CN102270140B

CN102270140B - Time-constrained Web service flow mining method

Info

Publication number: CN102270140B
Application number: CN 201110235664
Authority: CN
Inventors: 马应龙; 阎光伟; 张金龙
Original assignee: North China Electric Power University
Current assignee: North China Electric Power University
Priority date: 2011-08-17
Filing date: 2011-08-17
Publication date: 2013-10-30
Anticipated expiration: 2031-08-17
Also published as: CN102270140A

Abstract

The invention discloses a time-constrained Web service flow mining method in the technical field of Web service flow mining. The method comprises the steps of first establishing graphic descriptions of time-constrained Web service flows for all the candidate Web service flows and requested Web service flows in a Web flow library; subsequently, calculating task compatibility values between the candidate Web service flows and the requested Web service flows; afterwards, respectively establishing a standardized matrix for the graphic description of one of the requested Web service flows and the graphic description of each candidate Web service flow which is compatible in task with the requested Web service flow, and calculating a difference value between the graphic description of the requested Web service flow and the graphic description of each candidate Web service flow which is compatible in task with the requested Web service flow; and finally, selecting the graphic description of the candidate Web service flow which has the minimum difference value with the graphic description of the requested Web service flow, and reusing the Web service flow corresponding to the graphic description in the Web flow library. The time-constrained Web service flow mining method can ensure that the mined service flow can meet the demand of a basic task requested by clients and the desired time-constrained demand of users.

Description

Web service flow mining method constrained by time

Technical Field

The invention belongs to the technical field of Web service flow mining, and particularly relates to a time-constrained Web service flow mining method.

Background

With the continuous development of Service-Oriented Architecture (SOA) and technology, Web services have played an increasingly important role in the fields of e-commerce, enterprise application integration, etc. Web services are externally available distributed application components accessed through the Web. Through the Web service combination technology, enterprises can combine the existing relatively simple Web services into a Web service flow according to certain business flow logic. In order to meet the demands of complex and variable and intense market competition, on one hand, when an enterprise executes a specific Web service process, all departments inside the enterprise must cooperate with each other as effectively as possible to meet the requirements of service quality and time constraints desired by enterprise customers, so that the enterprise can stand in the intense market competition without fail. Therefore, the various Web service flows provided by enterprises for different business needs are often time-constrained. This is particularly true for e-commerce enterprises. On the other hand, the enterprise application system must be able to implement the business process as quickly and flexibly as possible according to the business requirements, so as to save the service cost and improve the benefits. Therefore, enterprises need to adopt a process mining technology to find out a Web service process capable of meeting the current latest user requirements from the existing Web service processes constrained by time. By realizing the repeated use of the Web service flow, the service cost is saved as much as possible, and the value added of the Web service flow is realized.

Description languages for describing the Web service flow are BPEL4WS, WSFL, XLANG, WSCI, WSCL, BPML, etc., which have syntax of XML language and are close to each other in description capability. Many systems and methods supporting Web service process modeling have been developed and designed over the years both at home and abroad. Representative systems include the eblow system developed by the HP laboratory and the SERV system developed by the university of new south williams, australia. The former supports graphical flow modeling and dynamic service composition, and the latter uses state diagrams to model service flow logic. However, current Web service flow research and applications focus primarily on service flow modeling, verification, simulation, and execution based on workflow techniques, rather than mining of service flows. The current tools and methods for process mining, such as the ProM process mining framework and the DAGAMA process discovery tool, mainly aim to ensure that the mined service processes can meet the basic task requirements required by the customer, but cannot ensure that the mined service processes can meet the requirements of efficient services desired by the user, that is, the time constraint requirements desired by the customer. This is an important issue to be solved for enterprises such as e-commerce, etc., which have the quality of service and efficiency as their survival bases.

Disclosure of Invention

Aiming at the defects that the current process mining technology cannot meet the requirements of users and the like in the background technology, the invention provides a time-constrained Web service process mining method.

