CN111652704A - Financial credit risk assessment method based on knowledge graph and graph deep learning - Google Patents
Financial credit risk assessment method based on knowledge graph and graph deep learning Download PDFInfo
- Publication number
- CN111652704A CN111652704A CN202010515785.1A CN202010515785A CN111652704A CN 111652704 A CN111652704 A CN 111652704A CN 202010515785 A CN202010515785 A CN 202010515785A CN 111652704 A CN111652704 A CN 111652704A
- Authority
- CN
- China
- Prior art keywords
- data
- knowledge
- user
- graph
- map
- 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
- 238000012502 risk assessment Methods 0.000 title claims abstract description 29
- 238000013135 deep learning Methods 0.000 title claims abstract description 26
- 238000000034 method Methods 0.000 title claims abstract description 19
- 230000006870 function Effects 0.000 claims abstract description 10
- 238000013528 artificial neural network Methods 0.000 claims abstract description 7
- 238000000605 extraction Methods 0.000 claims description 19
- 238000012545 processing Methods 0.000 claims description 18
- 239000013598 vector Substances 0.000 claims description 13
- 238000005516 engineering process Methods 0.000 claims description 8
- 238000013499 data model Methods 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 6
- 230000004927 fusion Effects 0.000 claims description 5
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000013145 classification model Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000004806 packaging method and process Methods 0.000 claims description 3
- 238000012856 packing Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 abstract description 7
- 238000007418 data mining Methods 0.000 description 3
- 230000004913 activation Effects 0.000 description 2
- 230000002457 bidirectional effect Effects 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention belongs to the field of financial credit risk assessment, and particularly discloses a financial credit risk assessment method based on a knowledge graph and deep learning of a graph, which comprises the following steps: acquiring historical credit data of a user; constructing a user knowledge graph according to the credit data; carrying out map deep learning on the user knowledge map by using a map neural network to obtain the characteristics of the knowledge map; using the characteristics of the knowledge graph to represent the credit characteristics of the user; and performing financial credit risk assessment on the user through a risk assessment model based on the user credit characteristics, and judging whether the user has risks through a softmax function. The financial credit risk of the user is evaluated by adopting a knowledge map and map deep learning mode, the structured data, the semi-structured data and the unstructured data in the historical credit data of the user can be preprocessed, extracted and analyzed, and the data is used as an important basis for evaluating the financial credit of the user, so that the evaluation quality is improved, and the evaluation efficiency is high.
Description
Technical Field
The invention relates to the field related to financial credit risk assessment, in particular to a financial credit risk assessment method based on a knowledge graph and graph deep learning.
Background
Since the 21 st century, with the rise of computer technology and data mining technology, personal credit evaluation is developing towards the direction of database, systematization and high-precision quantification. Credit assessment research using data mining technology is currently being valued by academic institutions and commercial banks both at home and abroad. The data mining technology can not only summarize rules from objective data and establish a personal credit scoring model, but also more comprehensively and scientifically evaluate personal credits from qualitative and quantitative angles; and the characteristics of quick processing of the computer can be fully utilized, the whole credit evaluation process is greatly accelerated, and the credit decision time is shortened.
Disclosure of Invention
The invention aims to provide a financial credit risk assessment method based on a knowledge graph and deep learning of a graph, so as to solve the problems in the background technology.
In order to achieve the purpose, the invention provides the following technical scheme:
a financial credit risk assessment method based on knowledge graph and graph deep learning comprises the following steps:
s1: acquiring historical credit data of a user;
s2: constructing a user knowledge graph according to the credit data;
s3: carrying out map deep learning on the user knowledge map by using a map neural network to obtain the characteristics of the knowledge map;
s4: using the characteristics of the knowledge graph to represent the credit characteristics of the user;
s5: and performing financial credit risk assessment on the user through a risk assessment model based on the user credit characteristics, and judging whether the user has risks through a softmax function.
As a preferred embodiment of the present invention, the historical credit data in S1 includes structured data, semi-structured data, and unstructured data.
