KR102158120B1 - System for estimating entity dynamics in linked data cloud using poisson process and self-organizing maps - Google Patents

Abstract

The present invention is a semantic big data for evaluating data changes in a linked data cloud, particularly a Poisson process for change estimation and self-organizing maps (SOM) as a visual change of high-dimensional data. , Probability and machine learning models, wherein the method for predicting entity dynamics in a connected data cloud according to an embodiment comprises: generating an A change estimation model based on a Poisson distribution , Measuring the dynamics for each entity of a linked data cloud using the generated change estimation model, and data recorded in the linked data cloud in consideration of the measured dynamics for each entity. It may include determining a priority for updating the copy.

Description

{SYSTEM FOR ESTIMATING ENTITY DYNAMICS IN LINKED DATA CLOUD USING POISSON PROCESS AND SELF-ORGANIZING MAPS}

The present invention is a semantic big data for evaluating data changes in a linked data cloud, particularly a Poisson process for change estimation and self-organizing maps (SOM) as a visual change of high-dimensional data. , Probability and machine learning models.

Linked Open Data (LOD) is a global information space that structurally represents and connects data.

Linked Open Data (LOD) cloud based on Linked Open Data (LOD) provides a flexible way to integrate data according to LOD rules.

Even though Linked Open Data (LOD) has grown remarkably, new content has been added or removed.

The change behavior of Linked Open Data (LOD) is very important in applications such as data caching, web incremental crawlers, and appropriate vocabulary recommendations for Linked Data Engineer.

However, depending on the scheduling strategy, there are some important challenges to be solved.

The first task is the sensitivity of the index model.

Changes to the Linked Open Data (LOD) source are not automatically propagated to the local copy because the index and cache may be inaccurate due to updated information.

On the other hand, in order to avoid communication overhead, most applications use local copies, so applications must synchronize their local copies with the original data.

Because the index model also relies on other resources such as network bandwidth, more bandwidth is required to obtain higher accuracy when performing updates.

The second task is the record of incomplete changes. In other words, it is related to the problem of not providing complete information when an element changes in many scenarios.

Rather than being automatically propagated across many of the linked open data (LOD) sources, the scheduling strategy must identify the changes.

Many quality measurement techniques have been proposed that help to detect updates when Linked Open Data (LOD) changes.

According to the third task, it is related to the problem that the scheduling strategy is not aware of the source availability.

Many of the linked open data (LOD) update strategies are unaware of the availability of the source. On the other hand, a broken link is one of the major challenges in using Linked Open Data (LOD).

Therefore, the update strategy is not aware of the broken link problem.

Resources are often moved, removed, or updated, so broken link issues and resources are unavailable. Linked Open Data (LOD) can play an important role because it is one of the major problems in using Linked Open Data (LOD).

The identified solution strategy is an HTTP metadata monitoring strategy, in which change detection can use HTTP metadata monitoring.

Here, the HTTP response header includes datestamp and ETag to detect whether the time has passed or the time has changed.

Another solution strategy for predicting change is Dynamic Linked Data Observatory (DYLDO), which takes the entire content and determines locally whether something has changed.

It crawls fixed-sized documents every week to find linked documents, and its implementation is based on the Opensource java project LDSpider, a multithreaded crawling framework for Linked Data.

As for notifications, there are several tools you can use to notify data set updates, such as DSNotify.

DSNotify is a generic framework that can assist human and machine actors to fix broken links, and uses time blocking techniques to identify instance matching problems.

While most notification strategies are based on an indexing infrastructure that monitors data sources, existing solution strategies cannot cope with the scalability and dynamics of a linked open data (LOD) cloud.

Korean Patent Registration No. 10-1675818 "Parameterized dynamic model for cloud migration" Korean Patent Registration No. 10-1678762 "Network entity of communication network"

An object of the present invention is to predict source availability based on a probability-based model and to update the local copy in a more effective manner.

It is an object of the present invention to provide assistance in answering and answering questions in web crawlers, data caching applications, large graph database searches, and health care.

An object of the present invention is to propose a change estimation model based on a Poisson distribution in order to measure the dynamics of an entity in a linked data cloud.

In order to measure the dynamic characteristics of a linked open data (LOD) cloud, an object of the present invention is to visualize changes over a certain period of time in a self-organizing map (SOM).

