CN101930369A - Task immigration-orientated component semantic matching method - Google Patents
Task immigration-orientated component semantic matching method Download PDFInfo
- Publication number
- CN101930369A CN101930369A CN 201010276348 CN201010276348A CN101930369A CN 101930369 A CN101930369 A CN 101930369A CN 201010276348 CN201010276348 CN 201010276348 CN 201010276348 A CN201010276348 A CN 201010276348A CN 101930369 A CN101930369 A CN 101930369A
- Authority
- CN
- China
- Prior art keywords
- assembly
- task
- satisfaction
- input
- component
- 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.)
- Granted
Links
Images
Abstract
The invention discloses a task immigration-orientated component semantic matching method, which comprises the following steps of: (1) performing semantic description on tasks, components and equipment by using a body network semantic description language; (2) filtering the candidate components by using a component filter; (3) calculating function satisfaction and service quality satisfaction of the candidate components by using a component matcher; (4) calculating the comprehensive satisfaction of the candidate components; and (5) outputting the components with highest comprehensive satisfaction. The method ensures seamlessness and continuity of user task immigration, reduces the participation of a user in a component selection process, and improves the satisfaction of the user.
Description
Technical field
The present invention relates to field of computer data processing, relate in particular to a kind of based on semanteme, take all factors into consideration assembly function satisfaction and service quality satisfaction assembly matching process as the assembly choice criteria.
Background technology
Along with the development of computing machine and wireless network, people no longer have been satisfied with the desktop computer of fixing in fixing place use and have carried out work, study and amusement, and are also more and more big for the demand of mobile working.Because to the demand of mobile working, user's task just can not be limited in fixing place and the hardware device and finish, we often need the migration of task.So-called task immigration, working as a task exactly is suspended on a computing machine or portable set, when we come the another one place, still this task recovery can be continued operation when using the another one computing terminal, and not need the user to repeat to do the task part of having done before.
Because the isomerism of general calculation entironment software and hardware, same task often needs to obtain different assemblies and carries out in different environment, therefore need an assembly matching mechanisms to come in new equipment, to reselect assembly for task, and, make task by seamless migration continuously with its portfolio restructuring task.And traditional assembly matching algorithm is based on the coupling of template, and the limitation of this matching process is low, the poor expandability of matching accuracy rate, and the assembly that obviously can not satisfy in the actual task transition process is selected.Existing component semantic matching algorithm adopts the assembly choice mechanism based on task model, though this method has matching accuracy rate preferably, user's participation is higher.
Summary of the invention
The objective of the invention is in task immigration, to guarantee the seamlessness and the continuity of user task migration, and minimizing user's the participation in the assembly selection course, improve user's satisfaction, and provide a kind of based on function satisfaction semanteme, the COMPREHENSIVE CALCULATING assembly and service quality satisfaction component semantic matching method as the oriented mission migration of assembly choice criteria.
A kind of component semantic matching method of oriented mission migration, its step is as follows:
(1) sets up the semantic description of task, assembly and equipment for the component semantic matching method in the task immigration;
Adopt the ontology describing language that task, assembly and equipment are carried out semantic description, abstract task, assembly and equipment modeling are become concrete data model;
Task semantic description method in the step (1) is as follows:
Task is described by functional requirement description, QoS requirement and task run state description three parts are formed, wherein:
Functional requirement is described and is comprised task function type, input resource requirement, output resource requirement;
QoS requirement is described and is comprised availability, interactivity, price, reputation, reliability, delay, error rate, throughput, security;
The task run state description is the set of a running status, and it is described and adopts following method mark:
TS={ τ
1, τ
2τ
kFormula 1
TS represents task run state set, τ
iRepresent the semantic description of i running status;
Component semantic describing method in the step (1) is as follows:
The running status that the functional description that assembly is provided by assembly, the service quality that provides are described, supported is described, appointed condition constraint specification four parts are formed, wherein:
The functional description that provides comprises assembly support function type, supports the output resource of input resource, support;
The service quality that provides is described and is divided into performance, reliability, security and four parts of cost;
Facility constraints is described and is comprised software constraint, hardware constraints, network constraint;
The running status of assembly support is described the set of the running status that is a support, and it is described and adopts following method mark:
CS={ τ
1, τ
2τ
kFormula 2
The wherein running status set of CS proxy component support, τ
iThe semantic description of i the running status that representative is supported;
Equipment semantic description method in the step (1) is as follows:
The equipment semantic description comprises software, hardware, network three parts, wherein:
Software description comprises operating system, virtual machine;
Hardware description comprises cpu type, CPU frequency, type of memory, memory size, hard disk type, hard-disk capacity;
Network description comprises network type, the network bandwidth.