The technical scheme of the invention is that a Web service flow mining method restricted by time is characterized by comprising the following steps:

step 1: establishing graph description of the Web service flow which is restricted by time for all the Web service flows and the request Web service flow in the Web flow library;

step 2: calculating a task compatibility value of the description of the request Web service flow diagram and the description of each candidate Web service flow diagram, and when the compatibility value is greater than a specified threshold value, the tasks of the request Web service flow diagram and the tasks of the candidate Web service flow diagrams are compatible;

and step 3: on the basis of the step 2, respectively establishing a standardized matrix for the description of the request Web service flow chart and the description of each candidate Web service flow chart compatible with the task of the request Web service flow chart;

and 4, step 4: calculating a difference value between the request Web service flowchart description and each candidate Web service flowchart description which is task-compatible with the request Web service flowchart description based on the standardized matrix;

and 5: and based on the difference value, selecting the candidate Web service flow diagram description with the minimum difference value with the request Web service flow diagram description, and repeatedly using the Web service flow corresponding to the flow diagram description in the Web flow library.

The graph description of the Web service flow can be determined by a set of nodes and a set of edges, wherein the nodes are Web service names in the service flow, and the edges are directed edges with time constraint marks between the two nodes.

The calculation formula of the task compatibility value is as follows:

taskcomp ({R, Q}_{i}) = \frac{| V_{R} \cap V_{Q_{i}} |}{| V_{R} \cup V_{Q_{i}} |}

wherein:

r is the description of a request Web service flow chart;

Q_ia description of the ith candidate Web service flow diagram;

taskcomp(R，Q_i) Is R and Q_iA task compatibility value of;

V_Ris the node set contained in R;

is Q_iThe set of nodes contained in (1).

The calculation formula of the standardized matrix described by the request Web service flow chart is as follows:

wherein:

m (i, j) is an element of a standardized matrix for requesting a Web service flowchart description;

v_iis R and Q_iThe ith node in the union of the contained Web service name sets;

v_jis R and Q_iThe j-th node in the union of the included sets of Web service names.

The calculation formula of the standardized matrix described by the candidate Web service flow chart is as follows:

wherein:

m' (i, j) is an element of a standardized matrix described by the candidate Web service flow diagram;

v_iis R and Q_rThe ith node in the union of the contained Web service name sets;

v_jis R and Q_rThe j-th node in the union of the contained Web service name sets;

[a，b]indicating edge (v)_i，v_j) A corresponding time constraint marker in R;

[a′，b′]indicating edge (v)_i，v_j) At Q_rThe corresponding time constraint flag in (1);

[ a, b ] -n [ a ', b' ] represents the portion of the time constraint that [ a, b ] and [ a ', b' ] overlap in time

Note, for example, [2,5] "d [3,7] ═ 3,5 ];

dur ([ a, b ]) represents the interval time value of the time stamp [ a, b ], i.e., Dur ([ a, b ]) is b-a;

if [ a, b ] -n [ a ', b' ] has no overlapping part in time, then

Dur([a，b]∩[a′，b′])＝0。

The calculation formula of the difference value is as follows:

d_r＝DiagonalSum((M-M′)×(M-M′)^T)

wherein:

d_ris a difference value;

M-M 'is a matrix obtained by subtracting corresponding element values of the matrix M and the matrix M';

(M-M′)^Trepresenting the transpose of matrix M-M'.

The method and the system can ensure that the excavated service flow can meet the basic task requirement required by the client and can also ensure that the excavated service flow can meet the time constraint requirement expected by the user.

Drawings

FIG. 1 is a schematic diagram of a time-constrained Web services flow mining algorithm;

FIG. 2 is a pictorial depiction of a request Web service flow and 2 candidate Web service flows;

FIG. 3 is a matrix difference calculation for a requested Web service flow and a1 st candidate Web service flow;

FIG. 4 is a matrix difference calculation for the requested Web service flow and the 2 nd candidate Web service flow.

Detailed Description

The preferred embodiments will be described in detail below with reference to the accompanying drawings. It should be emphasized that the following description is merely exemplary in nature and is not intended to limit the scope of the invention or its application.

The invention aims to provide a Web service flow mining method subject to time constraint aiming at the defects of the existing Web service flow mining technology, so that the mined service flow can meet the basic task requirement required by a client and the time constraint requirement expected by the client, thereby promoting the effective cooperative work among all business related departments in an enterprise, saving the service operation cost of the enterprise and improving the economic benefit of the enterprise under the condition of finishing the client requirement with high quality.

The technical scheme of the invention is a time-constrained Web service flow mining method. The method comprises the steps of converting Web service flows into a graph description form, further converting the graph description form into a corresponding standardized flow matrix for calculating the difference degree, and further selecting the Web service flows which most possibly meet customer requirements for reuse according to the difference value among the Web service flows.

