CN106228272A - A kind of method and device obtaining threshold value of warning - Google Patents
A kind of method and device obtaining threshold value of warning Download PDFInfo
- Publication number
- CN106228272A CN106228272A CN201610618224.8A CN201610618224A CN106228272A CN 106228272 A CN106228272 A CN 106228272A CN 201610618224 A CN201610618224 A CN 201610618224A CN 106228272 A CN106228272 A CN 106228272A
- Authority
- CN
- China
- Prior art keywords
- warning
- threshold value
- index
- sample
- threshold
- 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.)
- Pending
Links
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"
-
- 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
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
-
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Data Mining & Analysis (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The method and device obtaining threshold value of warning that the present invention provides, obtain warning index, set up the SVM prediction model corresponding with warning index, and obtain the threshold value of warning corresponding with warning index according to SVM prediction model, the problem obtaining dynamic self-adapting threshold value of warning is converted into forecasting problem relatively newly, solve prior art need manually to arrange the threshold value of warning of threshold value of warning and setting can not the technical problem of self-adaptative adjustment, arranging without artificial and adjust threshold value of warning, intelligence degree is high.
Description
Technical field
The present invention relates to communication technical field, be specifically related to a kind of method and device obtaining threshold value of warning.
Background technology
At present, some management software design patterns has warning module, such as employee's image warning module in office management system,
Goods entry, stock and sales forewarning index module sold in forewarning index module and JXC System in marketing system etc..And these are pre-
Alert module realizes the process of early warning and generally comprises: first by manually arranging threshold value of warning, then judge monitoring warning index or
Whether parameter reaches threshold value of warning, realizes early warning finally according to the result judged.
As can be seen here, the threshold value of warning in the warning module of prior art is mainly arranged by artificial, and uses this
There is problems such as " operations waste time and energy, can-not be automatically generated, cannot self-adaptative adjustment " in the method manually arranging threshold value of warning.Example
Threshold value of warning as arranged is the most scientific and reasonable, and this is artificially to obtain according to subjective experience mainly due to the threshold value of warning arranged
's.Additionally, the artificial threshold value of warning arranged can not self-adaptative adjustment, such as refer to for the monthly quantity in stock early warning in marketing system
Mark, each monthly meeting of management personnel periodically arranges corresponding monthly quantity in stock threshold value of warning.But outside if certain is monthly
(such as it is in great demand or period in red-letter day, carried out advertising campaign etc.) when boundary's environmental factors changes, then may need
Reset quantity in stock threshold value of warning.As can be seen here, manually arrange the wasting time and energy of threshold value of warning, can-not be automatically generated, cannot
Self-adaptative adjustment, therefore a kind of method and device that can automatically obtain threshold value of warning of offer is provided badly.
Summary of the invention
The invention provides a kind of method and device obtaining threshold value of warning, need manually to arrange pre-solving prior art
The threshold value of warning of alert threshold value and setting can not the technical problem of self-adaptative adjustment.
According to an aspect of the present invention, it is provided that a kind of method obtaining threshold value of warning, including:
Obtain warning index;
Set up the SVM prediction model corresponding with warning index;
The threshold value of warning corresponding with warning index is obtained according to SVM prediction model.
Further, set up the SVM prediction model corresponding with warning index to include:
Obtain the training sample corresponding with warning index;
According to training sample Training Support Vector Machines model, it is thus achieved that SVM prediction model.
Further, obtain the training sample corresponding with warning index to include:
Preset the influence factor's entry associated with warning index;
Obtain the early warning sample threshold corresponding with warning index, and the property value corresponding with early warning sample threshold, wherein, attribute
Value is the result corresponding with early warning sample threshold gathered according to influence factor's entry;
Using early warning sample threshold and the property value corresponding with early warning sample threshold as the training sample corresponding with warning index.
Further, after obtaining the training sample corresponding with warning index, according to training sample Training Support Vector Machines
Also include before model:
Training sample is normalized.
Further, influence factor's entry includes:
Time, place, weather, region level of development, red-letter day, advertising campaign, personnel, scale, historical data, colleague are with reference to index
One or more in factor entry.
Further, obtain the threshold value of warning corresponding with warning index according to SVM prediction model to include:
Gather the influence value corresponding with influence factor's entry of warning index association;
Influence value is inputted SVM prediction model, it is thus achieved that the threshold value of warning corresponding with warning index.