(2) give component filter with the description of task, assembly and the equipment of foundation in the step (1) and carry out the assembly filtration;
Adopt component filter that existing assembly is filtered, filtration comprises that function is filtered, QoS requirement filters, running status is filtered, facility constraints is filtered, wherein:
It is to compare according to the function type tabulation of assembly support and the function type of mission requirements that function is filtered, if the ancestors of function type in function type body tree that not have a function type in the tabulation of the function type of candidate's assembly support be mission requirements then filter this candidate's assembly;
It is the QoS requirement of judgement task and service quality that assembly the provides magnitude relationship in each dimension that QoS requirement filters, if have the magnitude relationship of at least one dimension to conflict mutually, then filter this candidate's assembly with the magnitude relationship that the QoS requirement of task is described;
Running status filters to be meant whether determination component supports the recovery of task status, filters ungratified assembly, its determination methods such as following:
Wherein t refers to the user task that need finish, and c refers to the candidate's assembly in the filter process, and (t c) is true time, and assembly satisfies the task status demand as Compatible; If false, then do not satisfy; TS
tRefer to the running status set of this task, CS
cRefer to the task run state set that candidate's assembly is supported, subsumes (τ
i, τ
j) judgement τ in task status body tree
jWhether be τ
jAncestors;
It is the facility constraints description of comparing component and the device description of target device that facility constraints is filtered, judge their numerical values recited relation and notion relation of inclusion between each dimension, if at least one dimension discontented foot-eye device software, hardware or network description then filter this candidate's assembly;
(3) give the component semantic adaptation, the function satisfaction of component semantic adaptation computation module and task and the service quality satisfaction of assembly with the component description subclass and middle setting up of the task of step (1) that obtain in the step (2).
The function satisfaction calculation procedure of assembly is as follows in the step (3):
3.1 traversal Ontological concept topological diagram calculates the bee-line of all nodes to root node, and all bee-lines are sorted, and gets wherein maximal value as the degree of depth of Ontological concept topological diagram;
3.2 according to the functional description of assembly and task, respectively with the input resource of task and assembly and output mapping resources on the corresponding node of Ontological concept topological diagram;
3.3 in the Ontological concept topological diagram, use the distance conduct input semantic distance of the input resource of Dijkstra shortest path first computation module to the shortest path between the input resource of task;
3.4 after if step 3.3 is calculated, do not exist assembly to import the path of resource node to task input resource node, then the input resource of calculation task is to the distance conduct input semantic distance of the shortest path between the input resource of assembly;
3.5 after if step 3.4 is calculated, do not have the path of task input resource node to assembly input resource node, then with the degree of depth of Ontological concept topological diagram as importing semantic distance;
3.6 utilize the input satisfaction of following formula computation module:
L wherein
InputBe the input semantic distance that step 3.3-step 3.5 calculates, D is the degree of depth of the Ontological concept topological diagram that calculates of step 3.2, and δ is a weighting coefficient;
3.7 utilize the output satisfaction of the method computation module of step 3.3-step 3.6;
3.8 utilize the function satisfaction of following formula computation module:
S
Cap=S
Input+ S
OutputFormula 5
S wherein
CapThe function satisfaction of expression assembly, S
InputThe input satisfaction of expression assembly, S
OutputThe output satisfaction of expression assembly;
The computing method of the service quality satisfaction in the step (3) such as following:
S wherein
QosBe meant the service quality satisfaction, P
iRefer to i the value after the quality of service attribute standardization, ρ
iRefer to i quality of service attribute in whole quality of service attribute, account for weighted value, its span is (0,1), and
The standardized method of quality of service attribute such as following:
Wherein v (qos) represents the value of a qos attribute, the mean value of this qos property value in mean (qoses) the expression candidate assembly, the standard deviation of this qos attribute in δ (qoses) the expression candidate assembly;
(4) function satisfaction and the service quality satisfaction that calculates according to step (3), the comprehensive satisfaction of computation module;
Its comprehensive satisfaction computing method such as following:
S=S
Cap* γ
Cap+ S
Qos* γ
QosFormula 8
Wherein S is the comprehensive satisfaction of assembly, S
CapBe the function satisfaction of assembly, γ
CapBe the weight parameter of assembly function satisfaction, S
QosBe the service quality satisfaction of assembly, γ
CapWeight parameter for Component service quality satisfaction;
(5) the highest assembly of output comprehensive satisfaction;
According to comprehensive satisfaction assembly is sorted from high to low, export the assembly of the highest assembly of comprehensive satisfaction as reconstruction task.
