evaluation result is low is solved, and the accuracy of the evaluation value of the target object is improved; on the other hand, the probability values corresponding to the view data matrixes are obtained by respectively inputting the view data matrixes into the learners corresponding to the view data matrixes; then, the evaluation value of the target object is obtained according to the probability value corresponding to each view data matrix, so that the problem that the evaluation efficiency of the target object is low due to the fact that the evaluation value of the target object needs to be obtained manually through a rule analysis method is solved, and the evaluation efficiency of the target object is improved; on the other hand, the problem that the evaluation value of the target object lacks stability due to the fact that the evaluation value of the target object needs to be obtained manually through a rule analysis method is solved.
Hereinafter, each step involved in the evaluation value calculation method of the target object of the exemplary embodiment of the present invention will be explained and explained in detail with reference to the drawings.
First, an application scenario of an exemplary embodiment of the present invention is explained and explained. Exemplary embodiments of the present invention are directed to an accurate, intelligent and real-time enterprise risk assessment method, which can comprehensively assess the risk of an enterprise from multiple aspects, such as enterprise basic information, related business statistics information and enterprise public opinion data, to obtain a more accurate assessment value.
In step S110, to-be-processed data corresponding to the target object is obtained, and a plurality of view data matrices are obtained according to the to-be-processed data.
In the present exemplary embodiment, first, data to be processed is explained and explained. Specifically, the data to be processed may include a first category of data, a second category of data, a third category of data, and the like, where the first category of data may be, for example, enterprise-based data, the second category of data may be statistical data of an enterprise, and the third category of data may be, for example, public opinion data of an enterprise, and the like. Wherein:
the enterprise-based data may include the underlying business information of the enterprise (e.g., enterprise name, enterprise registration time, enterprise operational age, operational scope, registered capital, etc.), knowledge products (trademark, copyright, and patent products owned by the enterprise, etc.), branches of the enterprise (affiliates, subsidiaries, or exclusive companies, etc.), and so on.
The statistical data may relate to financial, business operations, and business-related risk data for the business, among others. Because various types of statistical features are more, the following description is simply enumerated by the following items: the enterprise financial correlation comprises the invested amount and the investment frequency of the enterprise, the external investment amount of the enterprise and the like; the enterprise operation index comprises enterprise change information statistics, enterprise movable property mortgage change information statistics and the like; the enterprise related risk statistical data comprises the number of times of enterprise serious violation, the frequency of the serious violation, the number of times of enterprise administrative penalty, the frequency of the administrative penalty and the like.
Public opinion data may include information about the relevant opinions and ratings of businesses by first time everyone in an interactive platform (e.g., posts, banners, corporate networks, microblogs, etc.).
Further, in this exemplary embodiment, the data to be processed may be obtained according to identification information of the target object (for example, a uniform credit code or a taxpayer identification number of an enterprise, and then a plurality of view data matrices may be obtained according to the data to be processed. Specifically, referring to fig. 2, obtaining a plurality of view data matrices according to the data to be processed may include steps S210 to S230, which will be described in detail below.
In step S210, outlier cleaning and/or missing value completion are performed on the first category data and the second category data, respectively.
In step S220, the cleaned and/or supplemented first category data and second category data are normalized to obtain a first view data matrix and a second view data matrix.
In step S230, feature extraction is performed on the third category data to obtain a text feature, and vectorization processing is performed on the text feature to obtain a third view data matrix.
Hereinafter, steps S210 to S230 will be explained and explained. Firstly, aiming at enterprise basic data (first class data) and statistical characteristics (second class data), abnormal value cleaning and missing value completion are required, data standardization or normalization is carried out according to model requirements, and finally two different data sets (a first view data matrix and a second view data matrix) containing the enterprise basic data and the statistical characteristics are obtained respectively. Further, for the public opinion data (third category data) of the enterprise, a natural language text feature extraction and vectorization method can be used to convert the public opinion data into a data set (third view data matrix) which can be directly input into an algorithm model, for example, for the public opinion data of "company a illegally absorbs deposit to the public", the entity identification method is used to extract the main content of the word "company a and illegally absorbs deposit", and then the word is converted into a text vector [0,1, … 0] by a coding method (the text vector here is only for example and has no practical meaning), and the vector can be directly input into the algorithm for training.