According to a further aspect in the invention, it is provided that a kind of device obtaining threshold value of warning, including:
Warning index acquisition device, is used for obtaining warning index;
Forecast model sets up device, for setting up the SVM prediction model corresponding with warning index;
Threshold value of warning acquisition device, for obtaining the threshold value of warning corresponding with warning index according to SVM prediction model.
Further, it was predicted that model is set up device and included:
Training sample acquisition device, for obtaining the training sample corresponding with warning index;
Training devices, for according to training sample Training Support Vector Machines model, it is thus achieved that SVM prediction model.
Further, training sample acquisition device includes:
Influence factor's entry presets device, for presetting the influence factor's entry associated with warning index;
Property value acquisition device, for obtaining the early warning sample threshold corresponding with warning index, and with early warning sample threshold pair
The property value answered, wherein, property value is the result corresponding with early warning sample threshold gathered according to influence factor's entry;
Determine device, for using early warning sample threshold and the property value corresponding with early warning sample threshold as with warning index pair
The training sample answered.
The method have the advantages that
The method and device obtaining threshold value of warning that the present invention provides, obtains warning index, sets up prop up corresponding with warning index
Hold vector machine forecast model, and obtain the threshold value of warning corresponding with warning index according to SVM prediction model, newly
The problem obtaining dynamic self-adapting threshold value of warning is converted into forecasting problem by grain husk ground, and solving prior art needs manually to arrange pre-
The threshold value of warning of alert threshold value and setting can not the technical problem of self-adaptative adjustment, it is not necessary to artificial arrange and adjusts threshold value of warning, intelligence
Degree can be changed high.
In addition to objects, features and advantages described above, the present invention also has other objects, features and advantages.
Below with reference to figure, the present invention is further detailed explanation.
Accompanying drawing explanation
The accompanying drawing of the part building the application is used for providing a further understanding of the present invention, and the present invention's is schematic real
Execute example and illustrate for explaining the present invention, not building inappropriate limitation of the present invention.In the accompanying drawings:
Fig. 1 is the method flow diagram obtaining threshold value of warning of the preferred embodiment of the present invention;
Fig. 2 be the preferred embodiment of the present invention for the method flow diagram obtaining threshold value of warning simplifying embodiment;
Fig. 3 is the apparatus structure block diagram obtaining threshold value of warning of the preferred embodiment of the present invention.
Description of reference numerals:
10, warning index acquisition device;20, forecast model sets up device;30, threshold value of warning acquisition device.
Detailed description of the invention
Below in conjunction with accompanying drawing, embodiments of the invention are described in detail, but the present invention can be defined by the claims
Implement with the multitude of different ways covered.
With reference to Fig. 1, the preferred embodiments of the present invention provide a kind of method obtaining threshold value of warning, including:
Step S101, obtains warning index;
Step S102, sets up the SVM prediction model corresponding with warning index;
Step S103, obtains the threshold value of warning corresponding with warning index according to SVM prediction model.
The method obtaining threshold value of warning that the present invention provides, by obtaining warning index, sets up corresponding with warning index
SVM prediction model, and obtain the threshold value of warning corresponding with warning index according to SVM prediction model, relatively
Novelly the problem obtaining dynamic self-adapting threshold value of warning being converted into forecasting problem, solving prior art needs artificial setting
The threshold value of warning of threshold value of warning and setting can not the technical problem of self-adaptative adjustment, it is not necessary to artificial arrange and adjusts threshold value of warning,
Intelligence degree is high.
Why the present embodiment sets up the SVM prediction model corresponding with warning index, mainly can be according to it
Predict the threshold value of warning corresponding with warning index.In prior art, there is the threshold value of warning in the system of warning function module past
Past is all artificial the most default, therefore the threshold value of warning arranged may be the most scientific and reasonable;And it is of the prior art pre-
Alert threshold value once sets, substantially stationary, namely when system needs to adjust threshold value of warning, can only artificially manually change early warning threshold
Value.Certainly, using this manual type to arrange the dynamic self-adapting of threshold value of warning poor, intelligence degree is low, the present embodiment
By setting up SVM prediction model, threshold value of warning can be automatically obtained, the most scientific and reasonable, and the present embodiment prediction
Threshold value of warning change according to the change of the initial conditions of SVM prediction model, can carry out according to extraneous initial conditions
Self-adaptative adjustment, intelligence degree is high.
