CN104778543A - Business procedure monitoring method based on domain knowledge ontology - Google Patents
Business procedure monitoring method based on domain knowledge ontology Download PDFInfo
- Publication number
- CN104778543A CN104778543A CN201510160825.4A CN201510160825A CN104778543A CN 104778543 A CN104778543 A CN 104778543A CN 201510160825 A CN201510160825 A CN 201510160825A CN 104778543 A CN104778543 A CN 104778543A
- Authority
- CN
- China
- Prior art keywords
- monitoring
- domain knowledge
- operation flow
- knowledge
- business
- 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
Abstract
The invention discloses a business procedure monitoring method based on domain knowledge ontology. The method comprises steps as follows: a corresponding monitoring model is established according to knowledge, business procedure and monitoring demands of the business area; business procedure data are acquired in real time, and the monitoring model is used for monitoring. According to a monitoring procedure of the business procedure monitoring method, a business procedure state reasoned on the basis of monitoring rules is verified, and a monitoring result is obtained; the whole monitoring procedure and the business procedure are executed simultaneously, and the executing process of the business procedure can be monitored.
Description
Technical field
The present invention relates to domain knowledge body field, further relate to operation flow and perform and monitoring technique, be specially a kind of business process monitoring method based on domain knowledge body.
Background technology
Body is as the concept coming from philosophy, because having good concept hierarchy and the support to reasoning from logic, start the research of the object, attribute and the mutual relationship that utilize ontology thought Description of Knowledge field from the researcher of 20 th century later artificial intelligence fields.Body as a kind of can at the conceptual model of Description of Knowledge system on semantic and knowledge hierarchy and modeling tool, the concern of lot of domestic and foreign scientific research personnel has just been caused since being suggested, and obtain idea in many fields of computing machine and apply widely, as the process, Semantic Web etc. of Heterogeneous Information on knowledge engineering, digital library, software repeated usage, information retrieval and Web.Body is that the Formal Specification of shared ideas model illustrates, is described the semanteme of concept by the relation between concept.It is the concept definition collection in general meaning, for interpersonal or between tissue and tissue communication provides common vocabulary, achieves knowledge sharing to a certain degree and reuses, simultaneously for machine energy automatic Understanding semanteme provides the most basic support.The introducing of body, provides a new thinking for solving knowledge sharing and reuse in knowledge system.Common body inscape comprises:
● individual (example): " bottom " the in other words object on basis.
● class: the kind of set (sets), concept, object type things in other words.
● attribute: attribute, feature, characteristic, feature and parameter that object (and class) may have.
● relation: the mode that may have associated with each other between class and individuality.
● function term: in the middle of declarative statement, can be used to the labyrinth replacing the particular kind of relationship of concrete term to form.
● constraint (restriction): take formalization mode to state, about a certain description of asserting situation about must set up as input of acceptance.
● rule: the logical deduction that can draw can be asserted according to particular form a certain, the statement of if-then (cause-consequence) formula statement form for describing.
● axiom: take the asserting of certain logic form (comprising rule) jointly to form be exactly its body respective application field when described in whole theory.This definition is different from said in the middle of production grammar and formal logic " axiom ".In the middle of these subjects, among axiom, only comprise the statement that those are asserted to priori.With regard to usage here, also comprise among " axiom " according to the statement of axiom type the theory drawn of deriving.
● event (philosophy): the change of attribute or relation.
Domain body is the special body of one for describing designated field knowledge.Which giving a kind of formalized description of characteristic that domain entities concept and mutual relationship, domain activity and this field have and rule. we think, domain body is made up of attribute, object, relation and sub-domain body.
Operation flow is a series of activities jointly completed respectively by different people for reaching specific value goal.Not only have strict sequencing to limit between activity, and the content of activity, mode, responsibility etc. also all must have clear and definite arrangement and define, between different skills for occupation, carrying out changing hands handing-over to make different activity and become possibility.Between movable and activity, transfer over time and space can have larger span.
Along with the development of computer technology, increasing enterprise appliance computer operation flow technology improves overall work efficiency, also more and more comes into one's own to the management of operation flow.Wherein, enterprise can be helped to weigh the monitoring of operation flow and analyze the performance of operation flow, finding out key business problem by data, the speed of operation flow, quality and efficiency can be improved like this.At present, the monitoring of operation flow is mainly monitored operation flow by rule, as " the operation flow efficiency real-time management method based on eca rule " that the people such as Shi Meihong issued in 2013, have employed eca rule and manages operation flow data.Adopt rule to carry out business process monitoring method and generally there is some limitation following:
(1) ability to express of rule is more weak, the logic of some more complicated that are beyond expression;
(2) data structure of supervisory system needs for specific business Process Design, and operation flow there occurs change then to be needed to change data structure, cannot carry out multiplexing.