It should be noted that, the view data matrices are all presented in a vector form, for example, for the first type data and the second type data, the first view data matrix and the second view data matrix may be obtained through an encoding method, and the specifically adopted encoding method may be, for example, one-hot encoding, or other encoding methods, which is not limited in this example.
In step S120, the view data matrices are input to the learners corresponding to the view data matrices, respectively, and probability values corresponding to the view data matrices are obtained.
In the present exemplary embodiment, referring to fig. 3, inputting each of the view data matrices into the learner corresponding to each of the view data matrices, and obtaining the probability value corresponding to each of the view data matrices may include steps S310 to S340, which will be described in detail below.
In step S310, feature combination is performed on the first view data matrix, and normalization processing is performed on the first view data matrix after feature combination to obtain a standard data matrix.
In the present example embodiment, the transformation and construction of features included in the enterprise-based data may be accomplished through feature automation extraction scripts. Further, since most of the enterprise basic data relates to the category features (such as the type of the enterprise, the state of the enterprise, etc.), and such data cannot be learned directly by using a machine learning algorithm, the category features are converted by using OneHot coding. Then, the existing data (the first view data matrix) is used to construct important features (perform feature combination), for example, the enterprise establishment time is used to obtain the establishment time of the enterprise, the enterprise establishment time is used as a risk assessment index, the difference between the enterprise approval time and the enterprise establishment time is also used as an assessment index, the business states of the enterprise parent company and the subsidiary company are also used as an important assessment index, and other assessment indexes are not listed one by one here. Then, in order to obtain a better feature expression effect, the first view data matrix after feature combination may be normalized to obtain a standard data matrix.
In step S320, the standard data matrix and the category label corresponding to the standard data matrix are input to a first learner, so as to obtain a first probability value corresponding to the first view data matrix.
In this example embodiment, first, according to a category label corresponding to the standard data matrix, performing weighted summation on combination features included in the standard data matrix and weights corresponding to the combination features to obtain a first operation result; and secondly, mapping the first operation result by using an activation function to obtain the first probability value. In detail:
the normalized data matrix and the class label corresponding thereto are input into a first learner for training, wherein the first learner may be, for example, a Logistic Regression (LR) training. Specifically, the LR algorithm may first perform weighted summation on all features in the enterprise basic information data, and then map the output value to the [0,1] interval by using a Sigmod function to obtain a first probability value. It should be noted that the value of the interval represents the risk score of the enterprise, and a larger value indicates a larger risk of the enterprise.
In step S330, the second view data matrix is input into a second learner, and a second probability value corresponding to the second view data matrix is obtained. In detail:
in order to obtain a better effect, the second learner may select a LightGBM classification model (lightweight gradient spanning tree model) with strong engineering practice, and may complete model training based on statistical indexes. The LightGBM has no strict standardization and normalization requirements on input data, and can also process missing values and classification characteristics by itself, so that the LightGBM has strong robustness. All statistical indexes of the enterprise are used as input of the model, the LightGBM can efficiently use a histogram algorithm to find the optimal segmentation point of the enterprise characteristics to complete the splitting of the tree, the final output value is also in the [0,1] interval, and the probability value of the interval is larger, so that the larger the risk of the enterprise is.
In step S340, the third view data matrix is input into a third learner, and a third probability value corresponding to the third view data matrix is obtained. In detail:
first, the third learner may be, for example
Bayes (naive Bayes model), the risk assessment of enterprise public opinion text data can be completed through the model. Further, can be prepared by
The Bayes algorithm can obtain the risk probability value of the enterprise according to the Bayes formula, and the risk probability is also [0,1]]The larger the value, the higher the risk.
It should be further added that, the above steps S320 to S340 are performed simultaneously in a parallel manner, and there is no precedence order; it is identified here by S320, S330 and S340 only for convenience of description and not for precedence.
In step S130, an evaluation value of the target object is obtained according to the probability value corresponding to each view data matrix.
In the present exemplary embodiment, first, a first weight value corresponding to the first learner, a second probability value corresponding to the second learner, and a third probability value corresponding to the third learner are configured; secondly, carrying out weighted summation on the first weight value and the first probability value, the second weight value and the second probability value, and the third weight value and the third probability value to obtain an evaluation value of the target object; wherein the sum of the first weight value, the second weight value and the third weight value is 1. In detail:
the first, second, and third weight values may be assumed to be: { w1,w2,w3}; and has: w is a1+w2+w3＝1；
The first, second, and third probability values are: { y1,y2,y3}；
The final risk fusion score results are:
wherein the size of n is consistent with the number of models, so that n is 3; the weight is initialized randomly, and the optimal weight parameter is obtained by adopting an iterative optimization mode.