The invention provides a time-constrained Web service flow mining method, which mainly comprises the following steps:

1. and establishing a graph description of the time-constrained Web service flow for all the Web service flows in the Web flow library and the request Web service flow.

The graph description of a Web service flow can be determined through a node set and an edge set, wherein the nodes described by the graph are Web service names in the service flow, the edges described by the graph are directed edges between two nodes and used for representing the activation relation between the Web services corresponding to the two nodes, time constraint marks in the form of [ a, b ] on the directed edges indicate that the starting Web service needs to be completed within a to b time units to further activate the tail Web service, and a is less than or equal to b. The node set and the edge set of the graph description may be stored by two database tables, respectively.

The method specifically comprises the following substeps:

1.1 the request Web service flow is represented directly using the graph description. The request Web service flow is an ideal Web service flow formulated for given customer requirements, including basic task requirements and time constraint requirements.

1.2 for all candidate Web service flows described by a flow description language in a Web service flow library, extracting all Web service names and service activation relations from the Web service flow description by adopting a currently common XML parser tool, and extracting a time constraint relation between service activations according to a system record log executed before the Web service flow, thereby establishing and storing a graph description for each candidate Web service flow. By Q₀，Q₁，…Q_m-1A diagram depiction of all candidate Web service flows.

2. Compute request Web service flow graph description R and each candidate Web service flow graph description Q₀，Q₁，…Q_m-1A task compatibility value of.

The method specifically comprises the following substeps:

2.1 set the initial value of counter i to 0.

2.2 the formula for calculating the task compatibility value taskomp is:

taskcomp ({R, Q}_{i}) = \frac{| V_{R} \cap V_{Q_{i}} |}{| V_{R} \cup V_{Q_{i}} |}

wherein:

r is a request Web service flowchart description;

Q_iis the ith candidate Web service flowchart description (0 ≦ i ≦ m-1);

taskcomp(R，Q_i) Is R and Q_iA task compatibility value of;

V_Rand

respectively represent R and Q_iThe set of nodes contained in (1).

The task compatibility threshold was set to 0.6. Task compatibility is mainly to verify two service flowsThe ratio of the same service name that the program contains. If R and Q_iTaskocomp values of greater than or equal to this threshold indicate that they are task compatible; otherwise, the task is incompatible.

2.3 if R and Q_iIs task compatible, then Q will be_iAnd adding the task-consistent Web service flow chart description queue.

2.4 if i < m-1 at this time, let i become i +1, then go to execute sub 2.2, select the next candidate Web service flow Q from the Web service flow library_iTask compatibility comparisons were made with R. Otherwise, if i is larger than or equal to m-1, the compatibility calculation described by the flow chart is stopped.

3. A standardized matrix is established for each of the requested Web service flowchart descriptions and each of the candidate Web service flowchart descriptions that are task-compliant.

The service flow diagram description in the request Web service flow diagram description R task-compliant Web service flow diagram description queue is denoted as Q₀，Q₁，…Q_k-1，0≤k≤m。

R and Q_rThe normalized matrices (0. ltoreq. r. ltoreq. k-1) are denoted by M and M', respectively, which are n × n matrices. Wherein,

namely R and Q_rThe row and column elements in the normalized matrix of (1) are from R and Q_rThe union of node sets of (1). The n elements in the union are denoted v₀，v₁，…v_n-1。

The method specifically comprises the following substeps:

3.1 building a standardized matrix requesting a Web service flow graph description R based on the Web service flow graph description.

The normalized matrix for the request Web services flow diagram description R is denoted as M, where the value of each element M (i, j) is calculated as follows:

wherein:

v_iand v_jRepresents R and Q_rThe union of the contained sets of Web service names (i.e., the

) I and j satisfy the condition 0 ≤ i, j ≤ n-1, and

i.e. the cardinality of the union.

3.2 establishing a standardized matrix for each candidate Web service flow diagram description based on the diagram description of the Web service flow.