As a preferred technical solution of the present invention, S2 specifically includes: s21: carrying out processing pretreatment on the structured data, the semi-structured data and the unstructured data; s22: performing knowledge extraction on the preprocessed unstructured data, semi-structured data and structured credit data, and adding the extracted data into a database; s23: performing knowledge fusion on a knowledge base, including entity disambiguation and coreference resolution; s24: constructing a data model from bottom to top to finish structured and networked knowledge representation; s25: and carrying out knowledge reasoning and knowledge discovery according to the existing data model to complete the construction of the user knowledge graph.
As a preferred technical solution of the present invention, the data preprocessing in S21 includes: s211: reading a text: obtaining a complete set chars _ set, bios _ set and relations _ set of the radicals of the words; s212: traversing training data: packing the token _ id, token, bio, relations and headers in each sentence into the sentence as a list; s213: traversing the current sentence to id the sample data, and packaging a word list embedding _ ids, a list char _ ids of a radical id, a list bio _ ids of an entity label and a list scoping matrix heads of a relationship into the sentence; s214: processing sentence id data to ensure that the dimensionality of each sentence in batch data is equal, the dimensionality of the longest sentence is taken as the maximum dimensionality, and the insufficient filling is 0; the token is a word in a sentence, relations are entity relations, and headers are subscript positions of the corresponding relations.
As a preferred embodiment of the present invention, the extracting knowledge of the unstructured data in S22 includes: s221: extracting key data from the unstructured data based on relevance,
the degree of correlation isWherein, k (w)i,wj) As data wiAnd data wjCorrelation of (d), tfid (w)i) Is wiD is a word frequency and inverse frequency value with respect to the data wiAnd data wjEuclidean distance with respect to word vectors; s222: and performing entity recognition and relation extraction on sentences in the extracted key data by using deep learning.
As a preferred embodiment of the present invention, the extracting knowledge of the structured data in S22 includes: the semi-structured data is converted into associated data by performing conversion processing on the semi-structured data by using a D2R technology.
As a preferred technical solution of the present invention, the processing of the semi-structured data in S22 is based on attribute extraction, ontology information extraction, and open information extraction.
As a preferred technical scheme of the invention, the knowledge fusion is carried out on the knowledge base in S23, and the method comprises entity disambiguation and coreference resolution, wherein the entity disambiguation and coreference resolution are used for judging whether the same-name entity in the knowledge base represents different meanings with the same-name entity, and whether other named entities exist in the knowledge base and represent the same meanings with the same-name entity, and the coreference resolution adopts a decision tree algorithm to determine the characteristics of the entity coreference resolution and is calculated based on the similarity value comparison.
As a preferred technical solution of the present invention, the knowledge inference in S25 is an inference using description logic.
As a preferred technical solution of the present invention, S3 specifically includes: s31: map embedding is carried out on the user credit knowledge map spectrogram by using a Deepwalk algorithm, and vector representations of all nodes and edges of the knowledge map are obtained; s32: and inputting the vector representations of the nodes and the edges into a neural network of the graph for training, and learning the features of the nodes to obtain the feature vector representations of the user nodes.
As a preferred technical scheme of the invention, the risk assessment model in S5 is a multi-classification model, and the model function is a softmax classifierThe model inputs the credit characteristics of the user and judges the risk level according to the output value.
Compared with the prior art, the invention has the beneficial effects that:
the financial credit risk of the user is evaluated by adopting a knowledge map and map deep learning mode, the structured data, the semi-structured data and the unstructured data in the historical credit data of the user can be preprocessed, extracted and analyzed, and the data are used as important bases for evaluating the financial credit of the user, so that the evaluation quality is improved, the evaluation efficiency is high, different processing and extracting methods are adopted aiming at three different data, the key data in the data can be effectively extracted, and the processing efficiency and the recognition rate are further improved.
Detailed Description
The technical solutions in the embodiments of the present invention are clearly and completely described below, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1: the invention provides a technical scheme that: a financial credit risk assessment method based on knowledge graph and graph deep learning comprises the following steps:
s1: acquiring historical credit data of a user;
s2: constructing a user knowledge graph according to the credit data;
s3: carrying out map deep learning on the user knowledge map by using a map neural network to obtain the characteristics of the knowledge map;
s4: using the characteristics of the knowledge graph to represent the credit characteristics of the user;
s5: and performing financial credit risk assessment on the user through a risk assessment model based on the user credit characteristics, and judging whether the user has risks through a softmax function.