The inventive method guarantees the seamlessness and the continuity of user task migration, thereby the bottom details in the shield assembly selection course reduces the participation of user in the assembly selection course; Describe by in task and component description, introducing running status, make this method can support the state of executing the task to preserve and recovery well; This method is synthetically considered the satisfaction aspect two of function and the service quality in the assembly matching process, thereby helps to improve user's comprehensive satisfaction.This method is filtered assembly earlier when assembly is selected, thereby has reduced the calculated amount of satisfaction, has improved the efficient of assembly matching algorithm.
Description of drawings
Fig. 1 is the inventive method process flow diagram;
Fig. 2 is a task bulk junction composition of the present invention;
Fig. 3 is a component body structural drawing of the present invention;
Fig. 4 is an apparatus body structural drawing of the present invention.
Embodiment
Step of the present invention as shown in Figure 1.
(1) adopt body network language OWL that task, assembly, equipment are carried out semantic description:
Use body the build tool prot é g é that semantic body is made up, wherein task body construction such as Fig. 2, by functional requirement, QoS requirement, task run state, functional requirement is described by function type, input resource type, output resource type; QoS requirement then is made up of four parts, is respectively performance, security, cost, reliability, and partly also can segment in each service quality: performance is described by delay, handling capacity, interactivity; Security is described by data encryption, mandate; Cost is described by price; Reliability is described by availability, reputation, error rate; Describe according to above-mentioned task structure, we have designed 4 tasks: (a) video playback task is used for the task of displaying video; (b) music task is used for the task of playing back music; (c) MP3 format file playing task is used to play the task of MP3 format music; (d) picture browsing task is used for the task of browsing pictures;
Component body structure such as Fig. 3 are made up of function, service quality, facility constraints, status of support, and wherein function, service quality, state description are identical with task, and the concrete form of facility constraints is identical with device description; We have designed assembly in 13, are that assembly, the JAVA video playback assembly of management of media file is to use the video playback assembly of realizing based on JMF (Java Media Framework), the assembly that the fast video player module is a simple video playback etc. as the media management assembly;
Device description such as Fig. 4, device description is made up of hardware, software, network three parts, and hardware comprises internal memory, CPU, three descriptions of hard disk, and internal memory is made up of type and capacity; CPU is made up of type and dominant frequency; Hard disk is made up of type and capacity; Software has comprised the description of operating system, virtual machine information; Network is made up of the network type that is possessed on this equipment; We have designed 3 kinds of device descriptions, are respectively desktop computer, notebook, smart mobile phone.
(2) adopt third party library jena resolution component, task, apparatus body semantic description file, use component filter that assembly is filtered then:
Its filtration step is as follows:
2.1 from candidate's the component list, choose candidate's assembly that does not enter filter process, if there is not candidate's assembly of needs filtrations, then finish filtration stage, return the assembly candidate list after the filtration;
Filter 2.2 carry out functional requirement,, and return step 2.1 if do not satisfy then this assembly of deletion from the assembly candidate list;
Filter 2.3 carry out QoS requirement,, and return step 2.1 if do not satisfy then this assembly of deletion from the assembly candidate list;
Filter 2.3 carry out running status,, and return step 2.1 if do not satisfy then this assembly of deletion from the assembly candidate list;
Filter 2.4 carry out facility constraints,, and return step 2.1 if do not satisfy then this assembly of deletion from the assembly candidate list;
(4) the function satisfaction of computation module and service quality satisfaction:
4.1 the computing method of function satisfaction such as following:
S
Cap=S
Input+ S
OutputFormula 1
S wherein
CapThe function satisfaction of expression assembly, S
InputThe input satisfaction of expression assembly, S
OutputThe output satisfaction of expression assembly;
Input satisfaction according to Ontological concept topological diagram computation module:
L wherein
InputBe the input semantic distance, D is the degree of depth of Ontological concept topological diagram, and δ is a weighting coefficient, and its value depends on L
