CN102411594A - Method and device for obtaining information - Google Patents
Method and device for obtaining information Download PDFInfo
- Publication number
- CN102411594A CN102411594A CN2010102928280A CN201010292828A CN102411594A CN 102411594 A CN102411594 A CN 102411594A CN 2010102928280 A CN2010102928280 A CN 2010102928280A CN 201010292828 A CN201010292828 A CN 201010292828A CN 102411594 A CN102411594 A CN 102411594A
- Authority
- CN
- China
- Prior art keywords
- data
- information
- collection
- information entropy
- item
- 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 method and a device for obtaining information, which have the main technical scheme that: data used for obtaining information is determined in advance, and in addition, the time period for generating the data is divided into a plurality of sub time periods; and the following steps are executed by aiming at each sub time period: loading the data generated in the current sub time period; determining the first information entropy corresponding to each itemset obtained from at least one preset data attribute combination in the loaded data; determining the second information entropy corresponding to each itemset in the data generated in all sub time periods before the current sub time period; and updating the itemset set of information used for mark obtaining according to the first information entropy and the second information entropy corresponding to the first itemset. When the technical scheme is adopted, on one hand, the information obtaining efficiency is improved, and on the other hand, the system expanse is reduced.
Description
Technical field
The present invention relates to data processing field, relate in particular to a kind of method and device that obtains information.
Background technology
Along with the social informatization degree improves constantly, the data of information system amount constantly expands, and different industries needs to handle, analyze data stream a large amount of and that bring in constant renewal in.At present; The problem that every profession and trade faces is that data volume is very big, but wherein real valuable information seldom, therefore; How from the data of a large amount of and continual renovation, to excavate valuable information, become the difficult point of puzzlement every profession and trade so that follow-up business is instructed.
Data mining is exactly for complying with the data processing technique that the needs that from mass data, obtain valuable information arise at the historic moment.The Knowledge Discovery in the database (knowledge discovery indatabase) is claimed in data mining again; Be meant, unknown, information or pattern non-trivial and that potential using value arranged implicit, merged the theory and technology in a plurality of fields such as database, artificial intelligence, machine learning, statistics from a large amount of incomplete, noisy, fuzzy extracting data.Data Mining Tools can predict in the future trend and behavior, thereby supports people's decision-making well.
From mass data, obtain valuable information; General way is to utilize relational database at present; Detailed process is: the mass data unification that will be used for the information of obtaining is loaded into relational database, on this basis loaded data is carried out data mining in the hope of finding Useful Information then.Relational database is to be the database on basis with the relational model; Define various data relationships in this relational model, promptly utilized the relation of definition to come data of description, wherein; A relation both can be used for describing an entity and attribute thereof, also can be used for describing the contact of inter-entity.Therefore, data are handled, at first data source file is accomplished complete loading and form the data acquisition that satisfies the inspection of relational database normal form, then database table is carried out projection meter on composite attribute and calculate, obtain the counting statistics value according to relational database.In practical application; Adopt relational database from mass data, to obtain information; After need waiting the total data that is ready to use in the information of obtaining to produce to finish again property be loaded into and handle in the relational database, make the data volume that in relational database, need concern calculating assemble, the problem that causes thus is: on the one hand; Need to consume system resources such as a large amount of CPU, I/O, internal memory, system overhead is very big; On the other hand, need the data volume of disposable processing huge, processing procedure needs the time of labor, and information acquisition efficiency is low.
In sum, prior art is obtained information based on relational database from data, and information acquisition efficiency is low, and system overhead is big.
Summary of the invention
In view of this, the embodiment of the invention provides a kind of method and device that obtains information, adopts this technical scheme, has improved the efficient that information is obtained on the one hand, has reduced system overhead on the other hand.
The embodiment of the invention realizes through following technical scheme:
An aspect according to the embodiment of the invention provides a kind of method of obtaining information.
The method of the information of obtaining that provides according to the embodiment of the invention confirms to be used to the data of the information of obtaining in advance, and the time period that will produce said data is divided into a plurality of sub-time periods;
Carry out to each sub-time period:
Load the data that the current sub-time period produces;
The first information entropy that each item set pair that is obtained by predefined at least one data attribute combination in the said data of confirming to load is answered;
Second information entropy that each item set pair is answered described in the data of confirming to produce in current sub-time period sub-time periods of before all;
The first information entropy and second information entropy according to said each item set pair is answered are upgraded the item collection set that is used to identify the information of obtaining.
According to another aspect of the embodiment of the invention, a kind of device that obtains information is provided also.
The device of the information of obtaining that provides according to the embodiment of the invention comprises:
The data load unit is used to confirm to be used to the data of the information of obtaining, and the time period that will produce said data be divided into a plurality of sub-time periods, and load the data that the current sub-time period produces;
First information entropy is confirmed the unit, is used for confirming the first information entropy that each item set pair that said data that said data load unit loads are obtained by predefined at least one data attribute combination is answered;
Second information entropy is confirmed the unit, is used for second information entropy of confirming that each item set pair is answered described in said data load unit all sub-time period loaded data before the current sub-time period;
Collection set updating block is used for confirming that according to said first information entropy first information entropy that each item set pair that the unit is confirmed is answered and said second information entropy confirm that second information entropy of confirming the unit upgrades the item collection set that is used to identify the information of obtaining.
Above-mentioned at least one technical scheme that provides through the embodiment of the invention; Confirm to be used to the data of the information of obtaining in advance; And the time period that will produce data is divided into a plurality of sub-time periods; Carry out to each sub-time period: load the data that the current sub-time period produces; Confirm in the loaded data first information entropy of answering, confirm second information entropy that each item set pair is answered in the data that all the sub-time periods before the current sub-time period produce, and upgrade the item collection set that is used to identify the information of obtaining according to the first information entropy and second information entropy that each item set pair is answered by each item set pair that predefined at least one data attribute combination obtains.Adopt this technical scheme, data based its generation time that will be used to the information of obtaining is divided into a plurality of sub-time periods, once only loads the data of a time period; Be used to identify the item collection set of the information of obtaining based on the Data Update that produces in this time period; Compared with prior art, the task distribution that will from data, obtain information is a plurality of execution, has significantly reduced the data volume of each processing; Thereby improved the efficient that information is obtained, and reduced system overhead.
Other features and advantages of the present invention will be set forth in instructions subsequently, and, partly from instructions, become obvious, perhaps understand through embodiment of the present invention.The object of the invention can be realized through the structure that in the instructions of being write, claims and accompanying drawing, is particularly pointed out and obtained with other advantages.