In the present embodiment, the historical credit data in S1 includes structured data, semi-structured data, and unstructured data.
In this embodiment, S2 specifically includes: s21: carrying out processing pretreatment on the structured data, the semi-structured data and the unstructured data; s22: performing knowledge extraction on the preprocessed unstructured data, semi-structured data and structured credit data, and adding the extracted data into a database; s23: performing knowledge fusion on a knowledge base, including entity disambiguation and coreference resolution; s24: constructing a data model from bottom to top to finish structured and networked knowledge representation; s25: and carrying out knowledge reasoning and knowledge discovery according to the existing data model to complete the construction of the user knowledge graph.
In the present embodiment, the data preprocessing in S21 includes: s211: reading a text: obtaining a complete set chars _ set, bios _ set and relations _ set of the radicals of the words; s212: traversing training data: packing the token _ id, token, bio, relations and headers in each sentence into the sentence as a list; s213: traversing the current sentence to id the sample data, and packaging a word list embedding _ ids, a list char _ ids of a radical id, a list bio _ ids of an entity label and a list scoping matrix heads of a relationship into the sentence; s214: processing sentence id data to ensure that the dimensionality of each sentence in batch data is equal, the dimensionality of the longest sentence is taken as the maximum dimensionality, and the insufficient filling is 0; the token is a word in a sentence, relations are entity relations, and headers are subscript positions of the corresponding relations.
In this embodiment, the knowledge extraction of the unstructured data in S22 includes: s221: extracting key data from the unstructured data based on relevance,
the degree of correlation isWherein, k (w)i,wj) As data wiAnd data wjCorrelation of (d), tfid (w)i) Is wiD is a word frequency and inverse frequency value with respect to the data wiAnd data wjEuclidean distance with respect to word vectors; s222: and performing entity recognition and relation extraction on sentences in the extracted key data by using deep learning.
In this embodiment, the extracting knowledge of the structured data in S22 includes: the semi-structured data is converted into associated data by performing conversion processing on the semi-structured data by using a D2R technology.
In this embodiment, the semi-structured data processing in S22 is based on attribute extraction, ontology information extraction, and open information extraction.
In this embodiment, in S23, the knowledge base is subjected to knowledge fusion, including entity disambiguation and coreference resolution, where the entity disambiguation and coreference resolution are used to determine whether a same-name entity in the knowledge base represents a different meaning from the same-name entity, and whether other named entities exist in the knowledge base and represent the same meaning as the same-name entity, and the coreference resolution determines the characteristics of the entity coreference resolution by using a decision tree algorithm, and compares and calculates based on the similarity value.
In this embodiment, the step S24 further includes processing the data model, specifically including: word embedding, namely extracting features through bidirectional LSTM to obtain char _ logins, loading word vectors pre-trained by a skip-gram model to obtain worumbedding, and splicing the word embedding and char _ logins to be used as input of the model; performing feature extraction on input through bidirectional LSTM of three hidden layers to obtain LSTM _ out; and performing full connection with an activation function of relu on lstm _ out, and performing entity classification to obtain nerScors, wherein the relu activation function is shown as the following formula:
f(x)=max(0,x);
the dependency between tags is introduced using CRF by BIO tagging strategy. Calculating each word to obtain the scores of different labels; the tag sequence probability of the sentence is calculated. The ner _ loss is obtained by minimizing the cross entropy loss function. Finally, obtaining a label preNers with the highest score by using a viterbi algorithm; and performing word Embedding on the obtained labels to obtain labelEmbedding, and splicing the output lstm _ out and labelEmbedding to obtain rel _ inputs which are used as input of entity relation prediction.
In the present embodiment, the knowledge inference in S25 is inference using description logic.
In this embodiment, S3 specifically includes: s31: map embedding is carried out on the user credit knowledge map spectrogram by using a Deepwalk algorithm, and vector representations of all nodes and edges of the knowledge map are obtained; s32: and inputting the vector representations of the nodes and the edges into a neural network of the graph for training, and learning the features of the nodes to obtain the feature vector representations of the user nodes.
