CN103679272A - Vector graph based construction method for binary-decision-tree expert knowledge base - Google Patents
Vector graph based construction method for binary-decision-tree expert knowledge base Download PDFInfo
- Publication number
- CN103679272A CN103679272A CN201310500603.3A CN201310500603A CN103679272A CN 103679272 A CN103679272 A CN 103679272A CN 201310500603 A CN201310500603 A CN 201310500603A CN 103679272 A CN103679272 A CN 103679272A
- Authority
- CN
- China
- Prior art keywords
- vector
- decision tree
- polar plot
- software
- expert knowledge
- 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
Abstract
The invention discloses a vector graph based construction method for a binary-decision-tree expert knowledge base. The method comprises the steps that a) an existed expert knowledge vector graph is selected from a computer via vector drawing software (1), or a blank vector graph is created in the vector drawing software; b) a binary-decision-tree vector graph (3) is drawn in the blank vector graph in the step a) via the vector drawing software (1) and a binary-decision-tree drawing module (2), or the expert knowledge vector graph in the step a) is modified to form a binary-decision-tree vector graph (3), and the drawing result is stored; and c) the binary-decision-tree vector graph (3) is analyzed by calling vector graph analysis software (4), and the vector graph analysis software (4) writes the analysis result into the expert knowledge base (5). The method of the invention improves the accuracy and efficiency in construction of the expert knowledge base, and reduces the difficulty in maintaining the expert knowledge base.
Description
Technical field
The present invention relates to field of artificial intelligence, is a kind of y-bend decision tree expert knowledge library building method based on vector graphics specifically.
Background technology
Expert system rudiment is the forties in 20th century, and grows since 20 century 70s are fast-developing.Expert system has been widely used in the various fields such as chemistry, electronics, medical science, geology at present.Expert system is the Yi Ge branch of artificial intelligence field.One of early stage guide person of expert system, the Edward Feigenbaum professor of Stanford University, is defined as expert system " a kind of computer program of intelligence, its working knowledge and reasoning process solve the challenge that only has expert to solve ".Therefore expert system can be expressed as: expert system=knowledge base+inference machine.
The feature of early stage expert system is weak knowledge base, strong inference machine.Its target is to utilize powerful inference method, relies on a small amount of knowledge base and solves a large amount of problems.Wherein that the most famous is general problem solver (General Problem Solver).Until 20 century 70 people just recognize knowledge base too a little less than, even the function of inference machine is infinite powerful or cannot reach human expert's level.Therefore " weak knowledge base+strong inference machine " pattern starts to develop, and progressively forms modern " strong knowledge base+strong inference machine " pattern.This evolution has improved the importance of knowledge engineer (Knowledge Engineer) in expert system development.
Y-bend decision tree described in the application be in expert system for solving the classical way of some category classification problem, Fig. 2 is an example of y-bend decision tree.A decision tree only has a root node, and the node outside root node is called child node.Each child node only has father's node, does not have the node of child node to be called leaf node.For y-bend decision tree, other node Dou Youliangge branch except leaf node, the path of walking when the representative of Yi Ge branch is answered as "Yes", the path of walking when the representative of Yi Ge branch is answered as "No".Y-bend decision tree generally has two category nodes: decision node and conclusion node.Conclusion node is all on leaf node.Its reasoning process, from root node, enters lower floor's decision node according to judged result branch, until arrive a conclusion node.
Knowledge base in expert system is the electronical record of the knowledge in human expert's brains, and knowledge base is mainly linked up and obtained by knowledge engineer and human expert.The process of setting up expert system on the basis of human expert's knowledge is called knowledge engineering (Knowledge Engineering).This process is completed by knowledge engineer.Knowledge engineering is an apprentice of human expert's acquire knowledge, and they are encoded in expert system.The expertise that knowledge engineer constructs has two kinds of forms, and a kind of is software code form, and another kind is the file of database or other form.
The building method of this expert knowledge library can bring three problems.The one, human expert's knowledge engineer's Software Coding expert knowledge library out of failing to understand, cannot find, check owing to linking up and misread the wrong expertise of bringing into.The 2nd, maintenance expert's knowledge base difficulty, newly-increased, modification expertise needs knowledge engineer and human expert again to link up, and knowledge engineer need to re-start coding, with stylish communication, misreads and also may bring in knowledge base.The 3rd, because frequent communication causes the structure efficiency of expert knowledge library low.