Candidate Web service flow Q_rThe normalized matrix depicted in the figure is denoted M ', where the value of each element M' (i, j) is calculated as follows:

wherein:

v_iand v_jRespectively represent R and Q_rThe ith node and the jth node in the union of the contained Web service name sets, wherein i and j meet the condition that i is more than or equal to 0 and j is more than or equal to n-1;

[ a, b ] "d [ a ', b' ] represents the portion of the time constraint marker where [ a, b ] and [ a ', b' ] overlap in time, such as [2,5 ]" d [3,7] ═ 3,5 ];

dur ([ a, b ]) represents the interval time value of the time stamp [ a, b ], i.e., Dur ([ a, b ]) is b-a.

If [ a, b ] -n [ a ', b' ] has no overlapping part in time, then

Dur([a，b]∩[a′，b′])＝0。

4. Based on the normalized matrix, a difference value is calculated between the requested Web service flowchart description and each candidate Web service flowchart description that is task-compliant.

4.1 set counter r to 0.

4.2 calculation of R and Q_rDifference value d between_rDifference value d_rThe value can be calculated by the following formula:

d_r＝DiagonalSum((M-M′)×(M-M′)^T)

wherein:

M-M 'represents a matrix obtained by subtracting corresponding element values of the matrix M and the matrix M'; the matrix multiplication is carried out, and the result is still a matrix;

(M-M′)^Ta transposed matrix representing the matrix M-M';

diagnonalsum is used to compute the sum of the diagonal element values of the matrix, i.e., R and Q_rThe difference value between, i.e. d_r。

4.3 if r < k-1, then r < r +1, and 4.2. go to select the next candidate flow Q from the Web service flow library_rAnd performing task compatibility calculation and difference degree calculation with the R.

5. And based on the difference value, selecting the candidate Web service flow diagram description with the minimum difference value with the request Web service flow diagram description, and reusing the corresponding Web service flow described by the candidate Web service flow diagram in the Web flow library.

All candidate Web service flow chart descriptions compatible with the R task have the difference value recorded as d₀，d₁，…d_k-1. The candidate Web service flow most similar to R is marked as Q_xWherein the subscript x passes through d_xTo determine, the formula is as follows:

d_x＝min{d₀，d₁，…d_k-1}

that is, the candidate Web service flow corresponding to the smallest value among all the discrepancy values will be the reusable service flow most similar to the requested service.

The method for mining the Web service flows constrained by time is characterized in that the candidate Web service flows with the minimum difference with the requested Web service flows are finally mined for flow reuse by converting the Web service flows into the graph description of the Web service flows and comparing the difference between the flow matrix of the requested Web service flows and the flow matrix of the candidate Web service flows. The method provided by the invention is in line with the basic cognition of human, the greater the similarity is, the smaller the difference is. The operations of extracting process information, calculating a correlation matrix and the like related to the method can be realized by some mainstream software development tools at present, and the method also has lower calculation complexity, wherein the complexity in the worst case is O (m × n)³) Wherein m is the total number of candidate Web service flows in the flow library, and n is the total number of Web services in the flow library. The application of the method can achieve the expected target, not only can ensure that the excavated service flow can meet the basic task requirement required by a client, but also can ensure that the excavated service flow can meet the time constraint requirement expected by the user.

The present embodiment includes two major aspects:

1. generation of graph descriptions of time-constrained Web service flows

Reusable Web service flows are typically stored in a service flow library via a Web service flow description file. Taking BPEL service description language as an example, a Web service flow file is stored in an XML format file, which includes all the Web service names and the mutual activation relations between the Web services involved in the flow. Therefore, a graph description of the service flow can be generated by parsing a Web service flow file described by BPEL through an XML parser. There are many XML parsers currently available, such as IBM's XML4J, Oracle's XML Parser for Java, and others.

The method comprises the following specific steps:

1.1 sequentially analyzing each Web service flow file in the Web flow library, and extracting all Web service names and Web service activation relations from the Web service flows according to the structured activity description of the service flow description language. The structured activity description specifies the name of the Web service activity through a < case > tag, AND indicates the conditional relations such as sequence, OR-split, AND-split, circulation AND the like among different service activities through tags such as < sequence >, < switch >, < flow >, < loop > AND the like. 1.1 can be divided into the following sub-steps:

1.1.1 extract all Web service (campaign) names from the structured campaign description by means of < case > tags. All extracted service names are stored in a 1-dimensional array, and each element in the array is a service name.

1.1.2 extract all the activation relationships between service activities from the structured activity description through the conditional relationship tags described above. The relationships between all service activities are stored in a 4-dimensional array. The first two components of each column vector of the 4-dimensional array represent the activation relationship between two service activities.