In this embodiment, the risk assessment model in S5 is a multi-classification model, and the model function is softmax classifierThe model inputs the credit characteristics of the user and judges the risk level according to the output value, wherein the risk level comprises high, medium and low.
The financial credit risk of the user is evaluated by adopting a knowledge map and map deep learning mode, the structured data, the semi-structured data and the unstructured data in the historical credit data of the user can be preprocessed, extracted and analyzed, and the data are used as important bases for evaluating the financial credit of the user, so that the evaluation quality is improved, the evaluation efficiency is high, different processing and extracting methods are adopted aiming at three different data, the key data in the data can be effectively extracted, and the processing efficiency and the recognition rate are further improved.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (9)
1. A financial credit risk assessment method based on knowledge graph and graph deep learning is characterized by comprising the following steps:
s1: acquiring historical credit data of a user;
s2: constructing a user knowledge graph according to the credit data;
s3: carrying out map deep learning on the user knowledge map by using a map neural network to obtain the characteristics of the knowledge map;
s4: using the characteristics of the knowledge graph to represent the credit characteristics of the user;
s5: and performing financial credit risk assessment on the user through a risk assessment model based on the user credit characteristics, and judging whether the user has risks through a softmax function.
2. The financial credit risk assessment method based on knowledge-graph and map deep learning according to claim 1, wherein the historical credit data in S1 comprises structured data, semi-structured data and unstructured data.
3. The financial credit risk assessment method based on knowledge-graph and map deep learning according to claim 1, wherein the step S2 specifically comprises: s21: carrying out processing pretreatment on the structured data, the semi-structured data and the unstructured data; s22: performing knowledge extraction on the preprocessed unstructured data, semi-structured data and structured credit data, and adding the extracted data into a database; s23: performing knowledge fusion on a knowledge base, including entity disambiguation and coreference resolution; s24: constructing a data model from bottom to top to finish structured and networked knowledge representation; s25: and carrying out knowledge reasoning and knowledge discovery according to the existing data model to complete the construction of the user knowledge graph.
4. The financial credit risk assessment method based on knowledge-graph and map deep learning according to claim 3, wherein the data preprocessing in S21 comprises: s211: reading a text: obtaining a complete set chars _ set, bios _ set and relations _ set of the radicals of the words; s212: traversing training data: packing the token _ id, token, bio, relations and headers in each sentence into the sentence as a list; s213: traversing the current sentence to id the sample data, and packaging a word list embedding _ ids, a list char _ ids of a radical id, a list bio _ ids of an entity label and a list scoping matrix heads of a relationship into the sentence; s214: processing sentence id data to ensure that the dimensionality of each sentence in batch data is equal, the dimensionality of the longest sentence is taken as the maximum dimensionality, and the insufficient filling is 0; the token is a word in a sentence, relations are entity relations, and headers are subscript positions of the corresponding relations.
5. The financial credit risk assessment method based on knowledge-graph and map deep learning according to claim 3, wherein the knowledge extraction of the unstructured data in the S22 comprises: s221: extracting key data from the unstructured data based on relevance,
the degree of correlation isWherein, k (w)i,wj) As data wiAnd data wjCorrelation of (d), tfid (w)i) Is wiD is a word frequency and inverse frequency value with respect to the data wiAnd data wjEuclidean distance with respect to word vectors(ii) a S222: and performing entity recognition and relation extraction on sentences in the extracted key data by using deep learning.
6. The financial credit risk assessment method based on knowledge-graph and map deep learning according to claim 3, wherein the knowledge extraction of the structured data in S22 comprises: the semi-structured data is converted into associated data by performing conversion processing on the semi-structured data by using a D2R technology.
7. The financial credit risk assessment method based on knowledge-graph and deep learning of graph as claimed in claim 3, wherein the semi-structured data processing in the S22 is based on attribute extraction, ontology information extraction and open information extraction.
