CN107895235A - Financial modeling system based on decision tree - Google Patents
Financial modeling system based on decision tree Download PDFInfo
- Publication number
- CN107895235A CN107895235A CN201711207965.8A CN201711207965A CN107895235A CN 107895235 A CN107895235 A CN 107895235A CN 201711207965 A CN201711207965 A CN 201711207965A CN 107895235 A CN107895235 A CN 107895235A
- Authority
- CN
- China
- Prior art keywords
- data
- decision tree
- module
- financial
- conversion
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/25—Integrating or interfacing systems involving database management systems
- G06F16/254—Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses
-
- 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
-
- 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
- G06Q40/12—Accounting
- G06Q40/125—Finance or payroll
Abstract
The invention discloses the financial modeling system based on decision tree, including data acquisition unit, data verification units, ETL data processing servers, data warehouse, Data Mart and data analysis unit;The data acquisition unit is used for the report data for gathering the balance sheet of enterprise, profit flow table and cash flow statement, and by the data transfer collected to data verification units;Data verification units are checked and are changed to the form of initial data, while by the data transfer after conversion to ETL data processing servers;ETL data processing servers are extracted to initial data, converted, simultaneously by the data after conversion in data warehouse storage, data in data bins are divided into multiple Data Marts, and the financial data in Data Mart is analyzed by data analysis unit, form decision tree.The decision tree analysis report of the present invention can intuitively show the financial situation of enterprise, be easy to enterprise administrator intuitively efficiently to search and the reason for financial situation occur.
Description
Technical field
The invention belongs to financial analysis field, is related to a kind of financial modeling system based on decision tree.
Background technology
Financial statement analysis refers to using financial statement and other data as foundation and starting point, using special method, system
Analysis and past and present management performance, financial situation and its accommodation of evaluation enterprise, it is therefore an objective to which understanding past, evaluation are existing
In, prediction future, help interest relations groups to improve decision-making, the most basic function of financial statement analysis, be by substantial amounts of form
Data dress changes specific decision-making useful information into, reduces the incorrectness of decision-making, the result of financial statement analysis is that enterprise is repaid
Debt ability, profitability and made an appraisal to the ability to ward off risks, or find out the problem of existing.
The content of the invention
It is an object of the invention to provide a kind of financial modeling system based on decision tree, Analysis of Policy Making tree has
Analysing content limit, the various state limits of problem, the factor limit of decision problem, cause problem Producing reason limit, it is right
The question and answer conclusion limit answered a question, and then make it that the performance of financial situation is more accurate and visual, it is easy to enterprise administrator to check
With understanding financial situation.
The purpose of the present invention can be achieved through the following technical solutions:
Financial modeling system based on decision tree, including data acquisition unit, data verification units, ETL data
Processing server, data warehouse, Data Mart and data analysis unit;
The data acquisition unit is used for the form number for gathering the balance sheet of enterprise, profit flow table and cash flow statement
According to, and by the data transfer collected to data verification units;
The data verification units are used to the form of initial data is checked and changed, while by the data after conversion
It is transferred to ETL data processing servers;
The ETL data processing servers are used to be extracted the initial data after conversion, and the data extracted turn
The form of multidimensional data is turned to, while the data transfer after conversion to data warehouse is stored, the number stored in data bins
According to multiple Data Marts are divided into, then the financial data in Data Mart is analyzed by data analysis unit, formation is determined
Plan tree.
Further, the ETL data processing servers include data extraction module, data conversion module, data cleansing
Module and data load-on module, data extraction module are used for the data that data warehouse needs are extracted from data verification units,
And by data pick-up to data conversion area;Data cleansing module is that the source data quality of data transition zone is checked, is formed
Audit report, there are the data of mistake in processing, if gross error occur in data, will carry out data by system maintenance personnel on site
Reason and inspection;Data conversion module is used for the datagram for the data Cun Chudao transition zones for needing data warehouse physical data structure
In table, source data is arranged, is rejected, merged and verified before deposit;Data load-on module is to utilize data loading tool or API
The loading of programing operation interface carries out data loading, loads data into data warehouse.
Further, the data analysis unit includes decision tree generation module and decision tree pruning module.
Further, the decision tree generation module be according to the judgement of the different indexs of finance, according to financial analysis rule,
New decision condition or different conclusions are formed, if there is Rule of judgment, further according to this decision condition, generates new judgement bar
Part or conclusion, until generating all judgement conclusions, it will be ultimately formed a complete financial decision parsing tree;Decision tree beta pruning
Module is the process that the decision tree of generation is verified, corrected and corrected, the data mainly concentrated with new sample data
Caused preliminary rule in Decision Tree Construction is verified, the branch for influenceing pre- weighing apparatus accuracy is wiped out, by beta pruning, with
Processing is due to undue fitting problems caused by the noise in data and outlier.