It should be added here that when performing risk assessment on an enterprise, other multiple learners may also be selected, for example, four, five, or two learners may also be selected, and this example is not particularly limited. The embodiment of the invention takes 3 learners as an example, thereby avoiding the problem that the classification of enterprise source data is not careful enough and the accuracy of the evaluation value is low due to too few learners; meanwhile, the problem that due to the fact that the number of learners is too large, data processing is too complicated, and therefore evaluation efficiency is low is solved.
It should be further added that, the learners described above may be not only classification algorithms, but also clustering, semi-supervised learning algorithms, even deep learning models, and the like according to data labels, and this example is not particularly limited. In the process of training each learner, when the difference between the predicted evaluation value obtained by the learner and the standard evaluation value is large, the weight value of each learner needs to be adjusted, and the adjustment can be stopped until the difference between the predicted evaluation value and the standard evaluation value is smaller than a preset threshold value. Specifically, the magnitude of each weight value may be adjusted by setting a loss function, or may be adjusted in other manners, which is not limited in this example.
Fig. 4 schematically shows a flowchart of another evaluation value calculation method of a target object according to an exemplary embodiment of the present invention. Referring to fig. 4, the flowchart of the evaluation value calculation method of the target object may further include step S410 and step S420, which will be described in detail below.
In step S410, the target objects are sorted according to the evaluation values of the target objects to obtain a sorting result.
In step S420, according to the sorting result, a target object whose evaluation value is greater than a preset threshold is identified.
Hereinafter, steps S410 to S420 will be explained. Firstly, after the evaluation value of each target object is obtained, the target objects (to-be-evaluated enterprises) can be ranked according to the evaluation value to obtain a ranking result, and then the target objects (to-be-evaluated enterprises) with the evaluation values larger than a preset threshold are identified according to the ranking result. Specifically, the high-risk red early warning enterprises are marked by red when the division score is in the interval of [0.8,1 ]; the [0.6,0.8) interval represents that the risk is high, and can be marked by orange; 0.45,0.6) indicates possible risk, may be identified with yellow, etc.; less than 0.45 is less risky and may not be identified. And the sequencing result and the identification result can be displayed, so that the investor can refer to the risk of the enterprise to be evaluated according to the evaluation value.
Hereinafter, the training process of the above-described learner involved in the exemplary embodiment of the present invention is explained and explained with reference to fig. 5. Referring to fig. 5, training the learner may include the steps of:
step S510, obtaining multi-source data of an enterprise, and preprocessing the multi-source data to obtain a multi-view data matrix;
step S520, inputting the multi-view data matrix into each learner correspondingly to obtain a weight vector corresponding to each view data matrix;
step S530, fusing weight vectors corresponding to the view data matrixes through a learner fusion algorithm to obtain evaluation values of enterprises;
step S540, determining whether the evaluation value satisfies a threshold condition; if yes, jumping to step S560, if no, jumping to step S550;
step S550, adjusting the weight of each learner, and continuing to step S520;
and step S560, outputting and displaying the evaluation value of the enterprise.
Hereinafter, the evaluation value calculation method of the target object of the exemplary embodiment of the present invention is further explained and explained with reference to fig. 6. Referring to fig. 6, the evaluation value calculation method of the target object may include the steps of:
step S610, acquiring enterprise multi-source data according to identification information of an enterprise to be evaluated, and acquiring enterprise basic information, statistical characteristics and public opinion data according to the enterprise multi-source data;
step S620, preprocessing the enterprise basic information, the statistical characteristics and the public sentiment data to obtain a first view data matrix, a second view data matrix and a third view data matrix;
step S630, correspondingly inputting the first view data matrix, the second view data matrix and the third view data matrix into a first learner, a second learner and a third learner respectively to obtain a first weight vector, a second weight vector and a third weight vector;
step 640, fusing the first weight vector, the second weight vector and the third weight vector by using a model fusion algorithm based on the weight vectors to obtain an evaluation value of the enterprise to be evaluated;
and S650, sorting (ranking) the enterprises to be evaluated according to the evaluation value, and outputting a sorting result.