Summary of the invention
The technical problem to be solved in the present invention, is to provide a kind of accuracy and efficiency that improves expert knowledge library structure, reduces the y-bend decision tree expert knowledge library building method based on vector graphics of expert knowledge library maintenance difficulties.
Technical solution of the present invention, is to provide a kind of y-bend decision tree expert knowledge library building method based on vector graphics of following steps, comprises the following steps:
A, by vector plotting software, from computing machine, select already present expertise polar plot or a newly-built blank polar plot in vector plotting software;
B, utilize vector plotting software and y-bend decision tree graphics module in the blank polar plot of step 1, to draw y-bend decision tree polar plot, or in the expertise polar plot of step 1, revise y-bend decision tree polar plot, drawing result is preserved;
C, call polar plot analysis software y-bend decision tree polar plot is analyzed, polar plot analysis software writes analysis result in expert knowledge library.
Adopt above method, compared with prior art have the following advantages: adopt the present invention, by vector plotting software and the polar plot analysis software structure y-bend decision tree expert knowledge library that combines, the mode that knowledge engineer and human expert are constructed to expert knowledge library by communication changes the mode that human expert utilizes vector plotting instrument standalone configuration into, this method is simple and easy to use, workable, can, by human expert's independent operation, reduce the communication error between knowledge engineer and human expert; Improve the accuracy and efficiency of expert knowledge library structure, reduced the maintenance difficulties of expert knowledge library.
As improvement, in described step b, described graphics module refers to vector plotting software User Defined graph block built-in or vector plotting software support.
As improvement, in described step b, described y-bend decision tree graphics module at least comprises initial judgement, judges, is connecting line, no connecting line, conclusion totally five class graphics modules; Wherein initial judgement, judgement graphics module allow input enquirement information; Conclusion graphics module allows input conclusion information.Adopt the method, make the drafting of y-bend decision tree more convenient, further improved expert knowledge library structure efficiency.
As improvement, in described step c, described polar plot analysis software is the extender of application software or vector plotting software support independently, and wherein extender can be by vector plotting Bootload, operation.By the expansion of vector plotting software, can make its compatibility stronger, be convenient to loading and the operation of program.
Accompanying drawing explanation
Fig. 1 is the concrete enforcement block diagram that the present invention is based on the y-bend decision tree expert knowledge library building method of vector graphics.
Fig. 2 is the y-bend decision tree schematic diagram of the embodiment of the present invention.
Fig. 3 is that in Fig. 1, y-bend decision tree graphics module forms schematic diagram.
Fig. 4 is polar plot analysis software data processing step.
Shown in figure: 1, vector plotting software, 2, y-bend decision tree graphics module, 3, y-bend decision tree polar plot, 4, polar plot analysis software, 5, expert knowledge library; 6, y-bend decision tree; 21, initial judgement, 22, judgement, 23, be connecting line, 24, no connecting line, 25, conclusion.
Embodiment
Below in conjunction with the drawings and specific embodiments, the present invention is further described.
As shown in Figure 1, the y-bend decision tree expert knowledge library building method based on vector graphics of the present invention, comprises the following steps:
A, by vector plotting software 1, from computing machine, select already present expertise polar plot or a newly-built blank polar plot in vector plotting software;
B, utilize vector plotting software 1 and y-bend decision tree graphics module 2 in the blank polar plot of step 1, to draw y-bend decision tree polar plot 3, or in the expertise polar plot of step 1, revise y-bend decision tree polar plot 3, drawing result is preserved;
C, call 4 pairs of y-bend decision tree polar plots 3 of polar plot analysis software and analyze, polar plot analysis software 4 writes analysis result in expert knowledge library 5.
Vector plotting software has a lot of types, as the Illustrator vector software of Adobe company exploitation; FreeHand10; Corel DRAW etc.The Jiang Yi Visio of Microsoft vector plotting software of the present invention is implemented for example.Modification and maintenance for expert knowledge library are exactly the process that y-bend decision tree polar plot is revised and upgraded.
In described step b, described graphics module refers to vector plotting software 1 User Defined graph block built-in or that vector plotting software 1 is supported.
In described step b, described y-bend decision tree graphics module at least comprises initial judgement 21, judgement 22, is connecting line 23, no connecting line 24, conclusion 25 totally five class graphics modules; Wherein initial judgement 21, judgement 22 graphics modules allow input enquirement information; Conclusion 25 graphics modules allow input conclusion information.