8. The financial credit risk assessment method based on knowledge-graph and map deep learning according to claim 1, wherein the step S3 specifically comprises: s31: map embedding is carried out on the user credit knowledge map spectrogram by using a Deepwalk algorithm, and vector representations of all nodes and edges of the knowledge map are obtained; s32: and inputting the vector representations of the nodes and the edges into a neural network of the graph for training, and learning the features of the nodes to obtain the feature vector representations of the user nodes.
9. The financial credit risk assessment method based on knowledge-graph and map deep learning as claimed in claim 1, wherein the risk assessment model in S5 is a multi-classification model with model function of softmax classifierThe model inputs the credit characteristics of the user and judges the risk level according to the output value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010515785.1A CN111652704A (en) | 2020-06-09 | 2020-06-09 | Financial credit risk assessment method based on knowledge graph and graph deep learning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010515785.1A CN111652704A (en) | 2020-06-09 | 2020-06-09 | Financial credit risk assessment method based on knowledge graph and graph deep learning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111652704A true CN111652704A (en) | 2020-09-11 |
Family
ID=72351304
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010515785.1A Pending CN111652704A (en) | 2020-06-09 | 2020-06-09 | Financial credit risk assessment method based on knowledge graph and graph deep learning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111652704A (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112069808A (en) * | 2020-09-28 | 2020-12-11 | 深圳壹账通智能科技有限公司 | Financing wind control method and device, computer equipment and storage medium |
CN112100290A (en) * | 2020-09-15 | 2020-12-18 | 山东交通职业学院 | E-commerce information sharing system |
CN112463981A (en) * | 2020-11-26 | 2021-03-09 | 福建正孚软件有限公司 | Enterprise internal operation management risk identification and extraction method and system based on deep learning |
CN112529681A (en) * | 2021-02-08 | 2021-03-19 | 南京审计大学 | Credit risk transfer method based on credit subject correlation strength |
CN112598489A (en) * | 2020-12-14 | 2021-04-02 | 深圳市快付通金融网络科技服务有限公司 | Risk monitoring method and system based on financial application |
CN112597232A (en) * | 2020-12-29 | 2021-04-02 | 南京小灿灿网络科技有限公司 | Data distributed publishing method and system |
CN113487416A (en) * | 2021-07-26 | 2021-10-08 | 中国科学院大学 | Financial risk assessment method based on knowledge graph |
CN113569059A (en) * | 2021-09-07 | 2021-10-29 | 浙江网商银行股份有限公司 | Target user identification method and device |
CN113783874A (en) * | 2021-09-10 | 2021-12-10 | 国网电子商务有限公司 | Network security situation assessment method and system based on security knowledge graph |
CN116167867A (en) * | 2022-12-26 | 2023-05-26 | 中国人民财产保险股份有限公司 | Knowledge graph-based insurance business risk identification method and device and electronic equipment |
CN116703553A (en) * | 2023-08-07 | 2023-09-05 | 浙江鹏信信息科技股份有限公司 | Financial anti-fraud risk monitoring method, system and readable storage medium |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109345399A (en) * | 2018-10-23 | 2019-02-15 | 平安科技(深圳)有限公司 | Claims Resolution methods of risk assessment, device, computer equipment and storage medium |
CN110689423A (en) * | 2019-08-22 | 2020-01-14 | 平安科技(深圳)有限公司 | Credit assessment method and device |
-
2020
- 2020-06-09 CN CN202010515785.1A patent/CN111652704A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109345399A (en) * | 2018-10-23 | 2019-02-15 | 平安科技(深圳)有限公司 | Claims Resolution methods of risk assessment, device, computer equipment and storage medium |
CN110689423A (en) * | 2019-08-22 | 2020-01-14 | 平安科技(深圳)有限公司 | Credit assessment method and device |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112100290A (en) * | 2020-09-15 | 2020-12-18 | 山东交通职业学院 | E-commerce information sharing system |
CN112069808A (en) * | 2020-09-28 | 2020-12-11 | 深圳壹账通智能科技有限公司 | Financing wind control method and device, computer equipment and storage medium |
CN112463981A (en) * | 2020-11-26 | 2021-03-09 | 福建正孚软件有限公司 | Enterprise internal operation management risk identification and extraction method and system based on deep learning |