InnputComputing method, work as L
InputAssembly function input resource gets 0.33 in the expression Ontological concept topological diagram when the task function number is gone into the bee-line of resource, works as L
InputGet 0.67 when task function input resource is to the bee-line of assembly function input resource in the expression Ontological concept topological diagram, as if L
InputGet 1 during the degree of depth of expression Ontological concept topological diagram;
The computing method of assembly output satisfaction are identical with input satisfaction computing method;
4.2 the computing method of service quality satisfaction such as following:
S wherein
QosBe meant the service quality satisfaction, P
iRefer to i the value after the quality of service attribute standardization, ρ
iRefer to i quality of service attribute in whole quality of service attribute, account for weighted value, its span is (0,1), and
The standardized method of quality of service attribute such as following:
Wherein v (qos) represents the value of a qos attribute, the mean value of this qos property value in mean (qoses) the expression candidate assembly, the standard deviation of this qos attribute in δ (qoses) the expression candidate assembly;
(4) comprehensive satisfaction of computation module:
Its computing method such as following:
S=S
Cap* γ
Cap+ S
Qos* γ
QosFormula 5
Wherein S is the comprehensive satisfaction of assembly, S
CapBe the function satisfaction of assembly, γ
CapBe the weight parameter of assembly function satisfaction, its span is (0,1), S
QosBe the service quality satisfaction of assembly, γ
CapBe the weight parameter of Component service quality satisfaction, its span is (0,1).
In order to verify the validity of comprehensive satisfaction computing method, we have designed 4 tasks and 12 assemblies, and 1 equipment has calculated comprehensive satisfaction respectively to assembly, result such as following table, FS presentation function satisfaction wherein, QS represents the service quality satisfaction, CS represents comprehensive satisfaction:
(5) the highest assembly of output comprehensive satisfaction:
Adopt quick sorting algorithm, from high to low candidate's assembly is sorted, the highest assembly of output comprehensive satisfaction according to comprehensive satisfaction;
For the validity of the component semantic matching algorithm of verifying oriented mission migration, we mate with 150 assemblies, 300 assemblies, 450 assemblies respectively 4 tasks, the statistic algorithm time, as following table:
Claims (7)
1. the component semantic matching method of oriented mission migration, its step is as follows:
(1) sets up the semantic description of task, assembly and equipment for the component semantic matching method in the task immigration;
(2) give component filter with the description of task, assembly and the equipment of foundation in the step (1) and carry out the assembly filtration;
(3) give the component semantic adaptation, the function satisfaction of component semantic adaptation computation module and task and the service quality satisfaction of assembly with the component description subclass and middle setting up of the task of step (1) that obtain in the step (2);
(4) function satisfaction and the service quality satisfaction that calculates according to step (3), the comprehensive satisfaction of computation module;
(5) according to the assembly comprehensive satisfaction that obtains in the step (4) assembly is sorted the highest candidate's assembly of output comprehensive satisfaction.
2. the component semantic matching method of oriented mission migration according to claim 1, it is characterized in that: the semantic description to task in the step (1) has increased the task run state description on the basis of traditional functional requirement and QoS requirement description, the task run state is meant the state of preserving when being suspended before task immigration, it is described and adopts following method mark:
TS={τ
1,τ
2…τ
k}
TS represents task run state set, τ
iRepresent the semantic description of i running status.
3. the component semantic matching method of oriented mission migration according to claim 1, it is characterized in that: the semantic description to assembly in the step (1) has increased the running status description that assembly is supported on the basis that traditional function, service quality and facility constraints are described, to guarantee the preservation and the recovery of task status in the task immigration process, wherein the running status of assembly specifically refers to the key variables in the operational process of assembly, the running status of assembly support then is a finger assembly running status to understand, and it is described and adopts following method mark:
CS={τ
1,τ
2…τ
k}
The wherein running status set of CS proxy component support, τ
iThe semantic description of i the running status that representative is supported.