It should be noted that the present embodiment is not limited to use the acquisition of SVM prediction model corresponding with warning index
Threshold value of warning, such as, can also use method of least square or other fitting processs to obtain the threshold value of warning corresponding with warning index.
Alternatively, set up the SVM prediction model corresponding with warning index to include:
Obtain the training sample corresponding with warning index;
According to training sample Training Support Vector Machines model, it is thus achieved that SVM prediction model.
In order to obtain the most scientific and reasonable threshold value of warning, the present embodiment obtains early warning according to SVM prediction model
During threshold value, it is necessary first to gather training sample and SVM prediction model is trained.It should be noted that in order to improve
The precision of prediction of support machine forecast model, the quantity of the training sample that the present embodiment gathers should be big as far as possible, such that it is able to will
Parameter adjustment in SVM prediction model is to relative optimal value.
Alternatively, obtain the training sample corresponding with warning index to include:
Preset the influence factor's entry associated with warning index;
Obtain the early warning sample threshold corresponding with warning index, and the property value corresponding with early warning sample threshold, wherein, attribute
Value is the result corresponding with early warning sample threshold gathered according to influence factor's entry;
Using early warning sample threshold and the property value corresponding with early warning sample threshold as the training sample corresponding with warning index.
In prior art, the threshold value of warning corresponding with warning index is the most rule of thumb set, often considers some shadows
Ring the influence factor of warning index.Such as, for the quantity in stock warning index in marketing system, management personnel typically can be according to shadow
The factor (such as time, red-letter day, advertising campaign factor etc.) ringing quantity in stock warning index arranges threshold value of warning;And for selling system
Sales volume warning index in system, management personnel then may be according to factor (the such as region development affecting sales volume warning index
Level, sale population support, scale factor etc.) threshold value of warning is set, specifically it is shown in Table 1, it should be noted that table 1 is enumerated
The warning index that the influence factor's entry associated with warning index is primarily directed in realm of sale or marketing system is arranged, its
The influence factor's entry associated with warning index in his field or system is also by User Defined.
Wherein, in the present embodiment indication with go together with reference to the property value that index entry is corresponding specifically refer to scale identical or
The threshold value of warning that similar colleague is arranged for identical warning index.The most known beauty parlor A was for the sales volume index of certain month
The sales volume threshold value of warning arranged is 100,000, and now needs the beauty parlor B similar to A beauty parlor scale arranges of that month sales volume
During threshold value of warning, then it is referred to the of that month sales volume threshold value of warning arranged with B scale same or analogous beauty parlor of beauty parlor A
(100,000).Additionally, the property value corresponding with historical data factor entry of the present embodiment indication refers to corresponding with warning index
History threshold value of warning, the history threshold value of warning such as arranged for quantity in stock warning index is 50, then these data are and go through
The property value that history data factors entry is corresponding, in actual implementation process, when system acquisition is to multiple corresponding with warning index
History threshold value of warning time, then can take the meansigma methods of history threshold value of warning as the attribute corresponding with historical data factor entry
Value.
As can be seen here, different warning indexs often associates different influence factors, therefore the present embodiment is obtaining and early warning
During training sample corresponding to index, first preset the influence factor's entry associated with warning index, then acquisition and warning index
Corresponding early warning sample threshold, and the property value corresponding with early warning sample threshold, wherein, property value is according to influence factor's bar
The result corresponding with early warning sample threshold that mesh gathers, the most again by early warning sample threshold and corresponding with early warning sample threshold
Property value is as the training sample corresponding with warning index.
Table 1
The present embodiment is by presetting influence factor's entry of associating with warning index, and pre-corresponding with warning index of acquisition
Alert sample threshold, and the property value corresponding with early warning sample threshold, wherein, property value is to gather according to influence factor's entry
The result corresponding with early warning sample threshold, and using early warning sample threshold and the property value corresponding with early warning sample threshold as with
The training sample that warning index is corresponding, has taken into full account the influence factor associated with warning index affecting early warning sample threshold,
Property value from each dimension pair is corresponding with early warning sample threshold is acquired, for improving SVM prediction model
Accuracy and precision of prediction provide important Data Source basis.
Alternatively, after obtaining the training sample corresponding with warning index, according to training sample Training Support Vector Machines mould
Also include before type:
Training sample is normalized.