CN112598489B (en) * | 2020-12-14 | 2023-12-26 | 深圳市快付通金融网络科技服务有限公司 | Risk monitoring method and system based on financial application |
CN112598489A (en) * | 2020-12-14 | 2021-04-02 | 深圳市快付通金融网络科技服务有限公司 | Risk monitoring method and system based on financial application |
CN112597232A (en) * | 2020-12-29 | 2021-04-02 | 南京小灿灿网络科技有限公司 | Data distributed publishing method and system |
CN112529681A (en) * | 2021-02-08 | 2021-03-19 | 南京审计大学 | Credit risk transfer method based on credit subject correlation strength |
CN113487416A (en) * | 2021-07-26 | 2021-10-08 | 中国科学院大学 | Financial risk assessment method based on knowledge graph |
CN113569059A (en) * | 2021-09-07 | 2021-10-29 | 浙江网商银行股份有限公司 | Target user identification method and device |
CN113783874A (en) * | 2021-09-10 | 2021-12-10 | 国网电子商务有限公司 | Network security situation assessment method and system based on security knowledge graph |
CN113783874B (en) * | 2021-09-10 | 2023-08-29 | 国网数字科技控股有限公司 | Network security situation assessment method and system based on security knowledge graph |
CN116167867A (en) * | 2022-12-26 | 2023-05-26 | 中国人民财产保险股份有限公司 | Knowledge graph-based insurance business risk identification method and device and electronic equipment |
CN116703553A (en) * | 2023-08-07 | 2023-09-05 | 浙江鹏信信息科技股份有限公司 | Financial anti-fraud risk monitoring method, system and readable storage medium |
CN116703553B (en) * | 2023-08-07 | 2023-12-05 | 浙江鹏信信息科技股份有限公司 | Financial anti-fraud risk monitoring method, system and readable storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111652704A (en) | Financial credit risk assessment method based on knowledge graph and graph deep learning | |
CN104408153B (en) | A kind of short text Hash learning method based on more granularity topic models | |
CN113204952B (en) | Multi-intention and semantic slot joint identification method based on cluster pre-analysis | |
CN113868432B (en) | Automatic knowledge graph construction method and system for iron and steel manufacturing enterprises | |
CN110119355B (en) | Knowledge graph vectorization reasoning general software defect modeling method | |
Warghade et al. | Credit card fraud detection from imbalanced dataset using machine learning algorithm | |
CN115952292B (en) | Multi-label classification method, apparatus and computer readable medium | |
CN109614490A (en) | Money article proneness analysis method based on LSTM | |
CN116663539A (en) | Chinese entity and relationship joint extraction method and system based on Roberta and pointer network | |
CN112989803B (en) | Entity link prediction method based on topic vector learning | |
Xu et al. | Short text classification of chinese with label information assisting | |
CN113535928A (en) | Service discovery method and system of long-term and short-term memory network based on attention mechanism | |
An et al. | Generalized category discovery with large language models in the loop | |
CN117272142A (en) | Log abnormality detection method and system and electronic equipment | |
CN116452353A (en) | Financial data management method and system | |
CN116029300A (en) | Language model training method and system for strengthening semantic features of Chinese entities | |
CN116629387B (en) | Text processing method and processing system for training under missing condition | |
Zhang | A Novel Text Classification Model Combining Time Correlation Principle and Rough Set Theory | |
CN112463928B (en) | Technical list generation method and system for field evaluation prediction | |
Singh et al. | Axiomatic Analysis of Pre‐Processing Methodologies Using Machine Learning in Text Mining: A Social Media Perspective in Internet of Things | |
CN113569014B (en) | Operation and maintenance project management method based on multi-granularity text semantic information | |
CN116205217B (en) | Small sample relation extraction method, system, electronic equipment and storage medium | |
CN118194858A (en) | Two-stage optimized wireless network optimization field long entity identification method and system | |
Guo et al. | PLBR: A Semi-supervised Document Key Information Extraction via Pseudo-labeling Bias Rectification | |
Li et al. | Hierarchical Differential Amplifier Contrastive Learning for Semi-supervised Extractive Summarization |
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 |
Application publication date: 20200911 |
|
RJ01 | Rejection of invention patent application after publication |