1.2 extracting the time constraint relation between service activation according to the system record log executed before the Web service flow. The system log stores the start time and the end time of the execution of the Web service in the flow. For a time constraint [ a, b ] of a service activation relationship edge, a equals the end time of the start service of the edge minus the start time, and b equals the start time of the service at the end of the edge minus the start time of the start service. Further, the values a and b are stored in the last two components of the column vector corresponding to the activation relationship in the 4-dimensional array, respectively.

1.3A graph description of a Web service flow can be determined by saving a 1-dimensional array of Web service names and saving a 4-dimensional array of service activation relationships with time constraints. FIG. 2a is a flowchart depiction of a request for Web services, denoted R, of an embodiment of the present invention. FIGS. 2b and 2c are pictorial depictions of 2 candidate Web service flows, denoted Q, respectively₀And Q₁. They can be determined by a 1-dimensional array and a 4-dimensional array, respectively. For example, R is stored as follows:

the Web service name 1-dimensional array RNode [ ] [ { a2, A3, a4, A5, A6, a7, a9, a10} of R, and the activation relationship 4-dimensional array RRel [ ] [4] { { a2, A3, 2, 3}, { a2, a4, 2, 5}, { A3, A5, 4, 6}, { A3, A6, 3, 4}, { A5, a9, 3, 4}, { A6, a9, 2, 3}, { a4, a7, 3, 4}, { a7, a10, 4, 6}, { a9, a10, 2, 7} }. And related storage arrays Q0Node, Q0Rel, Q1Node, Q1Rel and the like of Q0 and Q1 can be correspondingly established.

Wherein:

Q0Node[]＝{A1，A2，A3，A4，A5，A7，A8，A9，A10，A11}；

Q0Rel[][4]＝{{A1，A2，3，5}，{A2，A3，2，4}，{A2，A4，3，5}，{A3，A5，4，6}，{A4，A7，3，5}，{A4，A8，6，9}，{A5，A9，3，5}，{A1，A2，3，5}，{A7，A11，4，5}，{A8，A11，2，3}，{A9，A10，3，6}，{A11，A10，2，3}}；

Q1Node[]＝{A1，A2，A3，A4，A6，A7，A8，A9，A10}；

Q1Rel[][4]＝{{A1，A2，2，3}，{A2，A3，2，4}，{A2，A4，2，4}，{A2，A8，4，6}，{A3，A6，3，5}，{A4，A7，3，4}，{A6，A9，2，3}，{A9，A10，2，6}，{A8，A10，5，7}，{A7，A10，6，8}}。

2. web-based garmentThe service flow diagram describes that candidate Web service flows which are most similar to the request Web service flow R are mined from a service flow library. This embodiment will use Q₀And Q₁From Q as a graph description of candidate Web service flows₀And Q₁Find the Web service flow most similar to R. The method comprises the following specific steps:

2.1 calculate the task compatibility value taskomp for R and all candidate Web service flows. The method comprises the following substeps:

2.1.1 reaction of Q₀And Q₁Join Web flow description queue, when m is 2. Setting a counter i with an initial value of 0;

2.1.2 calculate taskocomp (R, Q)_i) A value of (d);

2.1.3 if R and Q_iIs not less than 0.6, then Q is_iAdding a queue described by the task-consistent Web service flow diagram;

2.1.4 if i < m-1, let i ═ i +1, then return to 2.1.2; otherwise, executing 2.2;

in the present embodiment, R and Q are obtained by the above-mentioned substeps₀The compatibility values of (A) are:

taskcomp(R，Q₀)＝7/11＝0.636≥0.6

r and Q₁The compatibility values of (A) are:

taskcomp(R，Q₁)＝7/10＝0.7≥0.6

thus, Q₀And Q₁Are added to the queue described by the task-consistent Web service flow diagram.