The property value corresponding with influence factor's entry gathered due to this enforcement is probably result qualitatively, it is also possible to fixed
The result of amount, therefore the present embodiment is after obtaining the training sample corresponding with warning index, according to training sample training support to
Before amount machine model, training sample is normalized.And it is follow-up according to training sample Training Support Vector Machines
Model, the present embodiment is typically chosen and is carried out at quantification according to default rule by the property value corresponding with influence factor's entry
Reason.Such as, the present embodiment in order to the property value corresponding with region level of development factor entry (flourishing, relative flourishing, do not send out
Reach, the most undeveloped) when being normalized, the property value corresponding with influence factor's entry gathered can be respectively provided with
Between scope 0-100, referring in particular to table 2.Wherein, the normalization rule that the present embodiment is preset is by User Defined, and presets
Normalization after attribute-value ranges also determined according to demand by user.
Table 2
Alternatively, influence factor's entry includes:
Time, place, weather, region level of development, red-letter day, advertising campaign, personnel, scale, historical data, colleague are with reference to index
One or more in factor entry.
Influence factor's entry in the present embodiment is not limited to time, place, weather, region level of development, red-letter day, sales promotion work
Dynamic, personnel, scale, historical data, colleague with reference to one or more in index factor entry, such as can also include public praise,
Customer resources configuration etc..
Alternatively, obtain the threshold value of warning corresponding with warning index according to SVM prediction model to include:
Gather the influence value corresponding with influence factor's entry of warning index association;
Influence value is inputted SVM prediction model, it is thus achieved that the threshold value of warning corresponding with warning index.
Specifically, when needing to obtain the threshold value of warning corresponding with warning index, first the present embodiment gathers and refers to early warning
The influence value corresponding to influence factor's entry of mark association, then inputs SVM prediction model by influence value, it is thus achieved that with in advance
The threshold value of warning that alert index is corresponding.It is achieved thereby that according to the influence value corresponding with influence factor's entry that warning index associates certainly
Dynamic obtain dynamic threshold value of warning, the most scientific and reasonable but also intelligence degree high.
Simplify embodiment below for one the present invention obtains the method for threshold value of warning to illustrate further.
With reference to Fig. 2, the method obtaining threshold value of warning in the present embodiment includes:
Step S201, obtains warning index.
Specifically, the warning index that the present embodiment obtains can be one, it is also possible to for multiple, with specific reference to user's request
Obtain.And it is identical to obtain the method for corresponding threshold value of warning for each warning index, present embodiment assumes that and obtains
The warning index taken is sales volume warning index.
Step S202, presets the influence factor's entry associated with warning index.
Specifically, it is assumed that the influence factor's entry associated with sales volume warning index pre-set be historical data,
Territory development degree, red-letter day, promoting factor, colleague, with reference to index entry, are specifically shown in Table 1.
Step S203, obtains the early warning sample threshold corresponding with warning index, and the genus corresponding with early warning sample threshold
Property value, wherein, property value is the result corresponding with early warning sample threshold gathered according to influence factor's entry, and by early warning sample
Threshold value and the property value corresponding with early warning sample threshold are as the training sample corresponding with warning index.
Specifically, it is assumed that the present embodiment gets three corresponding with threshold value of warning group early warning sample threshold, and with each
Group property value corresponding to early warning sample threshold, wherein, property value be gather according to influence factor's entry with early warning sample threshold
Corresponding result, is specifically shown in Table 3.
Table 3
Step S204, is normalized training sample.
Specifically, the property value corresponding with influence factor's entry gathered due to this enforcement is probably result qualitatively, also
It is probably quantitative result, therefore the present embodiment is after obtaining the training sample corresponding with warning index, instruct according to training sample
Before practicing supporting vector machine model, training sample is normalized.Follow-up according to training sample training support
Vector machine model, the present embodiment is typically chosen and is carried out quantitatively according to default rule by the property value corresponding with influence factor's entry
Change process, specifically, the present embodiment by gather the property value corresponding with influence factor's entry be arranged at scope 0-100 it
Between, referring in particular to table 2, therefore after training sample is normalized by the present embodiment according to table 2, it is possible to obtain the training after normalization
Sample, is specifically shown in Table 4.