The evaluation value calculation method for the target object provided by the embodiment of the invention at least has the following advantages:
on one hand, the enterprise risk condition is comprehensively described from a plurality of different angles: most of traditional methods simply extract statistical indexes from enterprise operation abnormal data, and the method based on the statistical indexes is difficult to comprehensively and accurately describe enterprise risk conditions. Aiming at the problems, the invention provides an enterprise risk assessment method based on multi-view risk fusion. The method comprehensively considers enterprise risk abnormal factors from multiple aspects such as enterprise basic information, enterprise operation related statistical characteristics, enterprise public opinion data and the like, and can identify risk enterprises in multiple directions and accurately; meanwhile, the addition of enterprise public opinion data makes real-time prediction more possible.
On the other hand, accurate early warning that intelligence is accurate is unusual, risk enterprise: part of traditional enterprise wind control methods are difficult to implement and are based on a rule analysis problem, a large amount of manual participation is needed in the method, and the risk identification effect is lack of stability. The method provided by the invention trains an optimal model aiming at each different view data set, fuses risk complementary information of a plurality of models and obtains a final result. The method does not need to manually establish rules, is more intelligent, and ensures the stability and robustness of the model.
An exemplary embodiment of the present invention also provides an evaluation value calculation apparatus of a target object. Referring to fig. 7, the evaluation value calculation apparatus of the target object may include a first processing module 710, a second processing module 730, and an evaluation value calculation module 730. Wherein:
the first processing module 710 may be configured to obtain to-be-processed data corresponding to a target object, and obtain a plurality of view data matrices according to the to-be-processed data.
The second processing module 720 may be configured to input each of the view data matrices into a learner corresponding to each of the view data matrices, so as to obtain a weight probability value corresponding to each of the view data matrices.
The evaluation value calculating module 730 may be configured to obtain the evaluation value of the target object according to the probability value corresponding to each of the view data matrices.
In an exemplary embodiment of the present disclosure, the data to be processed includes a plurality of first category data, second category data, and third category data; obtaining a plurality of view data matrices according to the data to be processed comprises:
respectively carrying out abnormal value cleaning and/or missing value completion on the first category data and the second category data; normalizing the cleaned and/or completed first class data and the second class data to obtain a first view data matrix and a second view data matrix; and performing feature extraction on the third category data to obtain text features, and performing vectorization processing on the text features to obtain a third view data matrix.
In an exemplary embodiment of the present disclosure, inputting each of the view data matrices into a learner corresponding to each of the view data matrices, and obtaining a weight value and a probability value corresponding to each of the view data matrices includes:
performing characteristic combination on the first view data matrix, and performing standardization processing on the first view data matrix after the characteristic combination to obtain a standard data matrix; inputting the standard data matrix and the class label corresponding to the standard data matrix into a first learner to obtain a first probability value corresponding to the first view data matrix; inputting the second view data matrix into a second learner to obtain a second probability value corresponding to the second view data matrix; and inputting the third view data matrix into a third learner to obtain a third probability value corresponding to the third view data matrix.
In an exemplary embodiment of the disclosure, inputting the standard data matrix and the category label corresponding to the standard data matrix into a first learner, obtaining a first probability value corresponding to the first view data matrix comprises:
according to the category label corresponding to the standard data matrix, carrying out weighted summation on the combination characteristics included in the standard data matrix and the weight corresponding to each combination characteristic to obtain a first operation result; and mapping the first operation result by using an activation function to obtain the first probability value.
In an exemplary embodiment of the present disclosure, the first learner is a logistic regression model; the second learner is a lightweight gradient lifting tree model; the third learner is a naive bayes model.
In an exemplary embodiment of the present disclosure, obtaining the evaluation value of the target object according to the weight value and the probability value corresponding to each of the view data matrices includes:
configuring a first weight value corresponding to the first learner, a second probability value corresponding to the second learner, and a third probability value corresponding to the third learner;
weighting and summing the first weight value and the first probability value, the second weight value and the second probability value, and the third weight value and the third probability value to obtain an evaluation value of the target object;
wherein the sum of the first weight value, the second weight value and the third weight value is 1.