2.2 calculate the difference value of R and all candidate task-compatible Web service flowchart descriptions. The method comprises the following substeps:

2.2.1 for counting process convenience, naming in order the flow diagram description in the candidate Web service flow queue compatible with the R task: q₀，Q₁，…Q_k-1And a counter r is set to an initial value of 0. In the task-consistent service flow queue in this embodiment, k is 2, and Q is also included in the queue₀And Q₁；

2.2.2 establishing Q_rThe normalized matrix M';

2.2.3 for Q_rEstablishing a standardized matrix M of R;

2.2.4 calculation of d_rAnd d is equal to_rAdding a difference value queue;

2.2.5 if r < k-1, then the counter r is incremented by 1 and then returned to 2.2.2 to continue the computation of the disparity for the next task-compatible Web service flow. Otherwise, then go to 2.3;

in this embodiment, R and Q are given according to the aforementioned definition of the standardized matrix for generating the service flow₀The normalization matrices for the disparity calculation are shown in fig. 3a and fig. 3b, respectively. Also gives R and Q₁The normalization matrices for the disparity calculation are shown in fig. 4a and 4b, respectively.

In computing R and each Web service flow Q that is task-compatible therewith_rTwo auxiliary matrices are also generated during the process of the variance value of (2). One is a matrix obtained by subtracting the corresponding elements of the matrix M 'from the matrix M', and is denoted as matrix M "for convenience. The other auxiliary matrix is a matrix obtained by matrix multiplication of the transposed matrix of M "and M", and is denoted as M' "for convenience. In this embodiment, Q₀The corresponding two auxiliary matrices M 'and M' ″ are shown in FIG. 3c and FIG. 3d, Q₁The corresponding two auxiliary matrices M "and M'" are shown in fig. 4c and 4d, respectively.

Calculating d₀Is measured by dividing Q by the difference value of₀The sum of the values of the diagonal elements of the M' ″ matrix (i.e., the gray-labeled elements of the matrix) to obtain d₀8.27. Similarly, d₁Respectively is given a value of₁The sum of the values of the diagonal elements in the M' ″ matrix of (2) to obtaind₁＝5.4；

2.3 queue d from disparity values₀，d₁，…d_k-1Selecting a difference value d with the smallest difference value_x(ii) a Then, Q_xIt is a Web service flow diagram description that is compatible with the most similar task of R. The sub-steps of specifically selecting the minimum variance value are as follows:

2.3.1 setting a counter q, wherein the initial value is 1; setting a variable x, which is initially equal to 0;

2.3.3 if d_q＜d_xThen the value of q is assigned to x;

2.3.4 if q < k-1, let q ═ q +1, return to 2.3.3; otherwise, ending.

In this example, the disparity value queue contains only two disparity values, d₀，d₁. After passing execution 2.3, d_xThe integer variable x in (1) stores the number of the service flow with the smallest disparity value in the disparity value queue. In this example, d₁Is the smallest, so the value of x is 1. Thus, in the Web service flow library, the task-compliant Web service flow graph description Q₁The corresponding candidate Web service flow is most similar to the requested Web service flow. Therefore, the candidate Web service flow corresponding to the smallest value among all the discrepancy values will be the reusable service flow most similar to the requested service.

The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A Web service flow mining method restricted by time is characterized in that the method comprises the following steps:

step 1: establishing graph description of the Web service flow which is restricted by time for all the Web service flows and the request Web service flow in the Web flow library; the graph description of the Web service flow can be determined by a set of nodes and a set of edges, wherein the nodes are Web service names in the service flow, and the edges are directed edges with time constraint marks between the two nodes;

the process of establishing the graph description of the time-constrained Web service flow is as follows:

step 101: representing a request Web service flow directly using a graph description;

step 102: for all candidate Web service flows described by a flow description language in a Web service flow library, extracting all Web service names and service activation relations from the Web service flow description by adopting an XML (extensive Makeup language) parser tool, and extracting a time constraint relation between service activations according to a system record log executed before the Web service flow, thereby establishing and storing a graph description for each candidate Web service flow; by Q₀,Q₁,···Q_m-1A graph description representing all candidate Web service flows, wherein m is the total number of all candidate Web service flows in the Web flow library;

step 2: calculating a task compatibility value of the description of the request Web service flow diagram and the description of each candidate Web service flow diagram, and when the compatibility value is greater than a specified threshold value, the tasks of the request Web service flow diagram and the tasks of the candidate Web service flow diagrams are compatible; the specific process is as follows:

step 201: setting the initial value of a counter i as 0;

step 202: the calculation formula of the task compatibility value taskomp is as follows:

taskcomp (R, Q_{i}) = \frac{| V_{R} \cap V_{Q_{i}} |}{| V_{R} \cup V_{Q_{i}} |}

wherein:

r is a request Web service flowchart description;