Table 4
It should be noted that for the precision of prediction improving support machine forecast model, the number of the training sample that the present embodiment gathers
Amount should be big as far as possible, such that it is able to by the parameter adjustment in SVM prediction model to relative optimal value.
Step S205, according to training sample Training Support Vector Machines model, it is thus achieved that SVM prediction model.
Specifically, the present embodiment by using early warning sample threshold as the output of supporting vector machine model, will be with early warning sample
What this threshold value was corresponding associates the input as supporting vector machine model of property value corresponding to influence factor's entry, to support vector machine
Model is trained, thus obtains SVM prediction model.
Step S206, gathers the influence value corresponding with influence factor's entry of warning index association.
Specifically, first the present embodiment gathers the influence value corresponding with influence factor's entry of warning index association, then
Influence value is inputted SVM prediction model, it is thus achieved that the threshold value of warning corresponding with warning index.It is achieved thereby that according to
The influence value corresponding to influence factor's entry of warning index association obtains dynamic threshold value of warning automatically, the most scientific and reasonable but also
Intelligence degree is high.
Assume the influence factor bar corresponding with warning index (sales volume threshold value of warning) to be predicted that the present embodiment obtains
The influence value that mesh is corresponding is respectively historical data factor entry: 2,000,000;Colleague is with reference to index factor entry: 1,800,000;Region is sent out
Exhibition degree factor entry: flourishing;Red-letter day factor entry: festivals or holidays;Promoting factor entry: have.
Step S207, inputs SVM prediction model by influence value, it is thus achieved that the threshold value of warning corresponding with warning index.
Specifically, first the influence value of acquisition is normalized by the present embodiment, and using the influence value after normalization as
The input of the SVM prediction model trained, thus automatically obtain the threshold value of warning corresponding with warning index.By this
Inventive embodiment understands, and when the initial conditions (influence value) of SVM prediction model changes, it exports (early warning threshold
Value) follow change.That is, the present embodiment can not only obtain corresponding with warning index according to SVM prediction model automatically
Threshold value of warning, and the threshold value of warning obtained can carry out dynamic self-adapting according to extraneous environmental factors.
The method automatically obtaining threshold value of warning that the present invention provides, by obtaining warning index, sets up and warning index pair
The SVM prediction model answered, and obtain the early warning threshold corresponding with warning index according to SVM prediction model
Value, is converted into forecasting problem by the problem obtaining dynamic self-adapting threshold value of warning relatively newly, and solving prior art needs people
Work arranges the threshold value of warning of threshold value of warning and setting can not the technical problem of self-adaptative adjustment, it is not necessary to artificial arrange and adjusts early warning
Threshold value, intelligence degree is high.
With reference to Fig. 3, the device obtaining threshold value of warning that the preferred embodiments of the present invention provide, including:
Warning index acquisition device 10, is used for obtaining warning index;
Forecast model sets up device 20, for setting up the SVM prediction model corresponding with warning index;
Threshold value of warning acquisition device 30, for obtaining the early warning threshold corresponding with warning index according to SVM prediction model
Value.
Alternatively, it was predicted that model is set up device 20 and included:
Training sample acquisition device, for obtaining the training sample corresponding with warning index;
Training devices, for according to training sample Training Support Vector Machines model, it is thus achieved that SVM prediction model.
Alternatively, training sample acquisition device includes:
Influence factor's entry presets device, for presetting the influence factor's entry associated with warning index;
Property value acquisition device, for obtaining the early warning sample threshold corresponding with warning index, and with early warning sample threshold pair
The property value answered, wherein, property value is the result corresponding with early warning sample threshold gathered according to influence factor's entry;
Determine device, for using early warning sample threshold and the property value corresponding with early warning sample threshold as with warning index pair
The training sample answered.
The device obtaining threshold value of warning that the present invention provides, by obtaining warning index, sets up corresponding with warning index
SVM prediction model, and obtain the threshold value of warning corresponding with warning index according to SVM prediction model, relatively
Novelly the problem obtaining dynamic self-adapting threshold value of warning being converted into forecasting problem, solving prior art needs artificial setting
The threshold value of warning of threshold value of warning and setting can not the technical problem of self-adaptative adjustment, it is not necessary to artificial arrange and adjusts threshold value of warning,
Intelligence degree is high.
Specific works process and the operation principle of the device of the present embodiment acquisition threshold value of warning can refer to obtaining of the present embodiment
Take work process and the operation principle of the method for threshold value of warning.
These are only the preferred embodiments of the present invention, be not limited to the present invention, for those skilled in the art
For Yuan, the present invention can have various modifications and variations.All within the spirit and principles in the present invention, any amendment of being made,
Equivalent, improvement etc., should be included within the scope of the present invention.
Claims (9)
1. the method obtaining threshold value of warning, it is characterised in that including:
Obtain warning index;
Set up the SVM prediction model corresponding with described warning index;
The threshold value of warning corresponding with described warning index is obtained according to described SVM prediction model.
The method of acquisition threshold value of warning the most according to claim 1, it is characterised in that set up corresponding with described warning index
SVM prediction model include:
Obtain the training sample corresponding with described warning index;
According to described training sample Training Support Vector Machines model, it is thus achieved that SVM prediction model.
The method of acquisition threshold value of warning the most according to claim 2, it is characterised in that obtain corresponding with described warning index
Training sample include:
Preset the influence factor's entry associated with described warning index;
Obtain the early warning sample threshold corresponding with described warning index, and the property value corresponding with described early warning sample threshold,
Wherein, the result corresponding with described early warning sample threshold that according to described property value, described influence factor's entry gathers;
Using described early warning sample threshold and the property value corresponding with described early warning sample threshold as with described warning index pair
The training sample answered.
The method of acquisition threshold value of warning the most according to claim 3, it is characterised in that obtain corresponding with described warning index
Training sample after, also include according to before described training sample Training Support Vector Machines model:
Described training sample is normalized.
The method of acquisition threshold value of warning the most according to claim 4, it is characterised in that described influence factor's entry includes:
Time, place, weather, region level of development, red-letter day, advertising campaign, personnel, scale, historical data, colleague are with reference to index
One or more in factor entry.
The method of acquisition threshold value of warning the most according to claim 5, it is characterised in that according to described SVM prediction
Model obtains the threshold value of warning corresponding with described warning index and includes:
Gather the influence value corresponding with influence factor's entry of described warning index association;
Described influence value is inputted described SVM prediction model, it is thus achieved that the threshold value of warning corresponding with described warning index.
7. the device obtaining threshold value of warning, it is characterised in that
Warning index acquisition device, is used for obtaining warning index;
Forecast model sets up device, for setting up the SVM prediction model corresponding with described warning index;
Threshold value of warning acquisition device, corresponding with described warning index pre-for obtaining according to described SVM prediction model
Alert threshold value.
The device of acquisition threshold value of warning the most according to claim 7, it is characterised in that described forecast model sets up device bag
Include:
Training sample acquisition device, for obtaining the training sample corresponding with described warning index;
Training devices, for according to described training sample Training Support Vector Machines model, it is thus achieved that SVM prediction model.
The device of acquisition threshold value of warning the most according to claim 8, it is characterised in that described training sample acquisition device bag
Include:
Influence factor's entry presets device, for presetting the influence factor's entry associated with described warning index;
Property value acquisition device, for obtaining the early warning sample threshold corresponding with described warning index, and with described early warning sample
The property value that this threshold value is corresponding, wherein, according to described property value described influence factor's entry gather with described early warning sample
The result that threshold value is corresponding;
Determine device, for using described early warning sample threshold and the property value corresponding with described early warning sample threshold as with institute
State the training sample that warning index is corresponding.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610618224.8A CN106228272A (en) | 2016-08-01 | 2016-08-01 | A kind of method and device obtaining threshold value of warning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610618224.8A CN106228272A (en) | 2016-08-01 | 2016-08-01 | A kind of method and device obtaining threshold value of warning |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106228272A true CN106228272A (en) | 2016-12-14 |