Description of drawings
Accompanying drawing is used to provide further understanding of the present invention, and constitutes the part of instructions, is used to explain the present invention with the embodiment of the invention, is not construed as limiting the invention.In the accompanying drawings:
The method flow diagram one of the information of obtaining that Fig. 1 provides for the embodiment of the invention one;
The process flow diagram of definite first information entropy that Fig. 2 provides for the embodiment of the invention one;
The process flow diagram of confirming second information entropy that Fig. 3 provides for the embodiment of the invention one;
The renewal that Fig. 4 provides for the embodiment of the invention one is used to identify the process flow diagram of the item collection set of the information of obtaining;
The method flow diagram two of the information of obtaining that Fig. 5 provides for the embodiment of the invention one;
The method flow diagram two of the information of obtaining that Fig. 6 provides for the embodiment of the invention three;
The method flow diagram three of the information of obtaining that Fig. 7 provides for the embodiment of the invention three;
The method flow diagram four of the information of obtaining that Fig. 8 provides for the embodiment of the invention three;
The method flow diagram five of the information of obtaining that Fig. 9 provides for the embodiment of the invention three;
The method flow diagram six of the information of obtaining that Figure 10 provides for the embodiment of the invention three;
The method flow diagram seven of the information of obtaining that Figure 11 provides for the embodiment of the invention three;
The method flow diagram of the information of obtaining that Figure 12 provides for the embodiment of the invention four.
Embodiment
In order to provide efficient that raising information obtains and the implementation that reduces system overhead; The embodiment of the invention provides a kind of method and device that obtains information; Below in conjunction with Figure of description the preferred embodiments of the present invention are described; Should be appreciated that preferred embodiment described herein only is used for explanation and explains the present invention, and be not used in qualification the present invention.And under the situation of not conflicting, embodiment and the characteristic among the embodiment among the application can make up each other.
Embodiment one
According to the embodiment of the invention one; A kind of method of obtaining information is provided; Data based its generation time that this method will be used to the information of obtaining is divided into a plurality of sub-time periods; Once only load the data of a time period, be used to identify the item collection set of the information obtained based on the Data Update that produces in this time period, the task distribution that will from data, obtain information is a plurality of execution with reach raising information efficient of obtaining and the purpose that reduces system overhead.
In the method for the information of obtaining that the embodiment of the invention one provides, need confirm to be used to the data of the information of obtaining in advance, and the time period that will produce these data is divided into a plurality of sub-time periods.Preferably; Can be divided into a plurality of sub-time period of W constant duration T the time period that produce these data; Wherein, this time interval T obtains the required duration of information more than or equal to what estimate to obtain from each sub-time period, promptly guarantees before loading the data that the current sub-time period produces; From the data of last one sub-time period generation, obtaining information finishes; According to this optimal way, avoided loading corresponding data of current sub-time period after since corresponding data of last one sub-time period also be untreated finish (promptly also not obtaining information from data finishes) cause the problem of corresponding data processing hysteresis of current sub-time period; Thereby can guarantee the continuity of data processing, improve data processing efficiency.
Be to be understood that; The method of the sub-time period of division that more than provides is merely the preferred implementation that the embodiment of the invention one provides, in concrete the application, and can taking into account system processing power and concrete factors such as data processing amount; Confirm dividing mode flexibly, enumerate no longer one by one here.
After confirming to be used to the data of the information of obtaining and accomplishing the division of sub-time period, the method for the information of obtaining that the embodiment of the invention one provides is as shown in Figure 1, and sub-time period of each that obtains to division is carried out following steps 101 to step 104:
The data that step 101, current sub-time period of loading produce.
In this step 101, after confirming each sub-time period, this sub-time period is carried out timing, after this sub-time period finishes, mean data that should sub-time section are produced to finish, then put down in writing the data that the current sub-time period produces.In the practical application, can carry out timing, and trigger the data that load each sub-time period generation through time controller.
The first information entropy that each item set pair that is obtained by predefined at least one data attribute combination in the data that step 102, definite current sub-time period that loads produce is answered.
Before carrying out this step 102, preestablish the data attribute of the data of the information of will obtaining, one or more data attributes can be set according to actual needs, and obtain a collection by the data attribute combination that is provided with.For example, in particular cases, a data attribute only is set, then corresponding item collection also is one, and this set pair should data attribute; If N data attribute (N is more than or equal to 2) is set; Then can make up the item collection that obtains the corresponding different pieces of information attribute of a plurality of difference to this N data attribute; For example; 3 data attribute A, B, C are arranged, then can make up and obtain 7 kinds of item collection, these 7 item collection are respectively: { A}, { B}, { C}, { A, B}, { B, C}, { A, C}, { A, B, C}.
In this step 102, confirm that the detailed process of first information entropy will specify in the subsequent implementation example, wouldn't describe here.
In this step 103, if the current sub-time period is first sub-time period, second information entropy that then each item set pair is answered in the data of the sub-time period generation of all before the current sub-time period is 0.
In this step 103, confirm that the detailed process of second information entropy will specify in the subsequent implementation example, wouldn't describe here.
The first information entropy that each item set pair that step 104, basis are confirmed is answered and second information entropy are upgraded the item collection that is used to identify the information of obtaining and are gathered.
In this step 103, the detailed process of upgrading the item collection set that is used for identifying the information of obtaining will specify in the subsequent implementation example, wouldn't describe here.
So far, the process that the Data Update that produces according to sub-time period is used to identify the item collection set of the information of obtaining finishes, and obtains information the data that promptly produce from the current sub-time period and finishes.In the above-mentioned flow process, step 102 and step 103 do not have strict execution sequence, can first execution in step 103 in the practical application execution in step 102, perhaps executed in parallel again.
In the embodiment of the invention one, after confirming to be used to the data of the information of obtaining, handle, thereby the process of information is obtained in completion from the data of each sub-time period generation according to the data that the described flow process of Fig. 1 produces each sub-time period successively.
In the step 102 of the said flow process of Fig. 1, the process of the first information entropy that each set pair is answered in the said data of confirming to load, as shown in Figure 2, comprise the steps:
So far, the process of confirming the first information entropy that an item set pair is answered finishes.In the above-mentioned flow process, step 201 and step 202 do not have strict execution sequence, can first execution in step 202 in the practical application execution in step 201, perhaps executed in parallel again.
In the step 201 and step 202 of flow process shown in Figure 2, the data volume of data can be the bar number of data recording, also can be the storage size of data occupancy.
In the step 203 of flow process shown in Figure 2,, confirm the first information entropy that this set pair is answered, comprising according to the data volume and the said total amount of data that meet the data attribute that this set pair answers confirmed:
Confirm to meet the data volume of the data attribute that this set pair answers and the ratio of said total amount of data;
Utilize this ratio to multiply by the value that this ratio is taken the logarithm and obtained, the negative value of the product that obtains is confirmed as the first information entropy that this set pair is answered.