An object of the present invention is to deal with the scalability problem and dynamics of a linked data cloud (LOD).

The method for predicting entity dynamics in a connected data cloud according to an embodiment includes the steps of generating a change estimation model based on a Poisson distribution, and data connected using the generated change estimation model. Measuring the dynamics for each entity in the cloud (Linked Data Cloud), and determining a priority for updating the copy of data recorded in the linked data cloud in consideration of the measured dynamics for each entity. Can include.

The step of measuring the dynamics for each entity according to an embodiment includes: visualizing changes over a certain period of time through the self-organizing map (SOM) for each entity, and measuring the dynamics for each entity based on the visualization result. It may include.

In the measuring the dynamics for each entity according to an embodiment, a change estimate for each entity is calculated using a Poisson distribution model, and the change estimate is calculated based on past changes in data items. Including, and determining the priority may include predicting an update timing of the data copy based on the past change details.

Determining a priority for updating the data copy according to an embodiment includes: estimating a frequency of change for each entity based on the change estimate, and reserving a next source based on the estimated frequency Determining when to do so, and determining a priority for updating the data copy based on when to reserve.

The step of determining the priority for the update of the data copy according to an embodiment comprises calculating the number of times the number of items in the linked open data (LOD) cloud is changed in each snapshot based on the measured dynamics for each entity. And determining a priority of data copies to be updated based on the calculated number of times.

The measuring the dynamics for each entity according to an embodiment includes measuring a semantic drift that generates dynamics of linked open data (LOD) according to an increase or decrease of data, and the measured semantic drift. ) On the basis of a linked open data (LOD) cloud.

The system for predicting the dynamics of an entity according to an embodiment uses a change estimation model generator that generates an A change estimation model based on a Poisson distribution, and uses the generated change estimation model. A dynamics measuring unit that measures the dynamics of each entity in the linked data cloud, and prioritization for updating the copy of data recorded in the linked data cloud in consideration of the measured dynamics of each entity. It may include a scheduler to determine.

The dynamics measuring unit according to an embodiment may visualize a change over a certain period of time through the self-organizing map (SOM) for each entity, and measure the dynamics for each entity based on the visualization result.

The dynamics measurement unit according to an embodiment calculates a change estimate for each entity using a Poisson distribution model, calculates the change estimate based on past changes of data items, and the scheduler The update timing of the data copy can be predicted based on past changes.

The scheduler according to an embodiment estimates a frequency of change for each entity based on the change estimate, determines when to reserve a next source based on the estimated frequency, and based on the reservation time Priority for updating of the data copy can be determined.

The scheduler according to an embodiment calculates the number of times the number of items in the linked open data (LOD) cloud is changed in each snapshot based on the measured dynamics of each entity, and the data to be updated based on the calculated number of times You can prioritize your copies.

The dynamics measuring unit according to an embodiment measures a semantic drift that generates dynamics of linked open data (LOD) according to an increase or decreases in data, and based on the measured semantic drift, the LOD ( Linked Open Data) Can predict changes occurring in the cloud.

According to an embodiment, source availability may be predicted based on a probability-based model and local copies may be updated in a more effective manner.

According to an embodiment, it may provide assistance with web crawlers, data caching applications, large graph database searches, and answer-response questions in health care.

According to an embodiment, a change estimation model based on a Poisson distribution may be proposed in order to measure the dynamics of an entity in a linked data cloud.

According to an embodiment, in order to measure the dynamic characteristics of a linked open data (LOD) cloud, a self-organizing map (SOM) may visualize changes over a certain period of time.

According to an embodiment, it is possible to deal with the scalability problem and dynamics of a linked data cloud (LOD).

1 is a diagram showing a proposed approach for estimating a change in a linked open data (LOD) source.
2 is a diagram illustrating a method of predicting entity dynamics according to an embodiment.
3 and 4 are diagrams illustrating a method of measuring dynamics for each entity according to an embodiment.
5 and 6 are diagrams for determining priorities for updating data copies according to an exemplary embodiment.
7 is a diagram illustrating an entity dynamics prediction system according to an embodiment.

Specific structural or functional descriptions of the embodiments according to the concept of the present invention disclosed in this specification are exemplified only for the purpose of describing the embodiments according to the concept of the present invention, and embodiments according to the concept of the present invention They may be implemented in various forms and are not limited to the embodiments described herein.