In described step c, described polar plot analysis software 4 is extenders that independently application software or vector plotting software 1 is supported, wherein extender can be by vector plotting Bootload, operation.
Y-bend decision tree 6 is a kind of of y-bend decision tree polar plot 3 in specific embodiment, and by it, the two represents respectively it is the convenience in order to explain.
The present embodiment be take and adopted the Visio of Microsoft vector plotting software construction expert knowledge library the present invention will be described as example.
With regard to concrete expert knowledge library structure flow process, describe below:
The first step, human expert open the Visio of Microsoft software, a newly-built blank plotting file;
Second step, " initial judgement " graphics module of drag and drop, two " judgement " graphics modules, three " conclusion " graphics modules, three " being connecting line " graphics modules, three " no connecting line " graphics modules, to drawing district, carry out layout by these graphics modules by Fig. 2;
The 3rd step, as shown in Figure 2, the text message that " initial judgement " graphics module is set is: R1 resistance is greater than 50 ohm
The 4th step, as shown in Figure 2, the text message that " judgement " graphics module is set is: C1 frequency is greater than 5MHz and is less than 6MHz
The 5th step, as shown in Figure 2, the text message that " judgement " graphics module is set is: R2_1 magnitude of voltage is less than 9V
The 6th step, as shown in Figure 2, the text message that " conclusion " graphics module is set is: filtering circuit is abnormal;
The 7th step, as shown in Figure 2, the text message that " conclusion " graphics module is set is: circuit is normal;
The 8th step, as shown in Figure 2, the text message that " conclusion " graphics module is set is: R5 short circuit;
The 9th step, click tools > > add-in > > generate binary tree expert knowledge library menu item, and expert knowledge library generates complete.
The tenth step, open expert knowledge library, check result.The expert knowledge library that the present embodiment generates is as shown in the table:
| shape?ID | yesShape?ID | noShape?ID | shape?Type | shape?Text |
| 15 | 14 | 2 | Initial judgement | R1 resistance is greater than 50 ohm |
| 14 | 8 | 7 | Judgement | C1 frequency is greater than 5MHz and is less than |
| ? | ? | ? | ? | 6MHz |
| 8 | -1 | -1 | Conclusion | Filtering circuit is abnormal |
| 7 | -1 | -1 | Conclusion | Circuit is normal |
| 2 | 7 | 9 | Judgement | R2_1 magnitude of voltage is less than 9V |
| 9 | -1 | -1 | Conclusion | R5 short circuit |
In order to illustrate in greater detail the data processing step of polar plot analysis software, therefore adopt Fig. 4 to be explained in detail.
Below only just preferred embodiment of the present invention is described, but can not be interpreted as it is limitations on claims.The present invention is not only confined to above embodiment, and its concrete steps allow to change.In a word, all various variations of doing in the protection domain of independent claims of the present invention are all in protection scope of the present invention.
Claims (4)
1. the y-bend decision tree expert knowledge library building method based on vector graphics, is characterized in that: comprise the following steps:
A, by vector plotting software (1), from computing machine, select already present expertise polar plot or a newly-built blank polar plot in vector plotting software;
B, utilize vector plotting software (1) and y-bend decision tree graphics module (2) in the blank polar plot of step (1), to draw y-bend decision tree polar plot (3), or in the expertise polar plot of step 1, revise y-bend decision tree polar plot (3), drawing result is preserved;
C, call polar plot analysis software (4) y-bend decision tree polar plot (3) is analyzed, polar plot analysis software (4) writes analysis result in expert knowledge library (5).
2. the y-bend decision tree expert knowledge library building method based on vector graphics according to claim 1, it is characterized in that: in described step b, described graphics module refers to vector plotting software (1) User Defined graph block built-in or that vector plotting software (1) is supported.
3. the y-bend decision tree expert knowledge library building method based on vector graphics according to claim 1, it is characterized in that: in described step b, described y-bend decision tree graphics module at least comprises initial judgement (21), judgement (22), is connecting line (23), no connecting line (24), conclusion (25) totally five class graphics modules; Wherein initial judgement (21), judgement (22) graphics module allow input enquirement information; Conclusion (25) graphics module allows input conclusion information.