Summary of the invention
The present invention is directed to the above-mentioned defect of prior art, provide a kind of business process monitoring method of basic area ontologies.
Based on a business process monitoring method for domain knowledge body, comprise the following steps:
(1) corresponding monitoring model is set up according to the knowledge of business scope, operation flow and monitoring demand;
(2) monitoring model of Real-time Obtaining operation flow data also described in utilization is monitored.
Business process monitoring method of the present invention needs to investigate and determines relevant knowledge and the monitoring demand of business scope to be monitored before enforcement.
Described step (1) is specific as follows:
(1-1) investigate the relevant knowledge of business scope and set up domain knowledge ontology model;
(1-2) on ontology model, monitoring rules is set up according to the demand of monitoring;
(1-3) corresponding monitoring model is set up according to operation flow.
Described domain knowledge ontology model is for defining the field concept relevant to business, and the mutual relationship between every field concept.
First the relevant knowledge of business scope is expressed as the domain knowledge ontology model of OWL form when setting up, concrete corresponding relation is as follows:
A () field concept is mapped to class (Class);
B the instantiation of () concept is mapped to individuality (Individual), between concept and the instantiation of concept, relation automatic mapping becomes is (is-a) relation;
C the subclass relation between () concept and concept, relation of inclusion are mapped to subclass relation (Subclass);
D best property of attribute mapping that () concept has becomes data attribute (Data property);
E () concept becomes object relationship (Object property) with other relationship map between concept.
Described monitoring rules is according to the field concept defined in described domain knowledge ontology model and mutual relationship, and monitoring demand is formulated.
As preferably, during the monitoring rules set up in described step (1-2), directly monitoring demand schedule is reached the SWRL rule based on the domain knowledge body set up in step (1-1).
In the present invention, monitoring demand can be split into precondition, requirement and restrictive condition three part and be expressed as many SWRL rules, every bar SWRL rule is made up of prerequisite (Antecedent) and conclusion (Consequent) two parts, prerequisite meets then can infer conclusion establishment, and what conclusion represented is concrete operation flow state.
Described step (2) comprises concrete steps:
(2-1) the operation flow data of acquisition are stored in described domain knowledge ontology model;
(2-2) according to monitoring rules, reasoning is carried out to the operation flow data be stored in described domain knowledge ontology model, the knowledge that must make new advances;
(2-3) according to monitoring flow process, the knowledge that reasoning obtains is verified, obtain monitored results.
Described operation flow data comprise Workflow messages, and the relation between operation flow data and business scope knowledge, and described Workflow messages comprises base attribute, relation between step and step.
In step (2-1), the concept in operation flow data and domain knowledge body and relation are mapped, stored in domain knowledge ontology model after mapping.Just operation flow data are stored in domain knowledge body, and concrete grammar is as follows:
A () finds corresponding class;
B () is at such lower newly-built individuality;
C () carries out assignment to the data attribute of this individuality;
(d) this individuality newly-built and other there are individual object properties;
Time in described step (2-2) according to monitoring rules reasoning:
Carry out reasoning to all operation flow data in described domain knowledge ontology model, reasoning new knowledge form is out a genuine declarative sentence forever, and statement operation flow reaches a certain specific state.
The statement operation flow that whole flow process reasoning obtains reaches a certain specific state can be called path status, and is represented by graphic form.
Monitoring model in the present invention is actually a monitoring flow process, and this monitoring flow process is made up of multiple similar monitoring sub-process, and the detailed process of each monitoring sub-process is as follows:
A (), when precondition meets, enters monitoring sub-process;
Whether (b) verification monitoring requirement meets, if met, then exits monitoring sub-process; If do not met, then make prompting;
C () be interval (need setting according to embody rule field and monitoring, can adjust) at set intervals, repeat step (b);
D (), when exceeding restrictive condition, verifies monitoring requirement for the last time, if do not met, then make warning, then exits monitoring sub-process.