Since the embodiments according to the concept of the present invention can apply various changes and have various forms, the embodiments will be illustrated in the drawings and described in detail herein. However, this is not intended to limit the embodiments according to the concept of the present invention to specific disclosed forms, and includes changes, equivalents, or substitutes included in the spirit and scope of the present invention.

Terms such as first or second may be used to describe various elements, but the elements should not be limited by the terms. The above terms are only for the purpose of distinguishing one component from other components, for example, without departing from the scope of rights according to the concept of the present invention, the first component may be named as the second component, Similarly, the second component may also be referred to as a first component.

When a component is referred to as being "connected" or "connected" to another component, it is understood that it may be directly connected or connected to the other component, but other components may exist in the middle. Should be. On the other hand, when a component is referred to as being "directly connected" or "directly connected" to another component, it should be understood that there is no other component in the middle. Expressions describing the relationship between components, for example, "between" and "just between" or "directly adjacent to" should be interpreted as well.

The terms used in the present specification are only used to describe specific embodiments and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In the present specification, terms such as "comprise" or "have" are intended to designate that the specified features, numbers, steps, actions, components, parts, or combinations thereof exist, but one or more other features or numbers, It is to be understood that the presence or addition of steps, actions, components, parts or combinations thereof does not preclude the possibility of preliminary exclusion.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art to which the present invention belongs. Terms as defined in a commonly used dictionary should be interpreted as having a meaning consistent with the meaning in the context of the related technology, and should not be interpreted as an ideal or excessively formal meaning unless explicitly defined in this specification. Does not.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, the scope of the patent application is not limited or limited by these embodiments. The same reference numerals in each drawing indicate the same members.

1 is a diagram showing a proposed approach for estimating a change in a linked open data (LOD) source.

As shown in Fig. 1, connected data continues to evolve and new sources can be added or removed. The proposed approach in the present invention can cope with the dynamics of data and reserve resources.

Referring to FIG. 1, the core admin interface 100 includes a configuration interface 101 and a linked data explorer 102. Meanwhile, the extended core admin interface 111 adjacent to the core admin interface 100 includes a sparql interface 112 and a linked data cache 113 (Linked Data Cache).

The core admin interface 100 and the extended core admin interface 111 can share with the core resource 103 and the proposed methodology 114 according to the access control rule 110.

For example, the core resource 103 is a resource web service 104, an import/export web service 105, an endpoint service 106, and a SPARQL service. It may include (SPARQL Service, 107), a triple store (108), and a configuration (configuration, 109). In addition, the core resource 103 can input/output data from the database 124.

Meanwhile, the proposed methodology (114) can be described as follows.

Estimating changes to entities in a linked data cloud is important for applications that use publicly available data.

The proposed method is helpful for answering questions in web crawlers, data caching applications, large graph database searches, and health care.

In the present invention, a change estimation model based on a Poisson distribution may be generated to measure the entity dynamics of a connected data cloud.

Probability-based change estimation models can predict source availability and update local copies in a more efficient manner.

To measure the dynamic characteristics of a linked open data (LOD) cloud, a self-organizing map (SOM) can visualize changes over a period of time.

In the present invention, it is possible to deal with the scalability problem and dynamics of the connected data cloud.

In general, Linked Open Data (LOD) applications can pre-fetch data from the source and store it in a cache for fast response speed.

However, in the process, the data in the linked open data (LOD) cloud is constantly updated, and the cache-maintained copy must also be updated.

Meanwhile, due to limitations such as bandwidth and computation time, Linked Open Data (LOD) applications must prioritize copies of the data to be updated.

In order to make the best use of resource availability, a good scheduling strategy 122 is needed to fetch data and update the index 121.

Also, ideally in the linked open data (LOD) cloud, the cache needs to be updated as changes occur.

In the present invention, the proposed change estimation 115 strategy using a Poisson distribution model 117 can estimate changes in two snapshots of data.

In the present invention, the schedule plan must predict when to update the local copy of the original linked open data (LOD) based on the past change history of the data item. The scheduling strategy for this is based on the history of past changes from Linked Open Data (LOD) sources.

These strategies can estimate how often items change and then decide when to reserve the source. For example, if you schedule an item and detect four changes, you can estimate how often the source has changed based on the previously observed item history.