Further, the method for the beta pruning has two kinds:(1) first beta pruning is i.e. in construction process, when some node meets
Beta pruning condition, then directly stop the construction of this branch;(2) beta pruning afterwards, i.e., the first complete decision tree of construction complete, then by some
Condition traversal tree carries out beta pruning.
Beneficial effects of the present invention:
The Analysis of Policy Making tree of the present invention has analysing content limit, the various state limits of problem, the factor of decision problem
Limit, cause problem Producing reason limit, the question and answer conclusion limit to answering a question, and then cause the performance of financial situation more
It is accurate and visual, it is easy to enterprise administrator to check and understand financial situation.
The decision tree analysis report of the present invention can intuitively show the financial situation of enterprise, be easy to enterprise administrator straight
See efficiently to search and the reason for financial situation occur, and make corresponding counter-measure in time.
Brief description of the drawings
For the ease of it will be appreciated by those skilled in the art that the present invention is further illustrated below in conjunction with the accompanying drawings.
Fig. 1 is the financial modeling system schematic of the invention based on decision tree.
Embodiment
Financial modeling system based on decision tree, as shown in figure 1, including data acquisition unit, data verification list
Member, ETL data processing servers, data warehouse, Data Mart and data analysis unit;
The data acquisition unit is used for the form number for gathering the balance sheet of enterprise, profit flow table and cash flow statement
According to, and by the data transfer collected to data verification units;
The data verification units are used to the form of initial data is checked and changed, while by the data after conversion
It is transferred to ETL data processing servers;
The ETL data processing servers are used to be extracted the initial data after conversion, and the data extracted turn
The form of multidimensional data is turned to, while the data transfer after conversion to data warehouse is stored, the number stored in data bins
According to multiple Data Marts are divided into, then the financial data in Data Mart is analyzed by data analysis unit, formation is determined
Plan tree;
The ETL data processing servers include data extraction module, data conversion module, data cleansing module and data
Load-on module, data extraction module is used for the data that data warehouse needs are extracted from data verification units, and data are taken out
Get data conversion area;Data cleansing module is that the source data quality of data transition zone is checked, forms audit report, place
There are the data of mistake in reason, if gross error occur in data, will carry out data processing and inspection by system maintenance personnel on site;Number
In the data sheet for the data Cun Chudao transition zones for being used to need data warehouse physical data structure according to modular converter, before deposit
Source data arranged, rejected, merged and verified, wherein data conversion has three kinds of forms, and (1) data normalization is by isomery source
Attribute data, according to certain proportional zoom, unified standard is in unified specific section;(2) low level is replaced with high-rise concept
Or initial data, or Data Discretization is carried out, map the data into different levels;(3) attribute construction and conversion are i.e. by data source
Data analyzed, construct new attribute and be added in property set, it is also possible to which attribute transformation is unified to only by the attribute easily purchased
Vertical attribute;Data load-on module is to carry out data loading using data loading tool or the loading of API programing operations interface, by data
It is loaded into data warehouse;
The data analysis unit includes decision tree generation module and decision tree pruning module, and decision tree generation module is root
According to the judgement of the different indexs of finance, according to financial analysis rule, new decision condition or different conclusions are formed, if there is sentencing
Broken strip part, further according to this decision condition, new Rule of judgment or conclusion are generated, until generating all judgement conclusions, most at last
Form a complete financial decision parsing tree;Decision tree pruning module is that the decision tree of generation is verified, corrected and repaiied
Positive process, will mainly with caused preliminary rule in the data check Decision Tree Construction of new sample data concentration
The branch for influenceing pre- weighing apparatus accuracy is wiped out;By beta pruning, to handle due to mistake caused by the noise in data and outlier
Divide fitting problems.The method of the beta pruning has two kinds, and (1) first beta pruning is i.e. in construction process, when some node meets beta pruning bar
Part, then directly stop the construction of this branch;(2) beta pruning afterwards, i.e., the first complete decision tree of construction complete, then pass through some conditions time
Go through tree and carry out beta pruning.
Present invention disclosed above preferred embodiment is only intended to help and illustrates the present invention.Preferred embodiment is not detailed
All details are described, it is only described embodiment also not limit the invention.Obviously, according to the content of this specification,
It can make many modifications and variations.This specification is chosen and specifically describes these embodiments, is to preferably explain the present invention
Principle and practical application so that skilled artisan can be best understood by and utilize the present invention.The present invention is only
Limited by claims and its four corner and equivalent.