Q_iis the description of the ith candidate Web service flow chart, and i is more than or equal to 0 and less than or equal to m-1;

taskcomp(R,Q_i) Is R and Q_iA task compatibility value of;

V_Rand V_QiRespectively representing the node sets contained in R and Qi;

setting a task compatibility threshold if R and Q_iIf the taskomp value is greater than or equal to the set threshold, indicating that they are task-consistent, go to step 203; otherwise, the task is incompatible, execute step 204;

step 203: will Q_iAdding a task-consistent Web service flow chart description queue;

step 204: if at this time i<m-1, then let i = i +1 and go to sub-step 202 to select the next candidate Web service flow Q from the Web service flow library_iComparing task compatibility with R; otherwise, if i is larger than or equal to m-1, stopping the compatibility calculation described by the flow chart, and executing the step 3;

and step 3: on the basis of the step 2, respectively establishing a standardized matrix for the description of the request Web service flow chart and the description of each candidate Web service flow chart compatible with the task of the request Web service flow chart; the method comprises the following specific steps:

step 301: the service flow diagram description in the request Web service flow diagram description R task-compliant Web service flow diagram description queue is denoted as Q₀,Q₁,···Q_k-1，0≤k≤m；

Step 302: establishing a standardized matrix requesting a Web service flow diagram description R based on the diagram description of the Web service flow;

wherein:

v_iand v_jRepresents R and Q_rThe union of the contained sets of Web service names (i.e., V)_R∪V_Qr) I and j satisfy the conditions that i is greater than or equal to 0 and less than or equal to n-1, j is greater than or equal to 0 and less than or equal to n-1, and n = | V_R∪V_QrI, the cardinality of the union;

step 303: establishing a standardized matrix for each candidate Web service flow diagram description based on the diagram description of the Web service flow;

wherein:

v_iand v_jRespectively represent R and Q_rIth node in a union of the contained sets of Web service names

The point and the jth node, wherein i and j meet the condition that i is more than or equal to 0 and less than or equal to n-1, and j is more than or equal to 0 and less than or equal to n-1;

[a,b]indicating edge (v)_i,v_j) A corresponding time constraint marker in R;

[a′,b′]indicating edge (v)_i,v_j) At Q_rThe corresponding time constraint flag in (1);

Note, for example, [2,5] andgate [3,7] = [3,5 ];

dur ([ a, b ]) represents the interval time value of the time stamp [ a, b ], i.e., Dur ([ a, b ]) = b-a; dur ([ a, b ] # andgate [ a ', b') =0 if [ a, b ] andgate [ a ', b' ] has no overlapping portion in time;

and 4, step 4: calculating a difference value between the request Web service flowchart description and each candidate Web service flowchart description which is task-compatible with the request Web service flowchart description based on the standardized matrix; the method comprises the following specific steps:

step 401: setting the counter r to be 0;

step 402: calculating R and Q_rDifference value d between_rDifference value d_rThe value can be calculated by the following formula:

d_r=DiagonalSum((M-M′)×(M-M′)^T)

wherein:

M-M' represents the subtraction of the corresponding element values of matrix M and matrix M

A matrix; the matrix multiplication is carried out, and the result is still a matrix;

(M-M′)^Ta transposed matrix representing the matrix M-M';

diagnonalsum is used to compute the sum of the diagonal element values of the matrix, which is the R sum

Q_rThe difference value between, i.e. d_r；

Step 403: if r is<k-1, then r<r +1, go to step 402, select the next candidate flow Q from the Web service flow library_rPerforming task compatibility calculation and difference calculation with the R; otherwise, executing step 5;

and 5: based on the difference value, selecting the candidate Web service flow diagram description with the minimum difference value with the request Web service flow diagram description, and repeatedly using the Web service flow corresponding to the flow diagram description in a Web flow library; the method comprises the following specific steps:

step 501: all candidate Web service flow chart descriptions compatible with the R task are recorded as the difference value of R₀,d₁,···d_k-1(ii) a The candidate Web service flow most similar to R is marked as Q_xWherein the subscript x passes through d_xTo determine, the formula is as follows:

d_x=min{d₀,d₁,···d_k-1}

step 502: and selecting the candidate Web service flow corresponding to the minimum value in all the difference values as the reusable service flow which is most similar to the request service.