The ChangeRatio strategy according to the present invention can calculate the number of times the number of items in the linked open data (LOD) cloud is changed in each snapshot.

ChangeRatio defines the number of changes to the LOD (Linked Open Data), and can be expressed as [Equation 1].

[Equation 1]

In [Equation 1], the function indicates that a change has occurred in an item whose value is 1 when the value of item oi is changed, and has a value of 0 when it is not changed.

The ChangeRatio, calculated by [Equation 1], is useful for storing the change history of an item in Linked Open Data (LOD).

On the other hand, Linked Open Data (LOD) sources change frequently instead of looking for changes between two snapshots, so it's good to look for dynamic characteristics over a period of time.

In the present invention, it is possible to quantify a change in a scheduling strategy based on dynamics and measure the dynamics of a linked open data (LOD) cloud.

The dynamics (118) of Linked Open Data (LOD) can be expressed as mentioned in [Equation 1] above. This can quantify changes between two data sets, such as different snapshots of Linked Open Data (LOD) sources.

The dynamic function can quantify the evolution of linked open data (LOD) over time.

The quantification of these dynamics can be expressed by the following [Equation 2].

[Equation 2]

As data continues to grow and new entities are constantly added and removed, the dynamics of Linked Open Data (LOD) can measure the semantic drift that causes the dynamics of the data.

Although managing the scale of data is different, the present invention proposes self-organizing maps (SOM) 120 that solve the problem of scale and reduce the dimension to visualize the large scale of evolving data in an effective manner.

Linked data crawled in the cloud (Subject, Predicate, Object, and Context) is a form retrieved by NQuads 108 and data caching application 113 at regular intervals of time t.

For example, the entire data set can be expressed as [Equation 3].

[Equation 3]

In [Equation 3], it represents the number of triples included in the data set in the context c at time t.

Data features can be used in most scheduling strategies to improve the up-to-dateness of cached data.

The strategy proposed in the present invention predicts changes occurring in a linked open data (LOD) cloud using functions such as change rates and dynamic characteristics.

However, whenever a change occurs, the SPARQL endpoint 106 may be updated. Therefore, long-running applications that use cache and query results must resubmit queries regularly to ensure up-to-date results.

In the present invention, it is possible to solve the refresh query scheduling problem and avoid overload.

2 is a diagram illustrating a method of predicting entity dynamics according to an embodiment.

The entity dynamics prediction method according to an embodiment may generate an A change estimation model based on a Poisson distribution (step 201).

In addition, the entity dynamics prediction method according to an embodiment may measure entity-specific dynamics of a linked data cloud using a change estimation model (step 202).

In addition, the method for predicting entity dynamics according to an embodiment may determine a priority for updating a copy of data recorded in a linked data cloud (step 203).

3 and 4 are diagrams illustrating a method of measuring dynamics for each entity according to an embodiment.

First, referring to FIG. 3, the entity dynamics prediction method according to an embodiment may visualize changes over a certain period of time through a self-organizing map (SOM) for each entity in order to measure the dynamics for each entity (step 301).

In addition, the entity dynamics prediction method according to an embodiment may measure the entity-specific dynamics based on a visualization result (step 302).

Next, referring to FIG. 4, the method for predicting entity dynamics according to an embodiment may measure semantic drift that generates dynamics of linked open data (LOD) according to an increase or decrease in data (step 401).

In addition, the entity dynamics prediction method according to an embodiment may predict a change occurring in a linked open data (LOD) cloud based on the measured semantic drift (step 402).

5 and 6 are diagrams for determining priorities for updating data copies according to an exemplary embodiment.

First, according to FIG. 5, in the method for predicting entity dynamics according to an embodiment, in order to determine a priority for updating data copies, the frequency of changes for each entity may be estimated (step 501).

In addition, the entity dynamics prediction method according to an embodiment determines when to reserve the next source based on the estimated frequency (step 502), and prioritizes the update of the data copy recorded in the linked data cloud. The ranking can be determined (step 503).

Next, according to FIG. 6, the entity dynamics prediction method according to an embodiment calculates the number of times the number of items in the linked open data (LOD) cloud is changed in each snapshot in order to determine the priority for updating the data copy. Yes (step 601).

In addition, the entity dynamics prediction method according to an embodiment may determine the priority of the data copy to be updated based on the calculated number of times (step 602).