4. the component semantic matching method of oriented mission migration according to claim 1 is characterized in that: the filter algorithm of component filter comprises four-stage in the step (2):
4.1 function is filtered, compare according to the function type tabulation of assembly support and the function type of mission requirements, if the ancestors of function type in function type body tree that not have a function type in the tabulation of the function type of candidate's assembly support be mission requirements then filter this candidate's assembly;
4.2 service quality is filtered, the service quality that the QoS requirement of judgement task and assembly provide is in the magnitude relationship of each dimension, if have the magnitude relationship of at least one dimension to conflict mutually, then filter this candidate's assembly with the magnitude relationship that the QoS requirement of task is described;
4.3 running status is filtered, whether determination component supports the recovery of task status, filters ungratified assembly, its determination methods such as following:
Wherein t refers to the user task that need finish, and c refers to the candidate's assembly in the filter process, and (t c) is true time, and assembly satisfies the task status demand as Compatible; If false, then do not satisfy; TS
tRefer to the running status set of this task, CS
cRefer to the task run state set that candidate's assembly is supported, subsumes (τ
i, τ
j) judgement τ in task status body tree
jWhether be τ
jAncestors;
4.4 equipment filters, than the facility constraints description of assembly and the device description of target device, judge their numerical values recited relation and notion relation of inclusion between each dimension, if at least one dimension discontented foot-eye device software, hardware or network description then filter this candidate's assembly.
5. the component semantic matching method of oriented mission migration according to claim 1, it is characterized in that: the computing method of the function satisfaction in the step (3) are:
5.1 traversal Ontological concept topological diagram calculates the bee-line of all nodes to root node, and all bee-lines are sorted, and gets wherein maximal value as the degree of depth of Ontological concept topological diagram;
5.2 according to the functional description of assembly and task, respectively with the input resource of task and assembly and output mapping resources on the corresponding node of Ontological concept topological diagram;
5.3 in the Ontological concept topological diagram, use the distance conduct input semantic distance of the input resource of Dijkstra shortest path first computation module to the shortest path between the input resource of task;
5.4 after if step 5.3 is calculated, do not exist assembly to import the path of resource node to task input resource node, then the input resource of calculation task is to the distance conduct input semantic distance of the shortest path between the input resource of assembly;
5.5 after if step 5.4 is calculated, do not have the path of task input resource node to assembly input resource node, then with the degree of depth of Ontological concept topological diagram as importing semantic distance;
5.6 utilize the input satisfaction of following formula computation module:
L wherein
InputBe the input semantic distance that step 5.3-step 5.5 calculates, D is the degree of depth of the Ontological concept topological diagram that calculates of step 62, and δ is a weighting coefficient;
5.7 utilize the output satisfaction of the method computation module of step 5.3-step 5.6;
5.8 utilize the function satisfaction of following formula computation module:
S
cap=S
input+S
output
S wherein
CapThe function satisfaction of expression assembly, S
InputThe input satisfaction of expression assembly, S
OutputThe output satisfaction of expression assembly.
6. the component semantic matching method of oriented mission migration according to claim 5, it is characterized in that: the Ontological concept topological diagram in the step 5.1 is by the initial graph of setting with the resource body as the Ontological concept topological diagram, the root node of resource body tree is as the root node of Ontological concept figure, and the direction on its figure limit is pointed to the resource subclass by the resource parent in the resource body tree.
7. the component semantic matching method of oriented mission migration according to claim 1 is characterized in that: the computing method of comprehensive satisfaction are shown below in the step (4):
S=S
cap×γ
cap+S
qos×γ
qos
Wherein S is the comprehensive satisfaction of assembly, S
CapBe the function satisfaction of assembly, γ
CapBe the weight parameter of assembly function satisfaction, S
QosBe the service quality satisfaction of assembly, γ
CapWeight parameter for Component service quality satisfaction.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 201010276348 CN101930369B (en) | 2010-09-08 | 2010-09-08 | Task immigration-orientated component semantic matching method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN 201010276348 CN101930369B (en) | 2010-09-08 | 2010-09-08 | Task immigration-orientated component semantic matching method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN101930369A true CN101930369A (en) | 2010-12-29 |