In the embodiment of the invention, this ratio taken the logarithm to think to use logarithmic function to carry out section diffusion, because this functional value is for negative, so the negative value of the product that obtains is confirmed as the first information entropy that this set pair is answered.
In the step 103 of the said flow process of Fig. 1, the process of second information entropy that each set pair is answered in the data of confirming to produce in current sub-time period sub-time periods of before all, as shown in Figure 3, comprise the steps:
So far, the process of confirming second information entropy that an item set pair is answered finishes.In the above-mentioned flow process, step 301 and step 302 do not have strict execution sequence, can first execution in step 302 in the practical application execution in step 301, perhaps executed in parallel again.
In the step 301 and step 302 of flow process shown in Figure 3, the data volume of data can be the bar number of data recording, also can be the storage size of data occupancy.
In the step 303 of flow process shown in Figure 3,, confirm second information entropy that this set pair is answered, comprising according to the data volume and the said total amount of data that meet the data attribute that this set pair answers:
Confirm the data volume that this meets the data attribute that this set pair answers and the ratio of said total amount of data;
Utilize this ratio to multiply by the value that this ratio is taken the logarithm and obtained, the negative value of the product that obtains is confirmed as second information entropy that this set pair is answered.
In the embodiment of the invention, this ratio taken the logarithm to think to use logarithmic function to carry out section diffusion, because this functional value is for negative, so the negative value of the product that obtains is confirmed as second information entropy that this set pair is answered.
In the step 104 of the said flow process of Fig. 1, upgrade a process that collection is gathered that is used to identify the information of obtaining according to the first information entropy and second information entropy that each item set pair of confirming is answered, as shown in Figure 4, comprise the steps:
The first information entropy of step 401, definite correspondence and the second information entropy sum reach first collection set of first threshold, and first collection during wherein first collection gathered passes through corresponding first information entropy and second information entropy sign;
The first information entropy and second information entropy of so far, answering according to definite each item set pair are upgraded the process end that the item that is used to identify the information of obtaining collects set.
In the step 402 of flow process shown in Figure 4, utilize first collection set determining to upgrade the process of the item collection set that is used to identify the information of obtaining, specifically comprise:
Be included in the item collection that is used for identifying the information of obtaining as if first collection in first collection set and gather, then utilize in this first the collection set and replace the corresponding entry collection that this is used for identifying the item collection set of the information of obtaining through the first information entropy of correspondence and first collection of second information entropy sign;
If first collection during first collection gathered is not included in the item collection set that is used for identifying the information of obtaining, then deletes this item that is used for identifying the information of obtaining and collect the corresponding entry collection of set.
Further; If last sub-time period in the time period of the said data of non-generation of current sub-time period; Then in the above-mentioned steps 104, the first information entropy and second information entropy of promptly answering according to each item set pair are upgraded the item collection set that is used to identify the information of obtaining, and also comprise:
Definite corresponding first information entropy reaches second collection set of second threshold value, and second collection during wherein second collection gathered is through corresponding first information entropy sign;
Utilize said second collection set to upgrade the item collection set that is used to identify the information of obtaining.
Wherein: utilize said second collection set to upgrade the item collection set that is used to identify the information of obtaining, comprising:
With second collection that is not included in the item collection set that is used for identifying the information of obtaining in second collection set, add the said item collection set that is used for identifying the information of obtaining.
Through above embodiment the process of from data, obtaining information in the technical scheme provided by the invention has been carried out detailed description; For understanding the embodiment of the invention better, below complete process process when further combining to the data that are used to the information of obtaining for user's ticket writing describe.
Before the method for the information of obtaining that the execution embodiment of the invention one provides, be provided with as follows:
Set a collection implicit information entropy threshold value E
P0, wherein, the corresponding above-described first information entropy of implicit information entropy, threshold value E
P0Corresponding above-described second threshold value;
Set a collection information entropy threshold value E
p, wherein, information entropy is implicit information entropy and accumulative total information entropy sum, corresponding above-described second information entropy of accumulative total information entropy, threshold value E
pCorresponding above-described first threshold;
The setting-up time window number | W| wherein, the corresponding above-described sub-time period of time window, is used for each sub-time period is carried out timing, i.e. the time interval of the sliding time of time window corresponding sub-time period of interval;
Setting-up time windows library table, corresponding with the time window of setting respectively, be used to load the data that produce in the corresponding time window;
Set the potential frequent item set set of output ITEM; Wherein, In the set each collection can pass through tlv triple { collection, accumulative total information entropy, implicit information entropy } expression, the corresponding above-described item collection set that is used to identify the information of obtaining of this potential frequent item set set ITEM.
In the above setting up procedure, threshold value E
P0Can be with reference to following factor setting:
1, to the probability distribution interval of all implicit information entropys of item collection carry out segmentation (p '
i, i=1,2 ..., n), obtaining a collection implicit information entropy threshold value E according to following formula then
P0:
Wherein:
The probability distribution of all implicit information entropys of item collection is interval, promptly representes the distribution of the implicit information entropy of all collection, and interval end points is respectively a minimum value and a maximal value of all implicit information entropys that collect;
Segmentation is carried out in probability distribution interval to all implicit information entropys of item collection; The probability distribution interval division that is about to confirm is a plurality of sub-ranges sections, and sub-range section number can be definite according to the probability distribution burst length of reality, for example; The probability distribution interval is [0; 0.5], then can this probability distribution is interval for being divided into 5 sub-interval sections, each sub-range segment length is 0.1;
P '
iBe the right end points of i sub-interval section, n is the number of sub-range section.
2, to (each time window all load operation) under the stable situation, to the probability distribution interval of all accumulative total information entropys of item collection carry out segmentation (p "
i, i=1,2 ..., n), obtaining a collection information entropy threshold value according to following formula then:
Wherein:
P "
iBe the right end points of i sub-interval section, n is the number of sub-range section.
After more than accomplishing, being provided with, as shown in Figure 5, from data, obtain information, mainly comprise the steps 501 to step 508:
In this step 502, calculate the implicit information entropy and pass through following formula:
The implicit information entropy=-p
iLn p
i(special, p
i=0, then the implicit information entropy is 0).
Wherein: p
i=data the total amount that data volume/the current time window is corresponding of answering at this set pair of current time window.
In this step 502, the selected attribute of data source property control generator promptly defines the data attribute that is used to the information of obtaining in advance.
In this step 502, the process of calculating the implicit information entropy of each item collection that each combinations of attributes obtains is promptly calculated the process of each item collection first information entropys through above-mentioned steps 102, and this process is described in detail in the above-described embodiments, repeats no more here.