7 is a diagram illustrating an entity dynamics prediction system 700 according to an embodiment.

The entity dynamics prediction system 700 according to an embodiment may predict source availability based on a probability-based model and update the local copy in a more effective manner.

To this end, the entity dynamics prediction system 700 according to an embodiment may include a change estimation model generation unit 710, a dynamics measurement unit 720, and a scheduler 730.

Specifically, the change estimation model generator 710 according to an embodiment may generate an A change estimation model based on a Poisson distribution.

Next, the dynamics measurement unit 720 according to an embodiment may measure the dynamics of each entity of a linked data cloud using the generated change estimation model.

According to an example, the dynamics measurement unit 720 may visualize changes over a certain period of time through a self-organizing map (SOM) for each entity, and measure the dynamics for each entity based on the visualization result.

Finally, the scheduler 730 according to an embodiment may determine a priority for updating a copy of data recorded in the linked data cloud in consideration of the measured dynamics of each entity.

Meanwhile, the dynamics measurement unit 720 may calculate a change estimate for each entity using a Poisson distribution model, and may calculate the change estimate based on a past change history of a data item. In this case, the scheduler 730 may predict the update timing of the data copy based on the past change history.

According to an embodiment, the dynamics measuring unit 720 may measure semantic drift that generates dynamics of linked open data (LOD) according to an increase or decrease in data. In this case, the dynamics measurement unit 720 may predict a change occurring in a linked open data (LOD) cloud based on the measured semantic drift.

In addition, the scheduler 730 according to an embodiment may estimate the frequency of change for each entity based on the change estimate. In this case, the scheduler 730 may determine when to reserve the next source based on the estimated frequency, and determine a priority for updating the data copy based on the time to reserve.

In addition, the scheduler 730 according to an embodiment may calculate the number of times the number of items in a linked open data (LOD) cloud is changed in each snapshot based on the measured dynamics for each entity. In this case, the scheduler 730 may determine the priority of data copies to be updated based on the calculated number of times.

In the end, using the present invention, it is possible to predict source availability based on a probability-based model and update the local copy in a more effective manner.

In addition, when the present invention is used, it can provide assistance with web crawlers, data caching applications, large graph database searches, and answer-response questions in health care, and to measure entity dynamics in a linked data cloud. For this, we can propose a change estimation model based on the Poisson distribution.

In addition, when using the present invention, in order to measure the dynamic characteristics of a linked open data (LOD) cloud, a self-organizing map (SOM) can visualize changes over a certain period of time, and scalability problems of a linked data cloud (LOD) And can deal with dynamics.

The apparatus described above may be implemented as a hardware component, a software component, and/or a combination of a hardware component and a software component. For example, the devices and components described in the embodiments include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA), It can be implemented using one or more general purpose computers or special purpose computers, such as a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications executed on the operating system. In addition, the processing device may access, store, manipulate, process, and generate data in response to the execution of software. For the convenience of understanding, although it is sometimes described that one processing device is used, one of ordinary skill in the art, the processing device is a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that it may include. For example, the processing device may include a plurality of processors or one processor and one controller. In addition, other processing configurations are possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of these, configuring the processing unit to behave as desired or processed independently or collectively. You can command the device. Software and/or data may be interpreted by a processing device or to provide instructions or data to a processing device, of any type of machine, component, physical device, virtual equipment, computer storage medium or device. , Or may be permanently or temporarily embodyed in a transmitted signal wave. The software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the embodiment, or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -A hardware device specially configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of the program instructions include not only machine language codes such as those produced by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware device described above may be configured to operate as one or more software modules to perform the operation of the embodiment, and vice versa.

As described above, although the embodiments have been described by the limited drawings, various modifications and variations are possible from the above description to those of ordinary skill in the art. For example, the described techniques are performed in a different order from the described method, and/or components such as a system, structure, device, circuit, etc. described are combined or combined in a form different from the described method, or other components Alternatively, even if substituted or substituted by an equivalent, an appropriate result can be achieved.

Therefore, other implementations, other embodiments, and claims and equivalents fall within the scope of the claims to be described later.