Claims (5)
1. the financial modeling system based on decision tree, it is characterised in that including data acquisition unit, data verification list
Member, ETL data processing servers, data warehouse, Data Mart and data analysis unit;
The data acquisition unit is used for the report data for gathering the balance sheet of enterprise, profit flow table and cash flow statement, and
By the data transfer collected to data verification units;
The data verification units are used to the form of initial data is checked and changed, while by the data transfer after conversion
To ETL data processing servers;
The ETL data processing servers are used to be extracted the initial data after conversion, and the data extracted are converted into
The form of multidimensional data, while the data transfer after conversion to data warehouse is stored, the data stored in data bins point
For multiple Data Marts, then the financial data in Data Mart is analyzed by data analysis unit, forms decision tree.
2. the financial modeling system according to claim 1 based on decision tree, it is characterised in that the ETL numbers
Include data extraction module, data conversion module, data cleansing module and data load-on module, data pick-up according to processing server
Module is used for the data that data warehouse needs are extracted from data verification units, and by data pick-up to data conversion area;Number
It is that the source data quality of data transition zone is checked according to cleaning module, forms audit report, there are the data of mistake in processing,
If gross error occur in data, data processing and inspection will be carried out by system maintenance personnel on site;Data conversion module is used for will
Data warehouse physical data structure need data Cun Chudao transition zones data sheet in, before deposit source data arranged,
Reject, merge and verify;Data load-on module is to carry out data using data loading tool or the loading of API programing operations interface to add
Carry, load data into data warehouse.
3. the financial modeling system according to claim 1 based on decision tree, it is characterised in that the data point
Analysis unit includes decision tree generation module and decision tree pruning module.
4. the financial modeling system according to claim 3 based on decision tree, it is characterised in that the decision tree
Generation module is according to the judgement of the different indexs of finance, according to financial analysis rule, forms new decision condition or different knots
By if there is Rule of judgment, further according to this decision condition, generating new Rule of judgment or conclusion, all sentence until generating
Determine conclusion, will be ultimately formed a complete financial decision parsing tree;Decision tree pruning module is the decision tree progress to generation
The process of verification, correction and amendment, mainly produced with the data check Decision Tree Construction of new sample data concentration
Preliminary rule, the branch for influenceing pre- weighing apparatus accuracy is wiped out, by beta pruning, with handle due to the noise in data and from
Undue fitting problems caused by group's point.
5. the financial modeling system according to claim 4 based on decision tree, it is characterised in that the beta pruning
Method has two kinds:(1) first beta pruning when some node meets beta pruning condition, then directly stops this branch i.e. in construction process
Construction;(2) beta pruning afterwards, i.e., the first complete decision tree of construction complete, then travel through tree by some conditions and carry out beta pruning.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711207965.8A CN107895235A (en) | 2017-11-27 | 2017-11-27 | Financial modeling system based on decision tree |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711207965.8A CN107895235A (en) | 2017-11-27 | 2017-11-27 | Financial modeling system based on decision tree |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107895235A true CN107895235A (en) | 2018-04-10 |
Family
ID=61806831
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711207965.8A Pending CN107895235A (en) | 2017-11-27 | 2017-11-27 | Financial modeling system based on decision tree |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107895235A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109300393A (en) * | 2018-08-16 | 2019-02-01 | 中国平安人寿保险股份有限公司 | Financial data display methods, terminal device and the medium of automatic Program Synthesis |
CN111061704A (en) * | 2019-11-01 | 2020-04-24 | 东方微银科技(北京)有限公司 | Financial analysis report generation method and equipment |
CN112015724A (en) * | 2019-09-25 | 2020-12-01 | 国网湖北省电力有限公司黄石供电公司 | Method for analyzing metering abnormality of electric power operation data |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101075304A (en) * | 2006-05-18 | 2007-11-21 | 河北全通通信有限公司 | Method for constructing decision supporting system of telecommunication industry based on database |
CN103108343A (en) * | 2011-11-15 | 2013-05-15 | 中国移动通信集团设计院有限公司 | Method and device of building decision-making tree and method and device of network performance optimization |
CN103902816A (en) * | 2014-03-12 | 2014-07-02 | 郑州轻工业学院 | Electrification detection data processing method based on data mining technology |