CN101930369B CN101930369B (en) | 2013-03-20 |
Family
ID=43369565
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN 201010276348 Active CN101930369B (en) | 2010-09-08 | 2010-09-08 | Task immigration-orientated component semantic matching method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101930369B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103500160A (en) * | 2013-10-18 | 2014-01-08 | 大连理工大学 | Syntactic analysis method based on sliding semantic string matching |
CN104205056A (en) * | 2012-04-06 | 2014-12-10 | 国际商业机器公司 | Dynamic allocation of a workload across a plurality of clouds |
EP2663928A4 (en) * | 2011-01-10 | 2017-10-25 | Storone Ltd. | Large scale storage system |
CN108733726A (en) * | 2017-04-24 | 2018-11-02 | 西门子(中国)有限公司 | Network semantic model reconfiguration system based on dynamic event and method |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070033590A1 (en) * | 2003-12-12 | 2007-02-08 | Fujitsu Limited | Task computing |
CN101776995A (en) * | 2010-01-04 | 2010-07-14 | 北京航空航天大学 | Self-adaptive combined service optimization method and device |
-
2010
- 2010-09-08 CN CN 201010276348 patent/CN101930369B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070033590A1 (en) * | 2003-12-12 | 2007-02-08 | Fujitsu Limited | Task computing |
CN101776995A (en) * | 2010-01-04 | 2010-07-14 | 北京航空航天大学 | Self-adaptive combined service optimization method and device |
Non-Patent Citations (3)
Title |
---|
《IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS》 20070331 Zhaohui Wu,Qing Wu,Hong Cheng,Gang Pan,Minde Zhao,Jie Sun ScudWare: A Semantic and Adaptive Middleware Platform for Smart Vehicle Space 121-132 1-7 第8卷, 第1期 * |
《计算机科学(增刊)》 20021231 张鹏,李昭原 基于组件技术的事物处理研究与实现 246-248 1-7 第29卷, 第8期 * |
《软件学报》 20070831 周宇,马晓星,曹建农,余萍,吕建 普适环境下基于软件代理虚拟化的应用迁移 2038-2048 1-7 第18卷, 第8期 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2663928A4 (en) * | 2011-01-10 | 2017-10-25 | Storone Ltd. | Large scale storage system |
CN104205056A (en) * | 2012-04-06 | 2014-12-10 | 国际商业机器公司 | Dynamic allocation of a workload across a plurality of clouds |
CN104205056B (en) * | 2012-04-06 | 2017-12-15 | 国际商业机器公司 | Across the method and system of the dynamically distributes of the live load of multiple clouds |
CN103500160A (en) * | 2013-10-18 | 2014-01-08 | 大连理工大学 | Syntactic analysis method based on sliding semantic string matching |
CN103500160B (en) * | 2013-10-18 | 2016-07-06 | 大连理工大学 | A kind of syntactic analysis method based on the semantic String matching that slides |
CN108733726A (en) * | 2017-04-24 | 2018-11-02 | 西门子(中国)有限公司 | Network semantic model reconfiguration system based on dynamic event and method |
CN108733726B (en) * | 2017-04-24 | 2022-03-29 | 西门子(中国)有限公司 | Network semantic model reconstruction system and method based on dynamic events |
Also Published As
Publication number | Publication date |
---|---|
CN101930369B (en) | 2013-03-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Washizaki et al. | Studying software engineering patterns for designing machine learning systems | |
Kolda et al. | Counting triangles in massive graphs with MapReduce | |
US9565129B2 (en) | Resource provisioning planning for enterprise migration and automated application discovery | |
CN107644286A (en) | Workflow processing method and device | |
CN107885499A (en) | A kind of interface document generation method and terminal device | |
CN104516730B (en) | A kind of data processing method and device | |
US11762920B2 (en) | Composite index on hierarchical nodes in the hierarchical data model within a case model | |
CN106649345A (en) | Automatic session creator for news | |
CN112766649B (en) | Target object evaluation method based on multi-scoring card fusion and related equipment thereof | |
Gao et al. | The capability matching of web services | |
US10083031B2 (en) | Cognitive feature analytics | |
CN106990960A (en) | Code dispositions method and device | |
CN101930369B (en) | Task immigration-orientated component semantic matching method | |
CN115080275B (en) | Twin service assembly based on real-time data model and method thereof | |
Dakhil et al. | BIM client maturity: literature review | |
CN107729349B (en) | Method and device for automatically generating feature data set based on metadata | |
US10719299B2 (en) | Physical connection of network components in a graphical computer environment with one-way signal propagation | |
CN113505278A (en) | Graph matching method and device, electronic equipment and storage medium | |
CN111241380B (en) | Method and apparatus for generating recommendations | |
US20210103639A1 (en) | Model-based systems engineering model conversion with text requirements | |
CN111259975A (en) | Method and device for generating classifier and method and device for classifying text | |
Borrmann et al. | AEC Digital Twin Data-Why Structure Matters | |
Muppala et al. | Amazon SageMaker Best Practices: Proven tips and tricks to build successful machine learning solutions on Amazon SageMaker | |
CN109901990B (en) | Method, device and equipment for testing service system | |
Al-Shomar et al. | A novel framework for remote management of social media big data analytics |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant |