CN103678514A - Business trend prediction method and system - Google Patents
Business trend prediction method and system Download PDFInfo
- Publication number
- CN103678514A CN103678514A CN201310611664.7A CN201310611664A CN103678514A CN 103678514 A CN103678514 A CN 103678514A CN 201310611664 A CN201310611664 A CN 201310611664A CN 103678514 A CN103678514 A CN 103678514A
- Authority
- CN
- China
- Prior art keywords
- business datum
- predetermined period
- statistical history
- business
- prediction
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a business trend prediction method and system. The business trend prediction method includes: acquiring a statistics historical business data sequence of prediction items with prediction periods as units from a business center system; subjecting the statistics historical business data sequence to periodization removal processing according to business periods of the prediction items to acquire the processed statistic historical business data sequence; according to the processed statistic historical business data sequence, adopting a time sequence prediction method to have business trend within selected prediction periods of the prediction items predicted to acquire preliminary predication data; subjecting the preliminary prediction data to periodization restoration processing corresponding to the periodization removal processing to acquire the prediction data of the selected prediction periods. Through prediction of development trend on online business, operators can be assisted in knowing variation trend of business in advance and discovering unusual situations of the business, so that the operators can find coping strategies in time to improve service quality and competitiveness of the same.
Description
Technical field
The present invention relates to data analysis and signal process field, relate in particular to a kind of business trend forecasting method and system, business trend forecasting method of the present invention and system are particularly suitable for telecommunications industry.
Background technology
Fast development along with mobile interconnected industry, mobile phone becomes the necessary article of people's daily life gradually, the additional function of mobile phone from strength to strength, therefore, make its call function weakened gradually, along with the appearance of the diverse network communication technology, the competition that telecommunications industry faces and challenge are also increasing, in order to seek better development, each company has released omnifarious mobile phone value-added service one after another, and hope can be stablized the status of oneself under network tide of today impacts.The business of releasing for company, if the quantity that can break down to user's attention rate of business, service fulfillment quantity, business cancellation quantity, business, the sales volume of business etc. is made certain prediction in advance, can carry out certain assessment to the promotion effect of business and user satisfaction, and then help enterprise to understand in time user's request, this is conducive to the business that operator optimizes oneself better, robs a step first chance in the competition in market.For business on line, if can the development of business be monitored in real time, to reporting to the police timely such as the abnormal conditions such as fault sudden fast rush that are a certain business, make operator can take timely counter-measure, can reduce unnecessary loss.
The forecasting techniques of telecommunications industry application is at present mainly the prediction based on call volume, calling amount to the hotline of different time sections is predicted, thereby for job placement, recruitment, the shift report of call center, arrange etc. Data support is provided, thereby rationally control the operation cost of call center.For example, the patent documentation that publication number is CN101132447A is with regard to the hot line calling Forecasting Methodology of disclosed a kind of large-scale call center.
The prediction of this kind based on call volume do not have actual directive significance to carrying out of business, because, for telecom operators, these are not the key of market competition, real key is to provide more satisfied service for user, when improving consumer loyalty degree, attract new user, and provide better mobile phone value-added service just can accomplish this point, therefore, for telecom operators, what is more important can help it to understand timely business development trend, that finds timely to occur in service operation is abnormal, and then find timely countermeasure, promote the service quality of oneself, improve the competitive power of self.
Summary of the invention
One object of the present invention is to overcome deficiency of the prior art, and a kind of business trend forecasting method and system are provided, and can, based on historical business datum automatically for operator provides business trend prediction, help operator to understand in time business development trend.
For achieving the above object, the technical solution used in the present invention is: a kind of business trend forecasting method, comprising:
The predetermined period of take from business centring system obtains the statistical history business datum sequence of prediction term as unit;
To described statistical history business datum sequence, according to the service period of described prediction term, go periodization to process, obtain the statistical history business datum sequence after processing;
According to the statistical history business datum sequence after processing, the method that adopts time series forecasting to described prediction term the business trend in selected predetermined period predict, obtain tentative prediction data;
Described tentative prediction data are carried out going periodization to process corresponding reduction cycleization processing with described, obtain the predicted data of selected predetermined period.
Preferably, to described statistical history business datum sequence, going periodization to process comprises: described in employing difference method carries out described statistical history business datum sequence, go periodization to process;
Described tentative prediction data are carried out to reduction cycleization processing to be comprised:
From described statistical history business datum sequence, obtain the statistical history business datum of upper first-class order predetermined period, wherein, described first-class order predetermined period is for be positioned at predetermined period of same sequence position according to time sequencing and selected predetermined period in a upper service period, and a described upper service period is the service period at selected predetermined period place;
Calculate described tentative prediction data and upper first-class order predetermined period statistical history business datum and as described predicted data.
Preferably, the method of going periodization to process described in employing difference method carries out comprises: if predetermined period that described service period comprises varying number, the statistical history business datum of choosing adjacent two predetermined period of same sequence position from each little service period is carried out linear interpolation, so that all little service period are supplied as large service period, statistical history business datum sequence after formation is supplied, described in carrying out with the statistical history business datum sequence to after described supplying, go periodization to process, wherein, described large service period is the maximum service period of quantity that comprises predetermined period, little service period is the service period that the quantity that comprises predetermined period is less than large service period.
Preferably, described business trend forecasting method also comprises: from business centring system, obtain the original historical business datum sequence of prediction term, by described original historical business datum sequence is carried out to the service period that fast fourier transform obtains described prediction term;
The method of obtaining described statistical history business datum sequence from business centring system is:
Described original historical business datum sequence be take to the statistical history business datum that predetermined period occurs in each predetermined period as unit statistics, to form described statistical history business datum sequence.
Preferably, described business trend forecasting method also comprises:
Obtaining described prediction term after the predicted data of selected predetermined period, judge whether the state of described prediction term occurs extremely; If so, to User Alarms.
Preferably, describedly judge whether the state of described prediction term occurs extremely comprising:
The predetermined period of take from described service center system obtains the current statistical history business datum of described prediction term as unit, and described current statistical history business datum is described prediction term actual business datum occurring in selected predetermined period;
Calculate the difference of the selected current statistical history business datum of predetermined period and the predicted data of selected predetermined period;
If described difference surpasses default warning value, determine that the state of described prediction term occurs abnormal.
Another object of the present invention has been to provide a kind of business trend predicting system that can help operator to understand timely business development trend.
The technical solution used in the present invention is: a kind of business trend predicting system, comprising:
Data acquisition module, obtains the statistical history business datum sequence of prediction term for take predetermined period from business centring system as unit;
Go periodization module, for going periodization to process described statistical history business datum sequence according to the service period of described prediction term, obtain the statistical history business datum after processing;
Prediction module, for according to the statistical history business datum after processing, the method that adopts time series forecasting to described prediction term the business trend in selected predetermined period predict, obtain tentative prediction data;
Data restoring module, goes periodization to process corresponding reduction cycleization processing for described tentative prediction data are carried out with described, obtains the predicted data of selected predetermined period.
Preferably, described in, go periodization module to comprise:
Difference unit, for adopting difference method to go described in described statistical history business datum sequence is carried out according to the service period of described prediction term periodization to process;
Described data restoring module comprises:
Data capture unit, for obtain the statistical history business datum of first-class order predetermined period from described data acquisition module, wherein, described first-class order predetermined period is for be positioned at predetermined period of same sequence position according to time sequencing and selected predetermined period in a upper service period, and a described upper service period is the service period at selected predetermined period place;
Reduction unit, for calculate described tentative prediction data and upper first-class order predetermined period statistical history business datum and as the output of described predicted data.
Preferably, described in, go periodization module also to comprise:
Interpolating unit, be used in the situation that predetermined period that described service period comprises varying number, from each little service period, choose the statistical history business datum of adjacent two predetermined period of same sequence position and carry out linear interpolation, so that all little service period are supplied as large service period, statistical history business datum sequence after formation is supplied, statistical history business datum sequence for difference unit after to described supplying goes periodization to process described in carrying out, wherein, described large service period is the maximum service period of quantity that comprises predetermined period, little service period is the service period that the quantity that comprises predetermined period is less than large service period.
Preferably, described data acquisition module comprises:
Raw data acquiring unit, for obtaining original historical business datum sequence from business centring system;
Statistic unit, for described original historical business datum sequence being take to predetermined period as the statistical history business datum that unit statistics occurs in each predetermined period, forms described statistical history business datum sequence;
Described business trend predicting system also comprises:
Service period computing module, carries out fast fourier transform for the described original historical business datum sequence that raw data acquiring unit is provided, to obtain the service period of described prediction term.
Preferably, described business trend predicting system also comprises:
Judge module, for obtaining described prediction term after the predicted data of selected predetermined period, judges whether the state of described prediction term occurs extremely;
Alarm module, after occurring extremely for the state in judge module judgement prediction term, to User Alarms.
Preferably, described judge module comprises:
Current data acquiring unit, for obtain the current statistical history business datum of described prediction term from described data acquisition module, described current statistical history business datum is described prediction term actual business datum occurring in selected predetermined period;
Computing unit, for calculating the difference of the selected current statistical history business datum of predetermined period and the predicted data of selected predetermined period;
Comparing unit, for described difference and default warning value are compared, if described difference surpasses default warning value, determines that the state of described prediction term occurs abnormal.
Business trend forecasting method provided by the invention and system, by predicting going up the development trend of line service, and the early warning to ANOMALOUS VARIATIONS, help operator to understand in advance the variation tendency of business, the abnormal conditions of discovery business, and then make operator can find timely countermeasure, and promote the service quality of oneself, improve the competitive power of self.
Accompanying drawing explanation
Fig. 1 shows according to the main flow chart of business trend forecasting method of the present invention;
Fig. 2 shows a kind of specific implementation method process flow diagram that carries out reduction cycle processing in Fig. 1;
Fig. 3 shows the main flow chart that increases warning function on the basis of the business trend forecasting method shown in Fig. 1;
Fig. 4 shows a kind of specific implementation method process flow diagram that goes periodization to process in Fig. 1;
Fig. 5 shows according to the frame principle figure of business trend predicting system of the present invention;
Fig. 6 shows a kind of concrete enforcement structure of each module in Fig. 5;
Fig. 7 shows the frame principle figure that increases warning function on the basis of the business trend predicting system described in Fig. 5;
Fig. 8 shows a kind of concrete enforcement structure of each module in Fig. 7.
Embodiment
Describe embodiments of the invention below in detail, the example of described embodiment is shown in the drawings, and wherein same or similar label represents same or similar element or has the element of identical or similar functions from start to finish.Below by the embodiment being described with reference to the drawings, be exemplary, only for explaining the present invention, and can not be interpreted as limitation of the present invention.
As shown in Figure 1, the business trend forecasting method that the embodiment of the present invention provides, comprises the following steps:
Step S1: the predetermined period of take from business centring system obtains the statistical history business datum sequence of prediction term as unit.
Illustrate, if the selected predetermined period of user is every day, it is the business datum that user wishes every day predict future a period of time, and the data that the original historical business datum sequence of the prediction term that service center system provides is also every day, when statistics is just without calculating like this, can be directly using original historical business datum sequence as statistical history business datum sequence, if the selected predetermined period of user is for weekly, and the data that the original historical business datum sequence of the prediction term that service center system provides is every day, need to calculate data weekly according to original historical business datum sequence, to form the statistical history business datum sequence of Yi Yizhouwei unit, this step need to be processed all original historical business datum sequences when system is predicted for the first time, and prediction after this is only processed untreated original historical business datum sequence, at this, for telecommunications industry, this service center system is call center system, and this prediction term such as the user's attention rate, service fulfillment quantity, business that are certain business are cancelled in the quantity that quantity, business breaks down etc.
Step S2: go periodization to process described statistical history business datum sequence according to the service period of described prediction term, obtain the statistical history business datum sequence after processing.
At this, the data of business all present the business of certain periodicity, particularly telecommunications industry conventionally, and the service period of indication is the periodicity that the original historical business datum of prediction term itself presents; For different prediction term, service period is not quite similar, and directly according to statistical history business datum sequence, cannot effectively predict, therefore, need to before prediction, remove the periodicity of statistical history business datum sequence.In addition, when system is predicted for the first time, need to go periodization to process to all statistical history business datums, on this basis, prediction afterwards only needs the statistical history business datum to newly obtaining to go periodization to process.
Step S3: according to the statistical history business datum sequence after processing.
The method that adopts the time series forecasting be for example arma modeling to described prediction term the business trend in selected predetermined period predict, obtain tentative prediction data;
Step S4: described tentative prediction data are carried out going periodization to process corresponding reduction cycleization processing with described, obtain the predicted data of selected predetermined period.
At this, because the tentative prediction data that obtain in step S3 are according to going the statistical history business datum sequence after periodization is processed to obtain, therefore, need to be reverted to periodization and be processed previous value, just can obtain the actual predicted data of prediction term.
For above step S2, conventionally can adopt difference method to go periodization to process to statistical history business datum sequence, that answers in contrast carries out reduction cycle processing as shown in Figure 2 to tentative prediction data, can comprise:
Step S401: the statistical history business datum of obtaining upper first-class order predetermined period from statistical history business datum sequence, wherein, upper first-class order predetermined period is for being positioned at predetermined period of same sequence position according to time sequencing and selected predetermined period in the upper service period in selected predetermined period place service period.
For the ease of understanding, provide following embodiment 1:
The prediction term that user selects is the quantity of opening of flow business, predetermined period that user selects is every day, the service period of prediction term is one month, setting today is November 15, if user need to know November 16 flow business open quantity, be so selected predetermined period on November 16, upper first-class order predetermined period is October 16, if user need to know November 16,17 and 18 flow business open quantity, upper first-class order predetermined period is respectively October 16,17 days and 18 days.
Step S402: calculate tentative prediction data and upper first-class order predetermined period statistical history business datum and as predicted data.
Adopting difference method to go in the process of periodization processing, if predetermined period that service period comprises varying number, conventionally adopt the method for interpolation to supply, to meet the condition of difference, be specially: the statistical history business datum of choosing adjacent two predetermined period of same sequence position from each little service period is carried out linear interpolation, so that all little service period are supplied as large service period, statistical history business datum sequence after formation is supplied, again the statistical history business datum sequence after supplying being carried out to n jump divides, completing periodization processes, wherein, n is the quantity (in this n round numbers) of predetermined period of containing of large business periodic packets, above large service period is the maximum service period of quantity that comprises predetermined period, little service period is the service period that the quantity that comprises predetermined period is less than large service period.
In most cases, if predetermined period that service period comprises varying number, so large service period and little service period only differ from a predetermined period, if there is the situation of the little service period of poor different predetermined period numbers, can carry out so many wheels and choose the process that linear interpolation is carried out in same sequence position, until all little service period are supplied as large service period.
For the ease of understanding, still adopt above embodiment 1 to describe, service period is one month, so large service period just comprises 31 days (i.e. 31 predetermined period), little service period comprises 30 days (i.e. 30 predetermined period), so for example select the statistical history business datum between 14 of little service period days and 15 days to carry out linear interpolation, the mean value of getting two days is inserted between the statistical history business datum of 14 days and 15 days, form the statistical history business datum sequence that each service period all comprises the statistical history business datum of 31 days, use afterwards the statistical history business datum on November 15 (owing to comprising 30 days November, therefore, actually after interpolation within 15th, in statistical history business datum sequence, become the statistical history business datum of 16 days) deduct the statistical history business datum on October 16, the like, statistical history business datum sequence is completed to periodization to be processed.If carried out once prediction on November 14, data before November 14 all go periodization later so, only need to by the statistical history business datum on November 15, deduct the statistical history business datum on October 16 just passable, so just realize and gone periodization.For more special February, can insert again in the middle of the month statistical history business datum of 1 day or 2 days.
Service period for above prediction term is calculated when predicting for the first time by system conventionally, if user has been known the service period of prediction term by other approach, so just without carrying out following calculating.For this reason, as shown in Figure 4, when predicting for the first time, can obtain by the following method the service period of prediction term:
Step S201: judge that whether service period is known, carry out above step S2 as known;
Step S202: the original historical business datum sequence of the prediction term obtaining in obtaining step S1;
Step S203: whether the original historical business datum sequence that determining step S202 obtains meets the needs of the service period of calculating prediction term? perform step in this way S204, as otherwise execution step S202, continue to obtain the original historical business datum sequence with how original historical business datum;
Step S204: obtain the service period of prediction term by original historical business datum sequence being carried out to fast fourier transform, carry out afterwards above step S2.
On the basis of above business trend forecasting method, also can carry out the early warning of business trend to prediction term, for this reason, the method also comprises:
Obtaining described prediction term after the predicted data of selected predetermined period, judge whether the state of described prediction term occurs extremely; If so, to User Alarms.
Whether the state that as shown in Figure 3, judges prediction term occurs that abnormal method can comprise the following steps:
Step S5: the predetermined period of take from business centring system obtains the current statistical history business datum of prediction term as unit, current statistical history business datum is described prediction term actual business datum occurring in selected predetermined period, at this, selected predetermined period has been past tense owing to carrying out early warning, therefore the statistical history business datum sequence of, obtaining in step S1 has comprised above-mentioned current statistical history business datum;
Step S6: the difference of calculating the selected current statistical history business datum of predetermined period and the predicted data of selected predetermined period;
Step S7: judge that whether described difference surpasses default warning value, the state of determining in this way described prediction term occurs abnormal, as otherwise the state that continues monitoring and forecasting item whether occur extremely.
The embodiment of the present invention also provides a kind of business trend predicting system that can realize above business trend forecasting method, and as shown in Figure 5, it comprises:
Data acquisition module 1, obtains the statistical history business datum sequence of prediction term for take predetermined period from business centring system as unit;
Go periodization module 2, for going periodization to process described statistical history business datum sequence according to the service period of described prediction term, obtain the statistical history business datum after processing;
Prediction module 3, for according to the statistical history business datum after processing, the method that adopts time series forecasting to prediction term the business trend in selected predetermined period predict, obtain tentative prediction data;
Data restoring module 4, processes corresponding reduction cycleization processing for tentative prediction data are carried out with going periodization, obtains the predicted data of selected predetermined period.
As shown in Figure 6, above data acquisition module 1 can comprise:
Raw data acquiring unit 11, for obtaining original historical business datum sequence from business centring system;
Statistic unit 12, for original historical business datum sequence being take to predetermined period as the statistical history business datum that unit statistics occurs in each predetermined period, forms described statistical history business datum sequence.
As shown in Figure 6, with the periodization module 2 that gets on, can comprise difference unit 21, it adopts difference method according to the service period of prediction term, to go periodization to process statistical history business datum sequence.Accordingly, above data restoring module 4 can comprise:
Data capture unit 41, for obtaining the statistical history business datum of upper first-class order predetermined period from data acquisition module 1 (being specially its statistic unit 12);
Reduction unit 42, for calculate tentative prediction data and upper first-class order predetermined period statistical history business datum and as predicted data.
If predetermined period that the service period of prediction term comprises varying number, when adopting difference method to go periodization to process, as shown in Figure 6, this goes periodization module also to comprise:
Interpolating unit 22, be used in the situation that predetermined period that described service period comprises varying number, from each little service period, choose the statistical history business datum of adjacent two predetermined period of same sequence position and carry out linear interpolation, so that all little service period are supplied as large service period, form the statistical history business datum sequence after supplying;
Statistical history business datum sequence after above difference unit adopts difference method to described supplying goes periodization to process described in carrying out according to the service period of described prediction term.
If need system to calculate voluntarily the service period of prediction term when the prediction of carrying out for the first time prediction term, as shown in Figure 6, this goes periodization module also to comprise:
Service period computing unit 23, for the original historical business datum sequence that raw data acquiring unit 11 is provided, carry out fast fourier transform, to obtain the service period of prediction term, at this, if the original historical business datum sequence of obtaining current is not enough to the computing service cycle, continues to obtain from raw data acquiring unit 11 the original historical business datum sequence that comprises how original historical business datum.
As shown in Figure 7, this business trend predicting system also can comprise:
Alarm module 6, after occurring extremely for the state in judge module 5 judgement prediction term, to User Alarms.
As shown in Figure 8, above judge module 5 can comprise:
Current data acquiring unit 51, for being specially statistic unit 12 from data acquisition module 1() obtain the current statistical history business datum of prediction term;
Comparing unit 53, for difference and default warning value are compared, if difference surpasses default warning value, the state of determining described prediction term occurs abnormal, notice alarm module is reported to the police, at this, alarm module 6 can, in the situation that difference all occurs extremely again to User Alarms within a period of time of setting, exist error to carry out the situation of false alarm to avoid occurring because of a secondary data once in a while.
Embodiment shown in above foundation is graphic describes structure of the present invention, feature and action effect in detail; the foregoing is only preferred embodiment of the present invention; but the present invention does not limit practical range with shown in drawing; every change of doing according to conception of the present invention; or be revised as the equivalent embodiment of equivalent variations; when not exceeding yet instructions and illustrating contain spiritual, all should be in protection scope of the present invention.
Claims (12)
1. a business trend forecasting method, is characterized in that, comprising:
The predetermined period of take from business centring system obtains the statistical history business datum sequence of prediction term as unit;
To described statistical history business datum sequence, according to the service period of described prediction term, go periodization to process, obtain the statistical history business datum sequence after processing;
According to the statistical history business datum sequence after processing, the method that adopts time series forecasting to described prediction term the business trend in selected predetermined period predict, obtain tentative prediction data;
Described tentative prediction data are carried out going periodization to process corresponding reduction cycleization processing with described, obtain the predicted data of selected predetermined period.
2. business trend forecasting method according to claim 1, is characterized in that,
Describedly to described statistical history business datum sequence, go periodization to process to comprise:
Described in carrying out described statistical history business datum sequence, employing difference method go periodization to process;
Describedly described tentative prediction data are carried out to reduction cycleization process and to comprise:
From described statistical history business datum sequence, obtain the statistical history business datum of upper first-class order predetermined period, wherein, described first-class order predetermined period is for be positioned at predetermined period of same sequence position according to time sequencing and selected predetermined period in a upper service period, and a described upper service period is the service period at selected predetermined period place;
Calculate described tentative prediction data and upper first-class order predetermined period statistical history business datum and as described predicted data.
3. business trend forecasting method according to claim 2, is characterized in that, the method for going periodization to process described in described employing difference method carries out comprises:
If predetermined period that described service period comprises varying number, the statistical history business datum of choosing adjacent two predetermined period of same sequence position from each little service period is carried out linear interpolation, so that all little service period are supplied as large service period, statistical history business datum sequence after formation is supplied, described in carrying out with the statistical history business datum sequence to after described supplying, go periodization to process, wherein, described large service period is the maximum service period of quantity that comprises predetermined period, little service period is the service period that the quantity that comprises predetermined period is less than large service period.
4. business trend forecasting method according to claim 1, is characterized in that, described method also comprises:
From business centring system, obtain the original historical business datum sequence of prediction term, by described original historical business datum sequence is carried out to the service period that fast fourier transform obtains described prediction term;
The method of obtaining described statistical history business datum sequence from business centring system is:
Described original historical business datum sequence be take to predetermined period as the statistical history business datum that unit statistics occurs in each predetermined period, form described statistical history business datum sequence.
5. according to the business trend forecasting method described in any one in claim 1 to 4, it is characterized in that, described method also comprises:
Obtaining described prediction term after the predicted data of selected predetermined period, judge whether the state of described prediction term occurs extremely;
If so, to User Alarms.
6. business trend forecasting method according to claim 5, is characterized in that, describedly judges whether the state of described prediction term occurs extremely comprising:
The predetermined period of take from described service center system obtains the current statistical history business datum of described prediction term as unit, and described current statistical history business datum is described prediction term actual business datum occurring in selected predetermined period;
Calculate the difference of the selected current statistical history business datum of predetermined period and the predicted data of selected predetermined period;
If described difference surpasses default warning value, determine that the state of described prediction term occurs abnormal.
7. a business trend predicting system, is characterized in that, comprising:
Data acquisition module, obtains the statistical history business datum sequence of prediction term for take predetermined period from business centring system as unit;
Go periodization module, for going periodization to process described statistical history business datum sequence according to the service period of described prediction term, obtain the statistical history business datum after processing;
Prediction module, for according to the statistical history business datum after processing, the method that adopts time series forecasting to described prediction term the business trend in selected predetermined period predict, obtain tentative prediction data;
Data restoring module, goes periodization to process corresponding reduction cycleization processing for described tentative prediction data are carried out with described, obtains the predicted data of selected predetermined period.
8. business trend predicting system according to claim 7, is characterized in that,
The described periodization module of going comprises:
Difference unit, for adopting difference method to go described in described statistical history business datum sequence is carried out according to the service period of described prediction term periodization to process;
Described data restoring module comprises:
Data capture unit, for obtain the statistical history business datum of first-class order predetermined period from described data acquisition module, wherein, described first-class order predetermined period is for be positioned at predetermined period of same sequence position according to time sequencing and selected predetermined period in a upper service period, and a described upper service period is the service period at selected predetermined period place;
Reduction unit, for calculate described tentative prediction data and upper first-class order predetermined period statistical history business datum and as described predicted data.
9. business trend predicting system according to claim 8, is characterized in that, described in go periodization module also to comprise:
Interpolating unit, be used in the situation that predetermined period that described service period comprises varying number, from each little service period, choose the statistical history business datum of adjacent two predetermined period of same sequence position and carry out linear interpolation, so that all little service period are supplied as large service period, statistical history business datum sequence after formation is supplied, statistical history business datum sequence for difference unit after to described supplying goes periodization to process described in carrying out, wherein, described large service period is the maximum service period of quantity that comprises predetermined period, little service period is the service period that the quantity that comprises predetermined period is less than large service period.
10. business trend predicting system according to claim 7, is characterized in that, described data acquisition module comprises:
Raw data acquiring unit, for obtaining original historical business datum sequence from business centring system;
Statistic unit, for described original historical business datum sequence being take to predetermined period as the statistical history business datum that unit statistics occurs in each predetermined period, forms described statistical history business datum sequence;
The described periodization module of going also comprises:
Service period computing unit, carries out fast fourier transform for the described original historical business datum sequence that raw data acquiring unit is provided, to obtain the service period of described prediction term.
11. according to the business trend predicting system described in any one in claim 7 to 10, it is characterized in that, described system also comprises:
Judge module, for obtaining described prediction term after the predicted data of selected predetermined period, judges whether the state of described prediction term occurs extremely;
Alarm module, after occurring extremely for the state in judge module judgement prediction term, to User Alarms.
12. business trend predicting systems according to claim 11, is characterized in that, described judge module comprises:
Current data acquiring unit, for obtain the current statistical history business datum of described prediction term from described data acquisition module, described current statistical history business datum is described prediction term actual business datum occurring in selected predetermined period;
Computing unit, for calculating the difference of the selected current statistical history business datum of predetermined period and the predicted data of selected predetermined period;
Comparing unit, for described difference and default warning value are compared, if described difference surpasses default warning value, determines that the state of described prediction term occurs abnormal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310611664.7A CN103678514B (en) | 2013-11-26 | 2013-11-26 | Business trend prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310611664.7A CN103678514B (en) | 2013-11-26 | 2013-11-26 | Business trend prediction method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103678514A true CN103678514A (en) | 2014-03-26 |
| CN103678514B CN103678514B (en) | 2017-01-18 |
Family
ID=50316059
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310611664.7A Active CN103678514B (en) | 2013-11-26 | 2013-11-26 | Business trend prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103678514B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104618949A (en) * | 2015-02-13 | 2015-05-13 | 浪潮通信信息系统有限公司 | Complaint predicting method and device based on ARMA model |
| CN104636874A (en) * | 2015-02-12 | 2015-05-20 | 北京嘀嘀无限科技发展有限公司 | Method and equipment for detecting business exception |
| CN105426978A (en) * | 2014-09-04 | 2016-03-23 | 中兴通讯股份有限公司 | Service concurrency prediction method and service concurrency prediction system |
| CN105577402A (en) * | 2014-10-11 | 2016-05-11 | 北京通达无限科技有限公司 | Business exception monitoring method and business exception monitoring equipment based on historical data |
| CN107292649A (en) * | 2016-04-01 | 2017-10-24 | 阿里巴巴集团控股有限公司 | The method and apparatus of traffic forecast Data correction |
| CN109727073A (en) * | 2018-12-29 | 2019-05-07 | 携程旅游网络技术（上海）有限公司 | Flowing of access control method, system, electronic equipment and storage medium |
| CN110163417A (en) * | 2019-04-26 | 2019-08-23 | 阿里巴巴集团控股有限公司 | A kind of prediction technique of portfolio, device and equipment |
| CN110858346A (en) * | 2018-08-22 | 2020-03-03 | 阿里巴巴集团控股有限公司 | Data processing method, device and machine readable medium |
| CN111833114A (en) * | 2020-07-27 | 2020-10-27 | 北京思特奇信息技术股份有限公司 | Intelligent prediction method, system, medium and equipment for channel business development target |
| WO2021223515A1 (en) * | 2020-05-06 | 2021-11-11 | 中兴通讯股份有限公司 | Method and apparatus for monitoring data modality change, and device and storage medium |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109101378A (en) * | 2018-07-20 | 2018-12-28 | 郑州云海信息技术有限公司 | A kind of method and system of automatic test storage device history report tool |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101132447A (en) * | 2007-08-13 | 2008-02-27 | 中国移动通信集团安徽有限公司 | Hot line call prediction method for large call centre |
| US20090204553A1 (en) * | 2004-12-17 | 2009-08-13 | Gates Kevin E | Feature Reduction Method for Decision Machines |
| CN101631317A (en) * | 2009-08-19 | 2010-01-20 | 哈尔滨工业大学 | Multistep forecasting method for mobile communication telephone traffic |
| CN102869033A (en) * | 2012-09-21 | 2013-01-09 | 西南交通大学 | ARMA-ARCH model family based prediction method for GPRS (general packet radio service) data services of GSM (global system for mobile communications) communication system |
-
2013
- 2013-11-26 CN CN201310611664.7A patent/CN103678514B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090204553A1 (en) * | 2004-12-17 | 2009-08-13 | Gates Kevin E | Feature Reduction Method for Decision Machines |
| CN101132447A (en) * | 2007-08-13 | 2008-02-27 | 中国移动通信集团安徽有限公司 | Hot line call prediction method for large call centre |
| CN101631317A (en) * | 2009-08-19 | 2010-01-20 | 哈尔滨工业大学 | Multistep forecasting method for mobile communication telephone traffic |
| CN102869033A (en) * | 2012-09-21 | 2013-01-09 | 西南交通大学 | ARMA-ARCH model family based prediction method for GPRS (general packet radio service) data services of GSM (global system for mobile communications) communication system |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105426978A (en) * | 2014-09-04 | 2016-03-23 | 中兴通讯股份有限公司 | Service concurrency prediction method and service concurrency prediction system |
| CN105577402B (en) * | 2014-10-11 | 2019-05-24 | 北京通达无限科技有限公司 | Service exception monitoring method and equipment based on historical data |
| CN105577402A (en) * | 2014-10-11 | 2016-05-11 | 北京通达无限科技有限公司 | Business exception monitoring method and business exception monitoring equipment based on historical data |
| CN104636874A (en) * | 2015-02-12 | 2015-05-20 | 北京嘀嘀无限科技发展有限公司 | Method and equipment for detecting business exception |
| CN104618949B (en) * | 2015-02-13 | 2019-03-05 | 浪潮天元通信信息系统有限公司 | A kind of complaint prediction technique and device based on arma modeling |
| CN104618949A (en) * | 2015-02-13 | 2015-05-13 | 浪潮通信信息系统有限公司 | Complaint predicting method and device based on ARMA model |
| CN107292649A (en) * | 2016-04-01 | 2017-10-24 | 阿里巴巴集团控股有限公司 | The method and apparatus of traffic forecast Data correction |
| CN110858346A (en) * | 2018-08-22 | 2020-03-03 | 阿里巴巴集团控股有限公司 | Data processing method, device and machine readable medium |
| CN110858346B (en) * | 2018-08-22 | 2023-05-02 | 阿里巴巴集团控股有限公司 | Data processing method, apparatus and machine readable medium |
| CN109727073A (en) * | 2018-12-29 | 2019-05-07 | 携程旅游网络技术（上海）有限公司 | Flowing of access control method, system, electronic equipment and storage medium |
| CN109727073B (en) * | 2018-12-29 | 2021-03-05 | 携程旅游网络技术（上海）有限公司 | Access flow control method, system, electronic device and storage medium |
| CN110163417A (en) * | 2019-04-26 | 2019-08-23 | 阿里巴巴集团控股有限公司 | A kind of prediction technique of portfolio, device and equipment |
| CN110163417B (en) * | 2019-04-26 | 2023-09-01 | 创新先进技术有限公司 | Traffic prediction method, device and equipment |
| WO2021223515A1 (en) * | 2020-05-06 | 2021-11-11 | 中兴通讯股份有限公司 | Method and apparatus for monitoring data modality change, and device and storage medium |
| CN111833114A (en) * | 2020-07-27 | 2020-10-27 | 北京思特奇信息技术股份有限公司 | Intelligent prediction method, system, medium and equipment for channel business development target |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103678514B (en) | 2017-01-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103678514A (en) | Business trend prediction method and system | |
| JP6721627B2 (en) | Systems and methods for resource consumption analysis | |
| CN111143102B (en) | Abnormal data detection method and device, storage medium and electronic equipment | |
| Zhou et al. | Reliability-centered predictive maintenance scheduling for a continuously monitored system subject to degradation | |
| CN102111307B (en) | Method and device for monitoring and controlling network risks | |
| CN109639450A (en) | Fault alarming method, computer equipment and storage medium neural network based | |
| CN101442762B (en) | Method and apparatus for analyzing network performance and locating network fault | |
| CN101772742B (en) | An alarm analysis system and a method for providing statistics on alarms from a process control system | |
| US8532836B2 (en) | Demand response load reduction estimation | |
| CN109186888A (en) | A kind of method, apparatus and terminal device of running water pipe water leakage monitoring | |
| CN104349367A (en) | Mobile terminal and predicting reminding method and device of flow consumption of mobile terminal | |
| CN108920336A (en) | A kind of service abnormity prompt method and device based on time series | |
| CA2565528A1 (en) | Method and apparatus for triage of network alarms | |
| CN109992473A (en) | Monitoring method, device, equipment and the storage medium of application system | |
| CN109088775A (en) | abnormality monitoring method, device and server | |
| CN109327334A (en) | A kind of alarm method being directed to equipment performance, device and equipment | |
| CN111130867B (en) | Intelligent household equipment alarm method and device based on Internet of things | |
| CN108881646B (en) | Task tracking method, device, equipment and storage medium of call center | |
| CN110535670B (en) | NFV capacity planning method and electronic equipment | |
| CN112965876A (en) | Monitoring alarm method and device | |
| CN106600144A (en) | Anti-electricity-theft early warning analysis method with single anomaly analysis | |
| CN116010897A (en) | Method and device for detecting data abnormality, electronic equipment and storage medium | |
| CN102735450B (en) | For assessing the system and method for the operation of combustor | |
| CN109597700A (en) | A kind of disk life-span prediction method and relevant apparatus | |
| JP2005136811A (en) | Device and method for diagnosis on base station |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: Wangjiang Road high tech Development Zone Hefei city Anhui province 230088 No. 666 Applicant after: Iflytek Co., Ltd. Address before: Wangjiang Road high tech Development Zone Hefei city Anhui province 230088 No. 666 Applicant before: Anhui USTC iFLYTEK Co., Ltd. |
|
| COR | Change of bibliographic data | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20190527 Address after: 100084 Tsinghua Yuan, Beijing, Haidian District Co-patentee after: Iflytek Co., Ltd. Patentee after: Tsinghua University Address before: 230088 666 Wangjiang West Road, Hefei hi tech Development Zone, Anhui Patentee before: Iflytek Co., Ltd. |