Compared with prior art, tool of the present invention has the following advantages:
Monitoring flow process verifies monitoring rules reasoning operation flow state out and draw monitored results, and the implementation of monitoring flow process is parallel to the execution of operation flow, can be implemented in the monitoring in operation flow implementation;
Establish two-layer business process monitoring model, underlying model is made up of domain knowledge ontology model and corresponding monitoring rules (SWRL rule), domain knowledge ontology model is for expressing business scope knowledge and storage service flow data, monitoring rules is used for carrying out reasoning to history service flow data and obtaining the state of operation flow, upper layer model is set up according to operation flow the monitoring flow process being simultaneously parallel to again operation flow, monitoring flow process verifies underlying model reasoning operation flow state out, draws final monitored results.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of the business process monitoring based on domain knowledge body;
Fig. 2 is the domain knowledge ontology model of unstable angina clinical path;
Fig. 3 is the monitoring process flow diagram of unstable angina clinical path;
Fig. 4 is course of disease data hum patterns stored in domain knowledge ontology model after pre-service;
Fig. 5 carries out to ontology data the clinical path status information figure that reasoning obtains according to monitoring rules.
Embodiment
The present invention is explained further below in conjunction with the drawings and specific embodiments.
To be described the business process monitoring method based on domain knowledge body of unstable angina clinical path implementation status monitoring to the invention of this reality in the present embodiment.
Clinical path refers to doctor, nurse and other professionals, for certain sick kind or operation, based on evidence-based medicine EBM, to improve for the purpose of quality of medical care and guarantee of the medical safety, the sequencing having extreme duty order and correct time to require formulated, standardized diagnosis and treatment plan, to reduce rehabilitation delay and the wasting of resources, patient is made to obtain best medical treatment and nursing service.
The method for supervising process flow diagram of the present embodiment as shown in Figure 1, comprises the steps:
Step S101: the relevant knowledge in investigation unstable angina clinical path flow process and unstable angina field.
Table 1 gives the partial content of unstable angina clinical path guide.This clinical guidelines defines clearly should implement which type for the treatment of or inspection into footpath patient at the time point determined, as several stages of pointing out in table 1, as Preoperative Method stage and selection medication stage.Wherein in the Preoperative Method stage, being admitted to hospital in 0-3 days and must carrying out the inspections such as blood routine, bracket for blood grouping, routine urine examination,urine for routine.
Table 1
Step S102: according to the relevant knowledge in unstable angina field, establish domain knowledge ontology model (being called for short domain knowledge body) as shown in Figure 2, include 26 classes, 12 data attributes, 26 attributes of a relation and 3479 individualities (part does not show in fig. 2).
Wherein, class represents some groups of individuals, and the method for body type of service (mathematics) accurately describes out the condition that in such, member must have, and such as, in field, the individuality of whole patient all belongs to patient's class.Individual then be the concrete object of class, such as, Zhang San is the body one by one of patient's class.Attribute data attribute is the attribute of the numeric type that class has, such as, and the patient number of patient's class, admission number attribute.Object properties are the binary relations between individuality, and such as, attribute kit is containing the individuality of patient's class and course of disease class individuality being coupled together.
Step S103: on above-mentioned domain knowledge ontology model, according to monitoring demand, design corresponding SWRL rule, in the present embodiment, SWRL rule is as shown in table 2.
Table 2
Step S104: set up according to the flow process of unstable angina clinical path and monitor flow process (i.e. monitoring model) accordingly.
Fig. 3 gives the monitoring flow process of whole unstable angina, and whole monitoring flow process is made up of three monitoring sub-processes as seen from the figure, is the postoperative monitoring sub-process of the monitoring sub-process after being admitted to hospital, PCI and monitoring sub-process of leaving hospital respectively.Such as, the monitoring sub-process that PCI is postoperative, start to enter prompting flow process when monitoring after PCI this state postoperative meets, to in clinical path regulation carry out but not have execution diagnosis and treatment plan remind, and after monitoring that PCI is postoperative and meeting more than one day this state, start to enter warning flow process, to regulation in clinical path carry out but not have the diagnosis and treatment plan of execution to warn.
Step S105: obtain patient's course of disease data (i.e. operation flow data) in real time from electronic medical record system (EMR).Table 3 gives the part course of disease data of Real-time Obtaining from EMR.These course of disease data comprise many doctor's advice information, and every bar doctor's advice information comprises again concrete identification number, patient number, admission number, content and execution time.
Table 3
Step S106: pre-service is carried out to the course of disease data obtained, comprises Chinese word segmentation, Similarity Measure etc., defined knowledge (domain knowledge) in the content of doctor's advice and domain knowledge ontology model is mapped.