In an exemplary embodiment of the present disclosure, the evaluation value calculation apparatus of the target object further includes:
the target object sorting module may be configured to sort the target objects according to the magnitude of the evaluation value of the target object to obtain a sorting result;
and the target object identification module can be used for identifying the target object of which the evaluation value is greater than a preset threshold value according to the sorting result.
The details of each module in the above-described evaluation value calculation apparatus for a target object have been described in detail in the evaluation value calculation method for a corresponding target object, and therefore, are not described in detail here.
It should be noted that although in the above detailed description several modules or units of the device for action execution are mentioned, such a division is not mandatory. Indeed, the features and functionality of two or more modules or units described above may be embodied in one module or unit, according to embodiments of the invention. Conversely, the features and functions of one module or unit described above may be further divided into embodiments by a plurality of modules or units.
Moreover, although the steps of the methods of the present invention are depicted in the drawings in a particular order, this does not require or imply that the steps must be performed in this particular order, or that all of the depicted steps must be performed, to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step execution, and/or one step broken down into multiple step executions, etc.
In an exemplary embodiment of the present invention, there is also provided an electronic device capable of implementing the above method.
As will be appreciated by one skilled in the art, aspects of the present invention may be embodied as a system, method or program product. Thus, various aspects of the invention may be embodied in the form of: an entirely hardware embodiment, an entirely software embodiment (including firmware, microcode, etc.) or an embodiment combining hardware and software aspects that may all generally be referred to herein as a "circuit," module "or" system.
An electronic device 800 according to this embodiment of the invention is described below with reference to fig. 8. The electronic device 800 shown in fig. 8 is only an example and should not bring any limitations to the function and scope of use of the embodiments of the present invention.
As shown in fig. 8, electronic device 800 is in the form of a general purpose computing device. The components of the electronic device 800 may include, but are not limited to: the at least one processing unit 810, the at least one memory unit 820, and a bus 830 that couples the various system components including the memory unit 820 and the processing unit 810.
Wherein the storage unit stores program code that is executable by the processing unit 810 to cause the processing unit 810 to perform steps according to various exemplary embodiments of the present invention as described in the above section "exemplary methods" of the present specification. For example, the processing unit 810 may perform step S110 as shown in fig. 1: acquiring data to be processed corresponding to a target object, and obtaining a plurality of view data matrixes according to the data to be processed; step S120: inputting each view data matrix into a learner corresponding to each view data matrix to obtain a probability value corresponding to each view data matrix; step S130: and obtaining the evaluation value of the target object according to the probability value corresponding to each view data matrix.
The storage unit 820 may include readable media in the form of volatile memory units such as a random access memory unit (RAM)8201 and/or a cache memory unit 8202, and may further include a read only memory unit (ROM) 8203.
The storage unit 820 may also include a program/utility 8204 having a set (at least one) of program modules 8205, such program modules 8205 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may comprise an implementation of a network environment.
The electronic device 800 may also communicate with one or more external devices 900 (e.g., keyboard, pointing device, bluetooth device, etc.), with one or more devices that enable a user to interact with the electronic device 800, and/or with any devices (e.g., router, modem, etc.) that enable the electronic device 800 to communicate with one or more other computing devices. Such communication may occur via input/output (I/O) interfaces 850. Also, the electronic device 800 may communicate with one or more networks (e.g., a Local Area Network (LAN), a Wide Area Network (WAN), and/or a public network, such as the internet) via the network adapter 860. As shown, the network adapter 860 communicates with the other modules of the electronic device 800 via the bus 830. It should be appreciated that although not shown, other hardware and/or software modules may be used in conjunction with the electronic device 800, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, among others.
Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein may be implemented by software, or by software in combination with necessary hardware. Therefore, the technical solution according to the embodiment of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a usb disk, a removable hard disk, etc.) or on a network, and includes several instructions to make a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) execute the method according to the embodiment of the present invention.
In an exemplary embodiment of the present invention, there is also provided a computer-readable storage medium having stored thereon a program product capable of implementing the above-described method of the present specification. In some possible embodiments, aspects of the invention may also be implemented in the form of a program product comprising program code means for causing a terminal device to carry out the steps according to various exemplary embodiments of the invention described in the above section "exemplary methods" of the present description, when said program product is run on the terminal device.