Family
ID=57535688
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610618224.8A Pending CN106228272A (en) | 2016-08-01 | 2016-08-01 | A kind of method and device obtaining threshold value of warning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106228272A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108490126A (en) * | 2018-03-20 | 2018-09-04 | 安徽航途智能科技有限公司 | A kind of bus inflammable gas monitoring system and automatic monitoring method |
CN109599178A (en) * | 2018-11-30 | 2019-04-09 | 苏州麦迪斯顿医疗科技股份有限公司 | Threshold determination model determines method, apparatus, medical treatment detection device and storage medium |
CN111695847A (en) * | 2019-05-17 | 2020-09-22 | 上海寻梦信息技术有限公司 | Number section management method, system, equipment and storage medium for logistics electronic bill |
CN117726044A (en) * | 2024-02-05 | 2024-03-19 | 广东迈科医学科技股份有限公司 | Blood inventory dynamic early warning method and system |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101916998A (en) * | 2010-07-12 | 2010-12-15 | 东北电力科学研究院有限公司 | Support vector machine-based wind electric powder prediction device and method |
CN101994001A (en) * | 2010-09-14 | 2011-03-30 | 上海海事大学 | Support vector machine algorithm based method for predicting vibration aging effect |
EP2688015A1 (en) * | 2012-07-20 | 2014-01-22 | Tata Consultancy Services Limited | Method and system for adaptive forecast of energy resources |
-
2016
- 2016-08-01 CN CN201610618224.8A patent/CN106228272A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101916998A (en) * | 2010-07-12 | 2010-12-15 | 东北电力科学研究院有限公司 | Support vector machine-based wind electric powder prediction device and method |
CN101994001A (en) * | 2010-09-14 | 2011-03-30 | 上海海事大学 | Support vector machine algorithm based method for predicting vibration aging effect |
EP2688015A1 (en) * | 2012-07-20 | 2014-01-22 | Tata Consultancy Services Limited | Method and system for adaptive forecast of energy resources |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108490126A (en) * | 2018-03-20 | 2018-09-04 | 安徽航途智能科技有限公司 | A kind of bus inflammable gas monitoring system and automatic monitoring method |
CN108490126B (en) * | 2018-03-20 | 2021-06-15 | 安徽航途智能科技有限公司 | Bus inflammable gas monitoring system and automatic monitoring method |
CN109599178A (en) * | 2018-11-30 | 2019-04-09 | 苏州麦迪斯顿医疗科技股份有限公司 | Threshold determination model determines method, apparatus, medical treatment detection device and storage medium |
CN111695847A (en) * | 2019-05-17 | 2020-09-22 | 上海寻梦信息技术有限公司 | Number section management method, system, equipment and storage medium for logistics electronic bill |
CN117726044A (en) * | 2024-02-05 | 2024-03-19 | 广东迈科医学科技股份有限公司 | Blood inventory dynamic early warning method and system |
CN117726044B (en) * | 2024-02-05 | 2024-05-14 | 广东迈科医学科技股份有限公司 | Blood inventory dynamic early warning method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106228272A (en) | A kind of method and device obtaining threshold value of warning | |
CN109615122B (en) | Distribution range generation method and device, electronic equipment and storage medium | |
CN109636013B (en) | Distribution range generation method and device, electronic equipment and storage medium | |
CN106169152A (en) | The automatic accounting method of a kind of value-added tax and system | |
CN109508839A (en) | Order allocation method and device | |
CN102446171A (en) | Method and apparatus for evaluating quality score of promotion key word based on weighted average click-through rate | |
CN105260954A (en) | Construction cost list service system based on Internet-of-things technology | |
CN106296247A (en) | The online sort method of network information resource and device | |
AU2023204174A1 (en) | Media control method and system | |
WO2016197857A1 (en) | Position information providing method and device | |
CN112785089A (en) | Agent service configuration method and device, electronic equipment and storage medium | |
CN110443425A (en) | Intelligent railway station electric energy energy consumption prediction technique based on Prophet | |
CN102075896B (en) | Price previewing method and system | |
CN105976170A (en) | Method and device for automatically generating work plans | |
US20140180794A1 (en) | Methods and apparatus for generating an online marketing campaign | |
CN110009287A (en) | Premium determines method, apparatus, equipment and storage medium | |
CN103426097A (en) | Method suitable for keyword release bid of search engine | |
CN111478779A (en) | Method and device for determining unlimited set meal and computing equipment | |
CN107423905A (en) | A kind of energy performance appraisal method and system | |
CN107944645A (en) | A kind of feasibility analysis method of tobacco distribution line optimization | |
CN114428913A (en) | Data management method, device, equipment and storage medium | |
CN110400160B (en) | Method and device for identifying competitive product user, electronic equipment and storage medium | |
CN112380436A (en) | Intelligent recommendation system and method for enterprise intelligent energy solution | |
CN111966903A (en) | Application software function recommendation method and device | |
CN105591980A (en) | Bandwidth consumption management method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161214 |