4. the y-bend decision tree expert knowledge library building method based on vector graphics according to claim 1, it is characterized in that: in described step c, described polar plot analysis software (4) is the extender that independently application software or vector plotting software (1) are supported, wherein extender can be by vector plotting Bootload, operation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310500603.3A CN103679272A (en) | 2013-10-12 | 2013-10-12 | Vector graph based construction method for binary-decision-tree expert knowledge base |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310500603.3A CN103679272A (en) | 2013-10-12 | 2013-10-12 | Vector graph based construction method for binary-decision-tree expert knowledge base |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103679272A true CN103679272A (en) | 2014-03-26 |
Family
ID=50316756
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310500603.3A Pending CN103679272A (en) | 2013-10-12 | 2013-10-12 | Vector graph based construction method for binary-decision-tree expert knowledge base |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103679272A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108664644A (en) * | 2018-05-16 | 2018-10-16 | 微梦创科网络科技（中国）有限公司 | A kind of question answering system construction method, question and answer processing method and processing device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20010025283A1 (en) * | 2000-03-18 | 2001-09-27 | Sim Dong Gyu | Apparatus and method for vector descriptor representation and multimedia data retrieval |
| CN101923552A (en) * | 2009-12-31 | 2010-12-22 | 华南师范大学 | Method for quickly superposing polygon vector layers |
-
2013
- 2013-10-12 CN CN201310500603.3A patent/CN103679272A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20010025283A1 (en) * | 2000-03-18 | 2001-09-27 | Sim Dong Gyu | Apparatus and method for vector descriptor representation and multimedia data retrieval |
| CN101923552A (en) * | 2009-12-31 | 2010-12-22 | 华南师范大学 | Method for quickly superposing polygon vector layers |
Non-Patent Citations (1)
| Title |
|---|
| 宋斌；方葛丰；刘毅: "基于矢量图形的专家知识库生成技术", 《电子测量技术》, vol. 35, no. 9, 30 September 2012 (2012-09-30) * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108664644A (en) * | 2018-05-16 | 2018-10-16 | 微梦创科网络科技（中国）有限公司 | A kind of question answering system construction method, question and answer processing method and processing device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhuo et al. | Learning complex action models with quantifiers and logical implications | |
| CN102930023B (en) | Knowledge based engineering data quality solution | |
| CN110750649A (en) | Knowledge graph construction and intelligent response method, device, equipment and storage medium | |
| KR20160058947A (en) | Evaluating rules applied to data | |
| CN102270137B (en) | Method for acquiring ADL (architecture description language) and modeling tool | |
| US20160188299A1 (en) | System And Method For Automatic Extraction Of Software Design From Requirements | |
| CN108664241A (en) | A method of SysML models are subjected to simulating, verifying | |
| CN105930447A (en) | Method for converting tree-like nested data into plane data table | |
| Bruggink et al. | Termination analysis for graph transformation systems | |
| CN113220835A (en) | Text information processing method and device, electronic equipment and storage medium | |
| CN114327483A (en) | Graph tensor neural network model establishing method and source code semantic identification method | |
| CN102663108B (en) | Medicine corporation finding method based on parallelization label propagation algorithm for complex network model | |
| CN102708285B (en) | Coremedicine excavation method based on complex network model parallelizing PageRank algorithm | |
| US20230117325A1 (en) | System for generating compound structure representation | |
| CN103679272A (en) | Vector graph based construction method for binary-decision-tree expert knowledge base | |
| CN106383738A (en) | Task processing method and distributed computing framework | |
| CN110533162A (en) | It is a kind of to automatically generate the method and system that mapping is operated between deep learning frame | |
| CN109754087B (en) | Quantum program conversion method and device and electronic equipment | |
| Kulkarni et al. | Novel Approach to Abstract the Data Flow Diagram from Java Application Program | |
| CN107247813A (en) | A kind of network struction and evolution method based on weighting technique | |
| CN107358494A (en) | A kind of client requirement information method for digging based on big data | |
| Esmaeilpour et al. | Design pattern mining using distributed learning automata and DNA sequence alignment | |
| CN106649118A (en) | Generating method of SSA single path of Java code based on AST | |
| CN109117207B (en) | Data processing method of business process model | |
| JP6148427B1 (en) | Document classification apparatus, document classification method, and document classification program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20140326 |
|
| RJ01 | Rejection of invention patent application after publication |