According to the program product for realizing the method, the portable compact disc read only memory (CD-ROM) can be adopted, the program code is included, and the program product can be operated on terminal equipment, such as a personal computer. However, the program product of the present invention is not limited in this regard and, in the present document, a readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium include: an electrical connection having one or more wires, a portable disk, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
A computer readable signal medium may include a propagated data signal with readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A readable signal medium may also be any readable medium that is not a readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.
Program code embodied on a readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
Program code for carrying out operations for aspects of the present invention may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., through the internet using an internet service provider).
Furthermore, the above-described figures are merely schematic illustrations of processes involved in methods according to exemplary embodiments of the invention, and are not intended to be limiting. It will be readily understood that the processes shown in the above figures are not intended to indicate or limit the chronological order of the processes. In addition, it is also readily understood that these processes may be performed synchronously or asynchronously, e.g., in multiple modules.
Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.
Claims (10)
1. A target object evaluation value calculation method characterized by comprising:
acquiring data to be processed corresponding to a target object, and obtaining a plurality of view data matrixes according to the data to be processed;
inputting each view data matrix into a learner corresponding to each view data matrix to obtain a probability value corresponding to each view data matrix;
and obtaining the evaluation value of the target object according to the probability value corresponding to each view data matrix.
2. The evaluation value calculation method of a target object according to claim 1, wherein the data to be processed includes a plurality of first category data, second category data, and third category data;
obtaining a plurality of view data matrices according to the data to be processed comprises:
respectively carrying out abnormal value cleaning and/or missing value completion on the first category data and the second category data;
normalizing the cleaned and/or completed first class data and the second class data to obtain a first view data matrix and a second view data matrix; and
and performing feature extraction on the third category data to obtain text features, and performing vectorization processing on the text features to obtain a third view data matrix.
3. The method of calculating an evaluation value of a target object according to claim 2, wherein inputting each of the view data matrices into a learner corresponding to each of the view data matrices, respectively, and obtaining a probability value corresponding to each of the view data matrices includes:
performing characteristic combination on the first view data matrix, and performing standardization processing on the first view data matrix after the characteristic combination to obtain a standard data matrix;
inputting the standard data matrix and the class label corresponding to the standard data matrix into a first learner to obtain a first probability value corresponding to the first view data matrix; and
inputting the second view data matrix into a second learner to obtain a second probability value corresponding to the second view data matrix; and
and inputting the third view data matrix into a third learner to obtain a third probability value corresponding to the third view data matrix.
4. The method of calculating an evaluation value of a target object according to claim 3, wherein inputting the standard data matrix and the category label corresponding to the standard data matrix into a first learner, and obtaining a first probability value corresponding to the first view data matrix comprises:
according to the category label corresponding to the standard data matrix, carrying out weighted summation on the combination characteristics included in the standard data matrix and the weight corresponding to each combination characteristic to obtain a first operation result;
and mapping the first operation result by using an activation function to obtain the first probability value.
5. The evaluation value calculation method of a target object according to claim 3, wherein the first learner is a logistic regression model;
the second learner is a lightweight gradient lifting tree model;
the third learner is a naive bayes model.
6. The method of claim 3, wherein obtaining the evaluation value of the target object according to the probability value corresponding to each of the view data matrices comprises:
configuring a first weight value corresponding to the first learner, a second probability value corresponding to the second learner, and a third probability value corresponding to the third learner;
weighting and summing the first weight value and the first probability value, the second weight value and the second probability value, and the third weight value and the third probability value to obtain an evaluation value of the target object;
wherein the sum of the first weight value, the second weight value and the third weight value is 1.
7. The target object evaluation value calculation method according to claim 1, characterized in that the target object evaluation value calculation method further comprises:
sequencing the target objects according to the evaluation value of the target objects to obtain a sequencing result;
and identifying the target object with the evaluation value larger than a preset threshold value according to the sequencing result.
8. An evaluation value calculation apparatus of a target object, characterized by comprising:
the first processing module is used for acquiring data to be processed corresponding to a target object and obtaining a plurality of view data matrixes according to the data to be processed;
the second processing module is used for respectively inputting the view data matrixes into the learners corresponding to the view data matrixes to obtain probability values corresponding to the view data matrixes;
and the evaluation value calculation module is used for obtaining the evaluation value of the target object according to the probability value corresponding to each view data matrix.
9. A computer-readable storage medium on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements the evaluation value calculation method of the target object of any one of claims 1 to 7.