Step 506, for the item collection among the ITEM, calculate the accumulative total information entropy of each item collection, with the implicit information entropy sum>=E of accumulative total information entropy with corresponding current time window
pItem collection item
t, with (item
t, accumulative total information entropy, the implicit information entropy of current time window) and corresponding entry collection among the substitute I TEM; Otherwise, this collection of deletion item in ITEM
t
In the step 506, calculate the accumulative total information entropy and pass through following formula:
The accumulative total information entropy=-p
I-1Ln p
I-1(p especially,
I-1=0, then adding up information entropy is 0);
Wherein: p
I-1The data total amount that=this set pair is answered before the current time window data total amount/before the current time window is corresponding.
Particularly, in above-mentioned steps 502 and the step 506, data volume can be shown through the bar numerical table of ticket writing.
Especially, in this step 507, when time moving window number＞| during W|; According to the queue structure's the most forward time window of replacement storehouse table; Be about to the data deletion in first time window storehouse table, will be current the | first time window storehouse of the data importing of W|+1 sub-time period is shown, and the like.
According to flow process shown in Figure 5, at first real time data stream is imported the 1st time window storehouse table, each collection item that at least one data attribute combination that the property control generator is exported obtains
1t, calculate the implicit information entropy, will satisfy implicit information entropy>=E
P0An item collection tlv triple incorporate ITEM into; Then the 2nd sub-time period in the corresponding data importing moment; Accomplish the importing of the 2nd time window storehouse table user call bill data stream; The same according to the 1st time window storehouse list processing (LISP) method; Calculate the implicit information entropy of each collection that at least one data attribute combination of property control generator output obtains in the 2nd the time window storehouse table earlier; Incorporate the item collection that is not included among the ITEM into ITEM, and further calculate the accumulative total information entropy, will satisfy (accumulative total information entropy+current time window implicit information entropy)>=E
pAn item collection tlv triple replace original frequent candidate tlv triple, otherwise, this frequent candidate tlv triple of deletion in ITEM.Repeat this process, up to disposing the | W| time window storehouse table all imports data with data stream time window storehouse table and finish this moment.Promptly obtaining information from the data of confirming that are used for the information of obtaining finishes.
Import constantly in next data stream; Promptly confirm the data that are used to the information of obtaining that make new advances and when handling; (be the 1st time window storehouse table this moment to clear history moment time window storehouse table at most, by that analogy), the latest data conductance gone into this time window storehouse table.
Said process is through calculating the information entropy of each item collection under each sliding time window; Thereby confirm potential frequent item set set ITEM; Reached continuous time window excavate frequent item set, but kept carrying out Knowledge Discovery result's purpose basically in global data base.This algorithm has been practiced thrift the complexity of the information of obtaining greatly.
Be treated to example with the excavation of the language data in the communications field below and specify an embodiment; Suppose in database table, to be provided with 5 temporary tables (corresponding time window oral thermometer of difference: time_win1, time_win2...time_win5; | W|=5), the main process of obtaining information is following:
1, through 1 time window temporary table of 6 concurrent importing 3 general-purpose family ticket inventories to the of passage time_win1;
2, client's brand of exporting according to attribute controller; Data attributes such as talk times, call type, conversation sub-district, duration of call average; Calculate the implicit information entropy that collects that each data attribute combination obtains; As: the Xiang Jiwei that combination obtains: the Global Link Olympic Games 88 clients _ 88 (client's brand) _ local call (call type) _ 23005_03133 (conversation sub-district coding) _ 300～600sec (duration of call average); This collection occurs 120 times in this base station cell (crowd) ticket inventory, this base station cell (crowd) at this moment between the window inventory of always conversing be recorded as 2500, then at this moment between window; This combinations of attributes implicit information entropy=-(120/2500) ln (120/2500)=0.146, if get E
P0=0.12, then can this collection " the Global Link Olympic Games 88 clients _ 88_ local call _ 23005_03133 (conversation sub-district coding) _ 300～600sec " be incorporated into ITEM;
3, at interval after 10 minutes (the time window length of setting), import new 2 time window temporary tables of 30,000 customer voice inventories to the time_win2 that produces;
4, according to client's brand of attribute controller output, data attributes such as talk times, call type, conversation sub-district, duration of call average calculate the implicit information entropy that collects that various combinations of attributes obtain, for the implicit information entropy>=E that calculates
P0And be not included in the item collection item among the ITEM
It, then with (item
It0; The implicit information entropy) incorporate ITEM into, as: the item collection among the 2nd time window temporary table time_win2 " M-ZONE standard client _ 156_ local call _ 23014_04165 (conversation sub-district coding) _ 0～300sec " be not included among the ITEM, but the implicit information entropy>=E of this collection
P0, then this combinations of attributes item is incorporated ITEM into.
For each collection item among the ITEM Already in
It, calculate the accumulative total information entropy of each item collection, and utilize (accumulative total information entropy+current time window implicit information entropy)>=E
pItem collection upgrade the corresponding entry collection among the ITEM, otherwise (accumulative total information entropy+current time window implicit information entropy)＜E among the deletion ITEM
pItem collection.As: " the Global Link Olympic Games 88 clients _ 88_ local call _ 23005_03133 (conversation sub-district coding)-_ 300～600sec " implicit information entropy on the 2nd time window=-(180/2500) ln (180/2500)=0.189438, then this combinations of attributes accumulative total information entropy+implicit information entropy=0.146+0.18.Notice that at this this is the 2nd window, the accumulative total information entropy can directly be quoted the implicit information entropy of the 1st window, but if the 2nd, 3,4,5 window then needs to calculate according to the computing formula of accumulative total information entropy.
According to said process, just can effectively keep the renewal of ITEM discipline collection.
5, and the like to calculate accomplishing 5 windows.
Then this data processing of obtaining information is finished; If need to handle the data that next group obtains information; Promptly think then will occupy the 6th time interval arrival time first window time_win1 at most and empty, import the 6th time window client inventory; Upgrade ITEM according to above the 4th step again, end until data stream.
Embodiment two
According to the embodiment of the invention two, a kind of method of obtaining information is provided, this method of obtaining information is optimized the item collection set that is used to identify the information of obtaining that obtains according to the foregoing description one on the basis of the foregoing description one.
Particularly, in the method for utilizing the foregoing description one to provide, after the first information entropy of answering according to each item set pair respectively and second information entropy are upgraded the item collection set (being above-mentioned steps 104) that is used to identify the information of obtaining, also further carry out following steps:
According to the data attribute to be extracted of setting and a data attribute that collection set each item set pair is answered that is used for identifying the information of obtaining, this item that is used for identifying the information of obtaining is gathered the item collection of preserving carry out packet transaction.