Step S107: go to explain course of disease data with domain knowledge body, obtains individual.Fig. 4 gives the individuality that the course of disease data in table 3 obtain after explaining, wherein rectangle (i.e.) represents individual, value in angle brackets (i.e. < >) represents it is the class of this individuality, the individuality of band asterisk (i.e. *) represents newly-built individuality, under individual title is data attribute and value thereof, and the line of band arrow represents object relationship.As shown in Figure 4, newly-built " patient _ 1 ", " course of disease _ 1 ", " event _ 1 ", " event _ 2 " and " date _ 2013-4-21 " five individualities and " comprising " between them, " intervention type " and " execution time " object relationships.Wherein " event _ 1 " and " event _ 2 " is set up and " to be admitted to hospital " with the individuality that defines in domain knowledge ontology model and " intervention type " object relationship of " blood routine examination ".
Step S108: the monitoring rules according to table 2 uses Tableau algorithm and forward reasoning to carry out reasoning to the course of disease data separate Pellet inference machine of patient, obtains the executing state in Present clinical path.Fig. 5 gives the clinical path status information obtained by SWRL rule-based reasoning (i.e. clinical path state).After the reasoning to " course of disease _ 1 ", set up " course of disease _ 1 " and state individuality and " be admitted to hospital " and " meeting " object relationship of " laggard promoting circulation of blood routine inspection of being admitted to hospital ", show that the clinical path state of this course of disease is be admitted to hospital and carried out blood routine examination.
Step S109: the clinical path state of monitoring flow process to Fig. 5 according to Fig. 3 verifies, and obtains monitored results.Table 5 gives and course of disease state is verified to the monitored results obtained by monitoring flow process.
Table 5
Step S110: the monitoring flow process according to Fig. 3 is monitored operation flow, judges whether operation flow terminates, if terminate, then stops monitoring (namely monitoring end); Otherwise, return step S105, proceed monitoring.
Above-described embodiment has been described in detail technical scheme of the present invention and beneficial effect; be understood that and the foregoing is only most preferred embodiment of the present invention; be not limited to the present invention; all make in spirit of the present invention any amendment, supplement and equivalent to replace, all should be included within protection scope of the present invention.
Claims (7)
1., based on a business process monitoring method for domain knowledge body, it is characterized in that, comprise the following steps:
(1) corresponding monitoring model is set up according to the knowledge of business scope, operation flow and monitoring demand;
(2) monitoring model of Real-time Obtaining operation flow data also described in utilization is monitored.
2. the business process monitoring method based on domain knowledge body according to claim 1, is characterized in that, described step (1) is specific as follows:
(1-1) investigate the relevant knowledge of business scope and set up domain knowledge ontology model;
(1-2) on ontology model, monitoring rules is set up according to the demand of monitoring;
(1-3) corresponding monitoring model is set up according to operation flow.
3. the business process monitoring method based on domain knowledge body according to claim 2, is characterized in that, described domain knowledge ontology model is for defining the field concept relevant to business, and the mutual relationship between every field concept.
4. the business process monitoring method based on domain knowledge body according to claim 2, is characterized in that, described monitoring rules is according to the field concept defined in described domain knowledge ontology model and mutual relationship, and monitoring demand is formulated.
5. according to the business process monitoring method based on domain knowledge body in Claims 1 to 4 described in any one, it is characterized in that, described step (2) comprises concrete steps and is:
(2-1) the operation flow data of acquisition are stored in described domain knowledge ontology model;
(2-2) according to monitoring rules, reasoning is carried out to the operation flow data be stored in described domain knowledge ontology model, the knowledge that must make new advances;
(2-3) according to monitoring flow process, the knowledge that reasoning obtains is verified, obtain monitored results.
6. the business process monitoring method based on domain knowledge body according to claim 5, it is characterized in that, described operation flow data comprise Workflow messages, and the relation between operation flow data and business scope knowledge, described Workflow messages comprises base attribute, relation between step and step.