CN105447525A (en) * | 2015-12-15 | 2016-03-30 | 中国科学院软件研究所 | Data prediction classification method and device |
CN105630936A (en) * | 2015-12-22 | 2016-06-01 | 北京奇虎科技有限公司 | Unbalanced data processing method and device based on single-class decision tree |
CN105787059A (en) * | 2016-02-29 | 2016-07-20 | 四川长虹电器股份有限公司 | Data warehouse based financial data integration method |
CN106611295A (en) * | 2016-06-28 | 2017-05-03 | 四川用联信息技术有限公司 | Decision tree-based evolutionary programming algorithm for solving material purchasing problem in manufacturing industry |
-
2017
- 2017-11-27 CN CN201711207965.8A patent/CN107895235A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101075304A (en) * | 2006-05-18 | 2007-11-21 | 河北全通通信有限公司 | Method for constructing decision supporting system of telecommunication industry based on database |
CN103108343A (en) * | 2011-11-15 | 2013-05-15 | 中国移动通信集团设计院有限公司 | Method and device of building decision-making tree and method and device of network performance optimization |
CN103902816A (en) * | 2014-03-12 | 2014-07-02 | 郑州轻工业学院 | Electrification detection data processing method based on data mining technology |
CN105447525A (en) * | 2015-12-15 | 2016-03-30 | 中国科学院软件研究所 | Data prediction classification method and device |
CN105630936A (en) * | 2015-12-22 | 2016-06-01 | 北京奇虎科技有限公司 | Unbalanced data processing method and device based on single-class decision tree |
CN105787059A (en) * | 2016-02-29 | 2016-07-20 | 四川长虹电器股份有限公司 | Data warehouse based financial data integration method |
CN106611295A (en) * | 2016-06-28 | 2017-05-03 | 四川用联信息技术有限公司 | Decision tree-based evolutionary programming algorithm for solving material purchasing problem in manufacturing industry |
Non-Patent Citations (3)
Title |
---|
李爱国等: "《数据挖掘原理、算法及应用》", 31 January 2012, 西安电子科技大学出版社 * |
蔡丽艳: "《数据挖掘算法及其应用研究》", 28 February 2013, 电子科技大学出版社 * |
薛砚丹: ""基于决策树算法的高校财务管理与决策分析研究"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109300393A (en) * | 2018-08-16 | 2019-02-01 | 中国平安人寿保险股份有限公司 | Financial data display methods, terminal device and the medium of automatic Program Synthesis |
CN112015724A (en) * | 2019-09-25 | 2020-12-01 | 国网湖北省电力有限公司黄石供电公司 | Method for analyzing metering abnormality of electric power operation data |
CN111061704A (en) * | 2019-11-01 | 2020-04-24 | 东方微银科技(北京)有限公司 | Financial analysis report generation method and equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
So et al. | Factors affecting citation networks in science and technology: focused on non-quality factors | |
Laender et al. | Assessing the research and education quality of the top Brazilian Computer Science graduate programs | |
CN112231333A (en) | Ecological environment data sharing and exchanging method and system | |
CN107895235A (en) | Financial modeling system based on decision tree | |
Yulianto | Extract transform load (ETL) process in distributed database academic data warehouse | |
CN107491877A (en) | A kind of power network construction project Budget Performance method based on fuzzy overall evaluation | |
Ji et al. | Complexity analysis approach for prefabricated construction products using uncertain data clustering | |
US20220327398A1 (en) | Technology maturity judgment method and system based on science and technology data | |
Danks et al. | Measuring culture of innovation: A validation study of the innovation quotient instrument (part 2) | |
Mustafa et al. | Coupling of cryptocurrency trading with the sustainable environmental goals: Is it on the cards? | |
Anauati et al. | Differences in citation patterns across journal tiers: The case of economics | |
Lopes et al. | From little seeds to a big tree: a far-reaching assessment of the integrated reporting stream | |
Nazari et al. | An investigation on the impact of business intelligence over the performance of startup companies according to innovation and knowledge management as mediators | |
CN108140051A (en) | Data based on whole world retrieval generate the connection to global networks system of global commerce grading in real time | |
Hogan et al. | Market dominance, R&D grant funding, and innovation outcomes | |
Alshehadeh et al. | The impact of business intelligence tools on sustaining financial report quality in Jordanian commercial banks | |
CN107093018A (en) | Communication engineering project information method for visualizing and device based on health model | |
CN112231386A (en) | Visual interaction method, system, equipment and storage medium for railway scientific research data | |
Skoogh et al. | Time-consumption analysis of input data activities in discrete event simulation projects | |
CN103559585A (en) | Method and system for achieving library comprehensive performance evaluation | |
Gang et al. | Analysis of the information management system in the manufacturing process of cigarette enterprises using fuzzy AHP | |
CN107967325A (en) | A kind of landmark disease survey and assessment administering method and system | |
Obst | Utilizing the business model canvas to enable sustainability measurement on the business model level: an indicator framework supplementing the business model canvas. | |
Kalhor et al. | Diversity dilemmas: uncovering gender and nationality biases in graduate admissions across top North American computer science programs | |
CN106909691A (en) | A kind of efficient revenue data analysis method based on caching |
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: 20180410 |