The technical scheme that provides according to this embodiment two; After the data that are used to the information of obtaining being handled (being called one-level handles) through the foregoing description; Be met the potential frequent item set set ITEM of information entropy condition; According to the requirement of data-flow analysis ITEM item collection further being classified this moment extracts the bigger frequent item set of practical significance (being called two stage treatment), thereby represents the information of obtaining more intuitively.Also can this analysis result be summarized as data knowledge and incorporate the special knowledge storehouse, with the Knowledge Discovery of the potential frequent item set of further promotion data stream (being valuable information).
For example, the information among the potential frequent item set set ITEM that processing obtains through one-level is as shown in the table:
The rule-based approach knowledge of above domination is carried out the knowledge alienation to be handled; That is: the management system of creating a file is carried out taxonomic organization and storage according to knowledge requirement group classification framework or standard; With the similarity that comprises knowledge between identification each item collection; Item collection storage file organizational form after obtaining the knowledge alienation and handling, as shown in the table:
Wherein: Class_1 ... the data attribute that Class_M is corresponding different respectively, can be a data attribute, also can be the combination of a plurality of data attributes, concrete data attribute is confirmed according to business demand.
The item that carries out taxonomic organization, storage according to knowledge requirement group classification framework collects the decision support knowledge that has begun to take shape satisfied different knowledge requirement persons, has reached the effect that real-time, effective knowledge generates.And the sorting item collection that obtains this moment can pass through internalization and handle, that is: warm, renewal that the decision support knowledge that will tentatively obtain and special knowledge storehouse historical knowledge carry out, and how effectively additional increase Xiang Jigeng is, the description of compound knowledge.In IT system, then focus on unstructured information in the special knowledge storehouse with hierarchical structure, list structure tissue and storage, and be aided with the proper knowledge meaning of one's words and describe and represent to the different knowledge demander.
The item collection that comprises among the ITEM that obtains according to the embodiment of the invention one and embodiment two can instruct for follow-up business.For example, the item collection that comprises among the ITEM:
The Global Link Olympic Games 88 clients _ 88_ local call _ 23005_03133 (conversation sub-district coding) _ 300～600sec;
The Global Link Olympic Games 88 clients _ 356_ local call _ 23005_03133 (conversation sub-district coding) _ 2300～300sec;
Concentrate at these two items; Can obtain a situation of the Global Link Olympic Games 88 set meal clients duration of call contrast in the base station cell of appointment in the time period of appointment; For the business personnel, obtained same set meal client duration of call contrast situation, given the interval division of the duration of call when being convenient to instruct its design set meal product in the appointed place.
Above-mentionedly be merely a simple illustration, in the practical application, can carry out the follow-up business adjustment with reference to the information among the ITEM that obtains flexibly according to concrete business demand.As, then can to obtain base station cell expansion plan reference information or the like, enumerate no longer one by one for the webmaster personnel through comprising the item collection of base station cell call volume and base station equipment utilization factor data attribute combination here.
Embodiment three
Corresponding with the foregoing description one, the embodiment of the invention three provides a kind of device that obtains information, and is as shown in Figure 6, and this device that obtains information comprises:
Wherein:
First information entropy is confirmed unit 602, is used for the first information entropy that each item set pair that said data that specified data loading unit 601 loads obtain by predefined at least one data attribute combination is answered;
Second information entropy is confirmed unit 603, is used for second information entropy that each item set pair is answered described in specified data loading unit 601 all sub-time period loaded data before the current sub-time period;
Collection set updating block 604 is used for confirming that according to first information entropy the first information entropy that each item set pairs that unit 602 is confirmed are answered and second information entropy confirm that second information entropys that unit 603 is confirmed upgrade the item collection set that is used to identify the information of obtaining.
As shown in Figure 7, in the preferred embodiment of the present invention, the data load unit 601 that device shown in Figure 6 comprises can specifically comprise:
Time period is divided module 601A, is used to confirm to be used to the data of the information of obtaining, and is divided into a plurality of sub-time period of constant duration the time period that produces said data; Wherein, saidly from each sub-time period, obtain the required duration of information interval greater than what equal to estimate to obtain;
Load-on module 601B is used for each sub-time period that time period division module 601A divides is carried out timing, after the current sub-time period finishes, loads the data that the current sub-time period produces.
As shown in Figure 8, in the preferred embodiment of the present invention, the first information entropy that device shown in Figure 6 comprises is confirmed unit 602, comprising:
The first data volume determination module 602A, the said data that are used for confirm loading meet the total amount of data of said data of data volume and the loading of the data attribute that this set pair answers;
First information entropy determination module 602B is used for the data volume and the said total amount of data that meet the data attribute that this set pair answers confirmed according to the first data volume determination module 602A, confirms the first information entropy that this set pair is answered.
Further, first information entropy determination module 602B shown in Figure 8 specifically is used for:
Confirm the said data volume of the data attribute that this set pair answers and the ratio of said total amount of data of meeting;
Utilize said ratio to multiply by the value that said ratio is taken the logarithm and obtained, the negative value of the product that obtains is confirmed as the first information entropy that this set pair is answered.
As shown in Figure 9, in the preferred embodiment of the present invention, second information entropy that device shown in Figure 6 comprises is confirmed unit 603, comprising:
The second data volume determination module 603A is used for confirming that data that all the sub-time periods before the current sub-time period produce meet the total amount of data of the data that data volume and all the sub-time periods before the current sub-time period of the data attribute that this set pair answers produce;
The second information entropy determination module 603B is used for the data volume and the said total amount of data that meet the data attribute that this set pair answers confirmed according to the second data volume determination module 603A, confirms second information entropy that this set pair is answered.
Further, the second information entropy determination module 603B shown in Figure 9 specifically is used for:
Confirm the said data volume of the data attribute that this set pair answers and the ratio of said total amount of data of meeting;
Utilize said ratio to multiply by the value that said ratio is taken the logarithm and obtained, the negative value of the product that obtains is confirmed as second information entropy that this set pair is answered.
Shown in figure 10, in the preferred embodiment of the present invention, the item collection set updating block 604 that device shown in Figure 6 comprises comprises:
First collection set determination module 604A; Be used for confirming that the corresponding first information entropy and the second information entropy sum reach first collection set of first threshold, first collection in wherein said first collection set is through corresponding first information entropy and second information entropy sign;
The first update module 604B, the item collection set that is used to identify the information of obtaining is upgraded in first the collection set that is used to utilize first collection set determination module 604A to confirm.
Further, the first update module 604B shown in Figure 10 specifically is used for:
When first collection in first collection set is included in the item collection set that is used for identifying the information of obtaining, utilize in said first collection set through the first information entropy of correspondence and the said corresponding entry collection that is used for identifying the item collection set of the information of obtaining of first collection replacement of second information entropy sign;
When first collection in first collection set is not included in the item collection set that is used for identifying the information of obtaining, delete the said corresponding entry collection that is used for identifying the item collection set of the information of obtaining.
Shown in figure 11, in the preferred embodiment of the present invention, the item collection set updating block 604 that device shown in Figure 10 comprises also comprises:
Second collection set determination module 604C; Be used for when sub-time period of last of time period of the said data of non-generation of current sub-time period; Confirm that corresponding first information entropy reaches second collection set of second threshold value, second collection in wherein said second collection set is through corresponding first information entropy sign;
The second update module 604D, the item collection set that is used to identify the information of obtaining is upgraded in second the collection set that is used to utilize said second collection set determination module to confirm.
Further, the second update module 604D shown in Figure 11 is used for:
With second collection that is not included in the item collection set that is used for identifying the information of obtaining in second collection set, add the said item collection set that is used for identifying the information of obtaining.
Should be appreciated that unit and module that the above device that obtains information comprises are merely the logical partitioning of carrying out according to the function of this terminal realization, in the practical application, can carry out the stack or the fractionation of said units and module.And the function that the device of the information of obtaining that this embodiment three provides is realized is corresponding one by one with the method flow of the information of obtaining that the foregoing description one provides; The more detailed treatment scheme that realizes for this device; In the foregoing description one, done detailed description, be not described in detail here.
Embodiment four
Corresponding with the foregoing description two, the embodiment of the invention four provides a kind of device that obtains information, and is shown in figure 12, and this device that obtains information further comprises on the basis of the device shown in Figure 6 that the foregoing description three provides:
Should be appreciated that unit and module that the above device that obtains information comprises are merely the logical partitioning of carrying out according to the function of this terminal realization, in the practical application, can carry out the stack or the fractionation of said units and module.And the function that the device of the information of obtaining that this embodiment four provides is realized is corresponding one by one with the method flow of the information of obtaining that the foregoing description two provides; The more detailed treatment scheme that realizes for this device; In the foregoing description two, done detailed description, be not described in detail here.
In the embodiment of the invention, the device of the information of obtaining that the foregoing description three and embodiment four provide can be disposed in unit, for example small-sized network environment or test macro; Also can in cluster, dispose; For example big-and-middle-sized network environment; Can the unit (being the unit that comprises among the embodiment three) that carry out the one-level processing be deployed in respectively in each processing node, can be deployed in the management node carrying out two stage treatment unit (being the unit that further comprises among the embodiment four).
Above-mentioned at least one technical scheme that provides through the embodiment of the invention; Confirm to be used to the data of the information of obtaining in advance; And the time period that will produce data is divided into a plurality of sub-time periods; Carry out to each sub-time period: load the data that the current sub-time period produces; Confirm in the loaded data first information entropy of answering, confirm second information entropy that each item set pair is answered in the data that all the sub-time periods before the current sub-time period produce, and upgrade the item collection set that is used to identify the information of obtaining according to the first information entropy and second information entropy that each item set pair is answered by each item set pair that predefined at least one data attribute combination obtains.Adopt this technical scheme, data based its generation time that will be used to the information of obtaining is divided into a plurality of sub-time periods, once only loads the data of a time period; Be used to identify the item collection set of the information of obtaining based on the Data Update that produces in this time period; Compared with prior art, the task distribution that will from data, obtain information is a plurality of execution, has significantly reduced the data volume of each processing; Thereby improved the efficient that information is obtained, and reduced system overhead.
Obviously, those skilled in the art can carry out various changes and modification to the present invention and not break away from the spirit and scope of the present invention.Like this, belong within the scope of claim of the present invention and equivalent technologies thereof if of the present invention these are revised with modification, then the present invention also is intended to comprise these changes and modification interior.
Claims (23)
1. a method of obtaining information is characterized in that, confirms to be used to the data of the information of obtaining in advance, and the time period that will produce said data is divided into a plurality of sub-time periods;
Carry out to each sub-time period:
Load the data that the current sub-time period produces;
The first information entropy that each item set pair that is obtained by predefined at least one data attribute combination in the said data of confirming to load is answered;
Second information entropy that each item set pair is answered described in the data of confirming to produce in current sub-time period sub-time periods of before all;
The first information entropy and second information entropy according to said each item set pair is answered are upgraded the item collection set that is used to identify the information of obtaining.
2. the method for claim 1 is characterized in that, the time period that produces said data is divided into a plurality of sub-time periods, comprising:
Be divided into a plurality of sub-time period of constant duration the time period that produces said data;
Wherein, saidly from each sub-time period, obtain the required duration of information interval greater than what equal to estimate to obtain.
3. the method for claim 1 is characterized in that, the first information entropy that each set pair is answered in the said data of confirming to load comprises:
The total amount of data of said data that meets data volume and the loading of the data attribute that this set pair answers in the said data that confirm to load;
According to said data volume and the said total amount of data that meets the data attribute that this set pair answers, confirm the first information entropy that this set pair is answered.
4. method as claimed in claim 3 is characterized in that, according to said data volume and the said total amount of data that meets the data attribute that this set pair answers, confirms the first information entropy that this set pair is answered, and comprising:
Confirm the said data volume of the data attribute that this set pair answers and the ratio of said total amount of data of meeting;
Utilize said ratio to multiply by the value that said ratio is taken the logarithm and obtained, the negative value of the product that obtains is confirmed as the first information entropy that this set pair is answered.
5. the method for claim 1 is characterized in that, second information entropy that each set pair is answered in the data of confirming to produce in current sub-time period sub-time periods of before all comprises:
Confirm to meet in the data that all the sub-time periods before the current sub-time period produce the total amount of data of the data that data volume and all the sub-time periods before the current sub-time period of the data attribute that this set pair answers produce;
According to said data volume and the said total amount of data that meets the data attribute that this set pair answers, confirm second information entropy that this set pair is answered.
6. method as claimed in claim 5 is characterized in that, according to said data volume and the said total amount of data that meets the data attribute that this set pair answers, confirms second information entropy that this set pair is answered, and comprising:
Confirm the said data volume of the data attribute that this set pair answers and the ratio of said total amount of data of meeting;
Utilize said ratio to multiply by the value that said ratio is taken the logarithm and obtained, the negative value of the product that obtains is confirmed as second information entropy that this set pair is answered.
7. like claim 3 or 5 described methods, it is characterized in that said data volume is:
The bar number of data recording; Or
The storage size of data occupancy.
8. the method for claim 1 is characterized in that, upgrades the item collection set that is used to identify the information of obtaining according to the first information entropy and second information entropy that said each item set pair is answered, comprising:
Confirm that the corresponding first information entropy and the second information entropy sum reach first collection set of first threshold, first collection in wherein said first collection set is through corresponding first information entropy and second information entropy sign;
Utilize said first collection set to upgrade the item collection set that is used to identify the information of obtaining.
9. method as claimed in claim 8 is characterized in that, utilizes said first collection set to upgrade the item collection set that is used to identify the information of obtaining, and comprising:
Be included in the item collection that is used for identifying the information of obtaining as if first collection in first collection set and gather, then utilize in said first collection set and replace the said corresponding entry collection that is used for identifying the item collection set of the information of obtaining through the first information entropy of correspondence and first collection of second information entropy sign;
If first collection during first collection gathered is not included in the item collection set that is used for identifying the information of obtaining, then deletes the said item that is used for identifying the information of obtaining and collect the corresponding entry collection of set.
10. method as claimed in claim 8; It is characterized in that; If last sub-time period in the time period of the said data of non-generation of current sub-time period; The first information entropy and second information entropy of then answering according to said each item set pair are upgraded the item collection set that is used to identify the information of obtaining, and also comprise:
Confirm that corresponding first information entropy reaches second collection set of second threshold value, second collection in wherein said second collection set is through corresponding first information entropy sign;
Utilize said second collection set to upgrade the item collection set that is used to identify the information of obtaining.
11. method as claimed in claim 10 is characterized in that, utilizes said second collection set to upgrade the item collection set that is used to identify the information of obtaining, and comprising:
With second collection that is not included in the item collection set that is used for identifying the information of obtaining in second collection set, add the said item collection set that is used for identifying the information of obtaining.
12. the method for claim 1 is characterized in that, the first information entropy of answering according to said each item set pair respectively and second information entropy also comprise after upgrading the item collection set that is used to identify the information of obtaining:
Gather the data attribute that each item set pair is answered according to data attribute of setting to be extracted and the said item collection that is used for identifying the information of obtaining, the collection that the said item collection set that is used for identifying the information of obtaining is preserved carries out packet transaction.
13. a device that obtains information is characterized in that, comprising:
The data load unit is used to confirm to be used to the data of the information of obtaining, and the time period that will produce said data be divided into a plurality of sub-time periods, and load the data that the current sub-time period produces;
First information entropy is confirmed the unit, is used for confirming the first information entropy that each item set pair that said data that said data load unit loads are obtained by predefined at least one data attribute combination is answered;
Second information entropy is confirmed the unit, is used for second information entropy of confirming that each item set pair is answered described in said data load unit all sub-time period loaded data before the current sub-time period;
Collection set updating block is used for confirming that according to said first information entropy first information entropy that each item set pair that the unit is confirmed is answered and said second information entropy confirm that second information entropy of confirming the unit upgrades the item collection set that is used to identify the information of obtaining.
14. device as claimed in claim 13 is characterized in that, said data load unit comprises:
Time period is divided module, is used to confirm to be used to the data of the information of obtaining, and is divided into a plurality of sub-time period of constant duration the time period that produces said data; Wherein, saidly from each sub-time period, obtain the required duration of information interval greater than what equal to estimate to obtain;
Load-on module is used for each sub-time period of said time period division Module Division is carried out timing, after the current sub-time period finishes, loads the data that the current sub-time period produces.
15. device as claimed in claim 13 is characterized in that, said first information entropy is confirmed the unit, comprising:
The first data volume determination module, the said data that are used for confirm loading meet the total amount of data of said data of data volume and the loading of the data attribute that this set pair answers;
First information entropy determination module is used for the data volume and the said total amount of data that meet the data attribute that this set pair answers confirmed according to the said first data volume determination module, confirms the first information entropy that this set pair is answered.
16. device as claimed in claim 15 is characterized in that, said first information entropy determination module specifically is used for:
Confirm the said data volume of the data attribute that this set pair answers and the ratio of said total amount of data of meeting;
Utilize said ratio to multiply by the value that said ratio is taken the logarithm and obtained, the negative value of the product that obtains is confirmed as the first information entropy that this set pair is answered.
17. device as claimed in claim 13 is characterized in that, said second information entropy is confirmed the unit, comprising:
The second data volume determination module is used for confirming that data that all the sub-time periods before the current sub-time period produce meet the total amount of data of the data that data volume and all the sub-time periods before the current sub-time period of the data attribute that this set pair answers produce;
The second information entropy determination module is used for the data volume and the said total amount of data that meet the data attribute that this set pair answers confirmed according to the said second data volume determination module, confirms second information entropy that this set pair is answered.
18. device as claimed in claim 17 is characterized in that, the said second information entropy determination module specifically is used for:
Confirm the said data volume of the data attribute that this set pair answers and the ratio of said total amount of data of meeting;
Utilize said ratio to multiply by the value that said ratio is taken the logarithm and obtained, the negative value of the product that obtains is confirmed as second information entropy that this set pair is answered.
19. device as claimed in claim 13 is characterized in that, said collection set updating block comprises:
First collection set determination module; Be used for confirming that the corresponding first information entropy and the second information entropy sum reach first collection set of first threshold, first collection in wherein said first collection set is through corresponding first information entropy and second information entropy sign;
First update module, the item collection set that is used to identify the information of obtaining is upgraded in first the collection set that is used to utilize said first collection set determination module to confirm.
20. device as claimed in claim 19 is characterized in that, said first update module specifically is used for:
When first collection in first collection set is included in the item collection set that is used for identifying the information of obtaining, utilize in said first collection set through the first information entropy of correspondence and the said corresponding entry collection that is used for identifying the item collection set of the information of obtaining of first collection replacement of second information entropy sign;
When first collection in first collection set is not included in the item collection set that is used for identifying the information of obtaining, delete the said corresponding entry collection that is used for identifying the item collection set of the information of obtaining.
21. device as claimed in claim 19 is characterized in that, said collection set updating block also comprises:
Second collection set determination module; Be used for when sub-time period of last of time period of the said data of non-generation of current sub-time period; Confirm that corresponding first information entropy reaches second collection set of second threshold value, second collection in wherein said second collection set is through corresponding first information entropy sign;
Second update module, the item collection set that is used to identify the information of obtaining is upgraded in second the collection set that is used to utilize said second collection set determination module to confirm.
22. device as claimed in claim 21 is characterized in that, said second update module is used for:
With second collection that is not included in the item collection set that is used for identifying the information of obtaining in second collection set, add the said item collection set that is used for identifying the information of obtaining.
23. device as claimed in claim 13 is characterized in that, also comprises:
Packet processing unit; Be used for after the first information entropy of answering according to said each item set pair respectively and second information entropy are upgraded the item collection set that is used to identify the information of obtaining; Gather the data attribute that each item set pair is answered according to data attribute of setting to be extracted and the said item collection that is used for identifying the information of obtaining, the collection that the said item collection set that is used for identifying the information of obtaining is preserved carries out packet transaction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010292828 CN102411594B (en) | 2010-09-25 | 2010-09-25 | Method and device for obtaining information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010292828 CN102411594B (en) | 2010-09-25 | 2010-09-25 | Method and device for obtaining information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102411594A true CN102411594A (en) | 2012-04-11 |
| CN102411594B CN102411594B (en) | 2013-06-26 |
Family
ID=45913668
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201010292828 Active CN102411594B (en) | 2010-09-25 | 2010-09-25 | Method and device for obtaining information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102411594B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104809140A (en) * | 2014-01-29 | 2015-07-29 | 中国银联股份有限公司 | Method and system for counting trading data |
| CN105975623A (en) * | 2016-05-30 | 2016-09-28 | 中国能源建设集团湖南火电建设有限公司 | Method and system for obtaining organization information by means of query expressions |
| CN107290967A (en) * | 2017-08-07 | 2017-10-24 | 合肥工业大学 | The automatic loading method and system of a kind of hydraulic press service function |
| CN107729571A (en) * | 2017-11-23 | 2018-02-23 | 北京天广汇通科技有限公司 | Relation finds method and apparatus |
| CN109739880A (en) * | 2018-12-20 | 2019-05-10 | 中国联合网络通信集团有限公司 | Method for computing data and device |
| CN113168414A (en) * | 2018-10-11 | 2021-07-23 | 维萨国际服务协会 | Systems, methods, and computer program products for load balancing to process large data sets |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3491147B2 (en) * | 2000-03-30 | 2004-01-26 | Ｊｆｅスチール株式会社 | Defect detection method and defect detection device |
| US6918099B2 (en) * | 2003-02-19 | 2005-07-12 | Sun Microsystems, Inc. | Method and system for entropy driven verification |
| CN1770094A (en) * | 2005-10-17 | 2006-05-10 | 浙江大学 | High quality true random number generator |
| CN101378394A (en) * | 2008-09-26 | 2009-03-04 | 成都市华为赛门铁克科技有限公司 | Detection defense method for distributed reject service and network appliance |
| CN101494049A (en) * | 2009-03-11 | 2009-07-29 | 北京邮电大学 | Method for extracting audio characteristic parameter of audio monitoring system |
-
2010
- 2010-09-25 CN CN 201010292828 patent/CN102411594B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3491147B2 (en) * | 2000-03-30 | 2004-01-26 | Ｊｆｅスチール株式会社 | Defect detection method and defect detection device |
| US6918099B2 (en) * | 2003-02-19 | 2005-07-12 | Sun Microsystems, Inc. | Method and system for entropy driven verification |
| CN1770094A (en) * | 2005-10-17 | 2006-05-10 | 浙江大学 | High quality true random number generator |
| CN101378394A (en) * | 2008-09-26 | 2009-03-04 | 成都市华为赛门铁克科技有限公司 | Detection defense method for distributed reject service and network appliance |
| CN101494049A (en) * | 2009-03-11 | 2009-07-29 | 北京邮电大学 | Method for extracting audio characteristic parameter of audio monitoring system |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104809140A (en) * | 2014-01-29 | 2015-07-29 | 中国银联股份有限公司 | Method and system for counting trading data |
| CN105975623A (en) * | 2016-05-30 | 2016-09-28 | 中国能源建设集团湖南火电建设有限公司 | Method and system for obtaining organization information by means of query expressions |
| CN107290967A (en) * | 2017-08-07 | 2017-10-24 | 合肥工业大学 | The automatic loading method and system of a kind of hydraulic press service function |
| CN107290967B (en) * | 2017-08-07 | 2018-08-31 | 合肥工业大学 | A kind of automatic loading method and system of hydraulic press service function |
| US10571871B2 (en) | 2017-08-07 | 2020-02-25 | Hefei University Of Technology | Automatic loading system and method for service functions of hydraulic machine |
| CN107729571A (en) * | 2017-11-23 | 2018-02-23 | 北京天广汇通科技有限公司 | Relation finds method and apparatus |
| CN113168414A (en) * | 2018-10-11 | 2021-07-23 | 维萨国际服务协会 | Systems, methods, and computer program products for load balancing to process large data sets |
| CN109739880A (en) * | 2018-12-20 | 2019-05-10 | 中国联合网络通信集团有限公司 | Method for computing data and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102411594B (en) | 2013-06-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102411594B (en) | Method and device for obtaining information | |
| CN110750650A (en) | Construction method and device of enterprise knowledge graph | |
| CN109191312A (en) | A kind of anti-fraud air control method and device of Claims Resolution | |
| CN106030579A (en) | A method, system and computer program for scanning a plurality of storage regions within memory for a specified quantity of results | |
| CN106022708A (en) | Method for predicting employee resignation | |
| CN107832291A (en) | Client service method, electronic installation and the storage medium of man-machine collaboration | |
| CN110674206A (en) | Knowledge sharing platform applied to enterprises | |
| CN110178128A (en) | It is indicated using the bitmap of optimization to manage extensive incidence set | |
| Yavari et al. | Integrated decision making for parts ordering and scheduling of jobs on two-stage assembly problem in three level supply chain | |
| Jin et al. | DAKA: design activity knowledge acquisition through data-mining | |
| CN111752541B (en) | Payment routing method based on Rete algorithm | |
| binti Oseman et al. | Data mining in churn analysis model for telecommunication industry | |
| CN111292201A (en) | Method for pushing field operation and maintenance information of power communication network based on Apriori and RETE | |
| CN112507213B (en) | Method for recommending optimized system scheme based on behavior big data analysis | |
| CN115630151A (en) | Infrastructure engineering knowledge management method, system and storage medium | |
| CN104573034A (en) | CDR call ticket based user group division method and system | |
| Levin et al. | Towards modular redesign of networked system | |
| CN114154048A (en) | Method and device for constructing recommendation model, electronic equipment and storage medium | |
| CN111428756A (en) | Planning data fusion real-time state method and device based on time series information entropy | |
| Chen et al. | A guided memetic algorithm with probabilistic models | |
| Vu et al. | FTKHUIM: A Fast and Efficient Method for Mining Top-K High-Utility Itemsets | |
| Mason et al. | Using knowledge to transform enterprises | |
| CN113641654B (en) | Marketing treatment rule engine method based on real-time event | |
| CN115577050B (en) | Construction method of electric charge digital application platform | |
| Xu et al. | Churn prediction in telecom using a hybrid two-phase feature selection method |
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 |