Claims (12)

Generating an A change estimation model based on a Poisson distribution;
Measuring the dynamics of each entity of a linked data cloud using the generated change estimation model; And
In consideration of the measured dynamics of each entity, determining a priority for updating a copy of data recorded in the linked data cloud,
The step of measuring the dynamics for each entity
Calculating a change estimate for each entity using a Poisson distribution model, but calculating the change estimate based on past changes in data items
Including,
The step of determining the priority,
Predicting an update timing of the data copy based on the past change details
A method for predicting the dynamics of an entity in a connected data cloud comprising a. The method of claim 1,
The step of measuring the dynamics for each entity,
Visualizing changes over a certain period of time through the self-organizing map (SOM) for each entity, and measuring the dynamics for each entity based on the visualization result
A method for predicting the dynamics of an entity in a connected data cloud comprising a. delete The method of claim 1,
The step of determining the priority for updating the data copy,
Estimating a frequency of change for each entity based on the change estimate value; And
Determining when to reserve a next source based on the estimated frequency; And
Determining a priority for updating the data copy based on the reservation time
A method for predicting the dynamics of an entity in a connected data cloud comprising a. Generating an A change estimation model based on a Poisson distribution;
Measuring the dynamics of each entity of a linked data cloud using the generated change estimation model; And
In consideration of the measured dynamics of each entity, determining a priority for updating a copy of data recorded in the linked data cloud,
The step of determining the priority for updating the data copy,
Calculating the number of times the number of items in the linked open data (LOD) cloud is changed in each snapshot based on the measured dynamics of each entity; And
Determining the priority of data copies to be updated based on the calculated number of times
A method for predicting the dynamics of an entity in a connected data cloud comprising a. Generating an A change estimation model based on a Poisson distribution;
Measuring the dynamics of each entity of a linked data cloud using the generated change estimation model; And
In consideration of the measured dynamics of each entity, determining a priority for updating a copy of data recorded in the linked data cloud,
The step of measuring the dynamics for each entity
Measuring a semantic drift that generates dynamics of linked open data (LOD) according to an increase or decrease in data; And
Predicting a change occurring in a linked open data (LOD) cloud based on the measured semantic drift
A method for predicting the dynamics of an entity in a connected data cloud comprising a. A change estimation model generator for generating an A change estimation model based on a Poisson distribution;
A dynamics measuring unit for measuring the dynamics of each entity of a linked data cloud using the generated change estimation model; And
And a scheduler that determines a priority for updating a copy of data recorded in the linked data cloud in consideration of the measured dynamics of each entity,
The dynamic measurement unit,
Calculate the change estimate for each entity using a Poisson distribution model, but calculate the change estimate based on the past change history of the data item,
The scheduler,
A system for predicting the dynamics of an entity in a connected data cloud, characterized in that predicting an update timing of the data copy based on the past change history. The method of claim 7,
The dynamic measurement unit,
A system for predicting the dynamics of an entity in a connected data cloud, characterized in that for visualizing changes over a certain period of time through the self-organizing map (SOM) for each entity, and measuring the dynamics for each entity based on the visualization result. delete The method of claim 7,
The scheduler,
Based on the change estimate, estimate the frequency of change for each entity, determine when to reserve the next source based on the estimated frequency, and prioritize updating of the data copy based on the reservation time A system for predicting the dynamics of an entity in a connected data cloud, characterized in that to determine. A change estimation model generator for generating an A change estimation model based on a Poisson distribution;
A dynamics measuring unit for measuring the dynamics of each entity of a linked data cloud using the generated change estimation model; And
And a scheduler that determines a priority for updating a copy of data recorded in the linked data cloud in consideration of the measured dynamics of each entity,
The scheduler,
Based on the measured dynamics of each entity, the number of times the number of items in the linked open data (LOD) cloud is changed in each snapshot, and determining the priority of the data copy to be updated based on the calculated number of times. A system that predicts the dynamics of an entity in a connected data cloud. A change estimation model generator for generating an A change estimation model based on a Poisson distribution;
A dynamics measuring unit for measuring the dynamics of each entity of a linked data cloud using the generated change estimation model; And
And a scheduler that determines a priority for updating a copy of data recorded in the linked data cloud in consideration of the measured dynamics of each entity,
The dynamics measurement unit measures a semantic drift that generates dynamics of linked open data (LOD) according to an increase or decreases in data, and a linked open data (LOD) cloud based on the measured semantic drift. A system for predicting the dynamics of an entity in a connected data cloud, characterized in that predicting changes occurring in

Images