7. the business process monitoring method based on domain knowledge body according to claim 6, is characterized in that, time in described step (2-2) according to monitoring rules reasoning:
Carry out reasoning to all operation flow data in described domain knowledge ontology model, reasoning new knowledge form is out a genuine declarative sentence forever, and statement operation flow reaches a certain specific state.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510160825.4A CN104778543A (en) | 2015-04-07 | 2015-04-07 | Business procedure monitoring method based on domain knowledge ontology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510160825.4A CN104778543A (en) | 2015-04-07 | 2015-04-07 | Business procedure monitoring method based on domain knowledge ontology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104778543A true CN104778543A (en) | 2015-07-15 |
Family
ID=53619994
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510160825.4A Pending CN104778543A (en) | 2015-04-07 | 2015-04-07 | Business procedure monitoring method based on domain knowledge ontology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104778543A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111583058A (en) * | 2020-03-17 | 2020-08-25 | 国网浙江省电力有限公司杭州供电公司 | Power distribution network security analysis system generation method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1967572A (en) * | 2005-11-18 | 2007-05-23 | 中国科学院软件研究所 | Method and system of business process management applied to business-intensive fields |
| US20080294485A1 (en) * | 2007-05-22 | 2008-11-27 | Takuya Takeuchi | System, method and program for supporting creating a business process |
| CN103593752A (en) * | 2013-11-29 | 2014-02-19 | 武汉大学 | Business process adaptability configuring method and system based on situation constraint |
-
2015
- 2015-04-07 CN CN201510160825.4A patent/CN104778543A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1967572A (en) * | 2005-11-18 | 2007-05-23 | 中国科学院软件研究所 | Method and system of business process management applied to business-intensive fields |
| US20080294485A1 (en) * | 2007-05-22 | 2008-11-27 | Takuya Takeuchi | System, method and program for supporting creating a business process |
| CN103593752A (en) * | 2013-11-29 | 2014-02-19 | 武汉大学 | Business process adaptability configuring method and system based on situation constraint |
Non-Patent Citations (1)
| Title |
|---|
| JUNJIAN LIU 等: "An ontology-based real-time monitoring approach to clinical pathway", 《2014 7TH INTERNATIONAL CONFERENCE ON BIOMEDICAL ENGINEERING AND INFORMATICS》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111583058A (en) * | 2020-03-17 | 2020-08-25 | 国网浙江省电力有限公司杭州供电公司 | Power distribution network security analysis system generation method and device |
| CN111583058B (en) * | 2020-03-17 | 2023-09-05 | 国网浙江省电力有限公司杭州供电公司 | Method and device for generating power distribution network safety analysis system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Dave et al. | Explainable ai meets healthcare: A study on heart disease dataset | |
| Abidi | Healthcare knowledge management: The art of the possible | |
| Esmaeilzadeh et al. | Patients’ perceptions toward human–artificial intelligence interaction in health care: experimental study | |
| CN109920508A (en) | prescription auditing method and system | |
| Mertens et al. | Towards a decision-aware declarative process modeling language for knowledge-intensive processes | |
| Duma et al. | An ad hoc process mining approach to discover patient paths of an Emergency Department | |
| Fox et al. | A canonical theory of dynamic decision-making | |
| Schulz et al. | An ontological analysis of reference in health record statements. | |
| Hurley et al. | Ontology engineering to model clinical pathways: Towards the computerization and execution of clinical pathways | |
| Russo et al. | A business process model for integrated home care | |
| Liu et al. | Cross-department collaborative healthcare process model discovery from event logs | |
| US11302441B2 (en) | Patient treatment resource utilization predictor | |
| van Damme et al. | From lexical regularities to axiomatic patterns for the quality assurance of biomedical terminologies and ontologies | |
| Lidströmer et al. | Introductory approaches for applying artificial intelligence in clinical medicine | |
| CN104778543A (en) | Business procedure monitoring method based on domain knowledge ontology | |
| Sene et al. | Data mining for decision support with uncertainty on the airplane | |
| Lagioia et al. | The strange case of Dr. Watson: liability implications of AI evidence-based decision support systems in health care | |
| Mathew et al. | Distributed privacy preserving decision support system for predicting hospitalization risk in hospitals with insufficient data | |
| Manzoor et al. | Ontology-based clinical decision support system for predicting high-risk pregnant woman | |
| Abidi | A conceptual framework for ontology based automating and merging of clinical pathways of comorbidities | |
| Dantu et al. | Medical process modeling with a hybrid system dynamics Zachman Framework | |
| Iheme et al. | Mobile-bayesian diagnostic system for childhood infectious diseases | |
| Wojtusiak | Semantic data types in machine learning from healthcare data | |
| de Bruin et al. | Medical fuzzy control systems with fuzzy arden syntax | |
| Shaheen et al. | Argumentation-based explanations of multimorbidity treatment plans |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate 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: 20150715 |