10. An electronic device, comprising:
a processor; and
a memory for storing executable instructions of the processor;
wherein the processor is configured to execute the evaluation value calculation method of the target object of any one of claims 1 to 7 via execution of the executable instructions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910939902.4A CN110704803A (en) | 2019-09-30 | 2019-09-30 | Target object evaluation value calculation method and device, storage medium and electronic device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910939902.4A CN110704803A (en) | 2019-09-30 | 2019-09-30 | Target object evaluation value calculation method and device, storage medium and electronic device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110704803A true CN110704803A (en) | 2020-01-17 |
Family
ID=69197767
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910939902.4A Pending CN110704803A (en) | 2019-09-30 | 2019-09-30 | Target object evaluation value calculation method and device, storage medium and electronic device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110704803A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113076525A (en) * | 2021-03-15 | 2021-07-06 | 北京明略软件系统有限公司 | Population attribute value calculation method and device, storage medium and electronic equipment |
| CN113191784A (en) * | 2021-04-23 | 2021-07-30 | 北京金堤征信服务有限公司 | Abnormal enterprise identification method and device, electronic equipment and storage medium |
| CN113723540A (en) * | 2021-09-02 | 2021-11-30 | 济南大学 | Unmanned scene clustering method and system based on multiple views |
-
2019
- 2019-09-30 CN CN201910939902.4A patent/CN110704803A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113076525A (en) * | 2021-03-15 | 2021-07-06 | 北京明略软件系统有限公司 | Population attribute value calculation method and device, storage medium and electronic equipment |
| CN113191784A (en) * | 2021-04-23 | 2021-07-30 | 北京金堤征信服务有限公司 | Abnormal enterprise identification method and device, electronic equipment and storage medium |
| CN113723540A (en) * | 2021-09-02 | 2021-11-30 | 济南大学 | Unmanned scene clustering method and system based on multiple views |
| CN113723540B (en) * | 2021-09-02 | 2024-04-19 | 济南大学 | Unmanned scene clustering method and system based on multiple views |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kotu et al. | Predictive analytics and data mining: concepts and practice with rapidminer | |
| CN107403345A (en) | Best-selling product Forecasting Methodology and system, storage medium and electric terminal | |
| CN110704803A (en) | Target object evaluation value calculation method and device, storage medium and electronic device | |
| CN108734296A (en) | Optimize method, apparatus, electronic equipment and the medium of the training data of supervised learning | |
| CN112883990A (en) | Data classification method and device, computer storage medium and electronic equipment | |
| CN112348321A (en) | Risk user identification method and device and electronic equipment | |
| CN112561320A (en) | Training method of mechanism risk prediction model, mechanism risk prediction method and device | |
| CN109660533A (en) | Method, apparatus, computer equipment and the storage medium of identification abnormal flow in real time | |
| CN111754317A (en) | Financial investment data evaluation method and system | |
| CN112990281A (en) | Abnormal bid identification model training method, abnormal bid identification method and abnormal bid identification device | |
| CN114048325A (en) | Case processing method and device based on knowledge graph and storage medium | |
| CN112434862B (en) | Method and device for predicting financial dilemma of marketing enterprises | |
| CN117725926A (en) | Evaluation method and system based on natural language processing and violation inspection | |
| CN112989050A (en) | Table classification method, device, equipment and storage medium | |
| CN111353728A (en) | Risk analysis method and system | |
| CN111815435A (en) | Visualization method, device, equipment and storage medium for group risk characteristics | |
| CN112102062A (en) | Risk assessment method and device based on weak supervised learning and electronic equipment | |
| CN115131039B (en) | Enterprise risk assessment method based on nonlinear dimension reduction, computer equipment and storage medium | |
| CN111626315A (en) | Model training method, object recognition method, device, medium, and electronic apparatus | |
| CN114493853A (en) | Credit rating evaluation method, credit rating evaluation device, electronic device and storage medium | |
| CN113935802A (en) | Information processing method, device, equipment and storage medium | |
| US20210312223A1 (en) | Automated determination of textual overlap between classes for machine learning | |
| CN114417974A (en) | Model training method, information processing method, device, electronic device and medium | |
| CN113870007A (en) | Product recommendation method, device, equipment and medium | |
| CN111191688A (en) | User staging number management method and device and electronic equipment |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200117 |