CN108596370A - A method of the metric analysis prediction registration success rate based on Reference Environment - Google Patents
A method of the metric analysis prediction registration success rate based on Reference Environment Download PDFInfo
- Publication number
- CN108596370A CN108596370A CN201810303689.3A CN201810303689A CN108596370A CN 108596370 A CN108596370 A CN 108596370A CN 201810303689 A CN201810303689 A CN 201810303689A CN 108596370 A CN108596370 A CN 108596370A
- Authority
- CN
- China
- Prior art keywords
- humidity
- temperature
- orientation
- reference environment
- height
- 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"
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 method for the metric analysis prediction registration success rate based on Reference Environment that the invention discloses a kind of.It includes training stage and evaluation and test stage, and the training stage refers to:By the conditions for acquiring Reference Environment, the geographical location of consulting activity spot, the Feature Selection Model of space size, orientation, temperature, humidity, height, air quality are trained using depth learning technology, feature is extracted according to Feature Selection Model, final training obtains prediction model;The evaluation and test stage refers to:By the measurement to Reference Environment, the geographical location of extraction consulting activity spot, space size, orientation, temperature, humidity, height, the feature of air quality carry out prediction of result using prediction model.The beneficial effects of the invention are as follows:The mode of measurement is simple, and data are also easy to obtain, and can form automatic Prediction, realize the automatic Prediction to the successful possibility of registering.
Description
Technical field
The present invention relates to machine learning correlative technology fields, refer in particular to a kind of metric analysis prediction based on Reference Environment
The method for success rate of registering.
Background technology
Increase as social rhythm accelerates competitive pressure, adult training and education with record of formal schooling training have been universally accepted.Training
Mechanism receives visiting registration consulting of largely visiting daily, thus success is registered after the how effective fast prediction consulting of research
There is positive meaning, can not only help to carry out consulting activity arrangement decision improvement effect from objective environment factor angle, and
And training organization can be instructed to make rational planning for Ground arrangement, it is influenced with the environment with affinity, wins consulting registration people to training
Instruct the degree of recognition of mechanism.At present to the prediction of consulting registration success rate, tested and assessed with being based on conversation content test and appraisal and psychology index
Based on two major classes, process is partial to subjective factor, and data are not easy automatic collection.
Invention content
The present invention is in order to overcome the above deficiencies in the prior art, it is simple and can be automatic to provide a kind of measurement
The method of the metric analysis prediction registration success rate based on Reference Environment of prediction.
To achieve the goals above, the present invention uses following technical scheme:
A method of the metric analysis prediction registration success rate based on Reference Environment, including training stage and evaluation and test rank
Section, the training stage refer to:By acquiring the conditions of Reference Environment, consulting is trained using deep learning technology
The geographical location of movable spot, the Feature Selection Model of space size, orientation, temperature, humidity, height, air quality, according to
Feature Selection Model extracts feature, and final training obtains prediction model;The evaluation and test stage refers to:By to Reference Environment
Measurement, the geographical location of extraction consulting activity spot, the spy of space size, orientation, temperature, humidity, height, air quality
Sign carries out prediction of result using prediction model.
The present invention proposes a kind of method predicted registration success rate, by the factor item for acquiring Reference Environment
Part, using deep learning technology extract geographical location in relation to consulting activity spot, space size, orientation, temperature, humidity,
Highly, the feature of air quality carries out model training and marking, final prediction probability is obtained by model.The present invention is based on
Be objective environmental factor, and measure mode it is simple, data also be easy obtain, automatic Prediction can be formed, in conjunction with machine
Study, extracts appropriate environmental characteristic, using trained data model, carries out prediction probability, realizes and successfully may be used to registering
The automatic Prediction of energy property.
Preferably, the training stage, steps are as follows:
(1) the metric data collection for establishing Reference Environment, by the site surrounding of passing registration consulting of nearest two naturology years
Situation is recorded and the result of corresponding registration success or not is associated;Geographical location, sky to consulting activity spot
Between size, orientation, temperature, humidity, height, the feature of air quality manually given a mark;Set the ground of consulting activity spot
Reason position, space size, orientation, temperature, humidity, height, air quality are divided into five grades, wherein five grades are right respectively
The numerical value answered is 0,1,2,3,4;
(2) use step (1) in metric data collection, it includes the geographical location of consulting activity spot, space size,
The artificial marking in orientation, temperature, humidity, height, air quality utilizes the ground of deep learning model training consulting activity spot
Manage position, space size, orientation, temperature, humidity, height, air quality characteristic model;
(3) using the obtained characteristic model of training to registration result extract geographical location in relation to consulting activity spot,
The feature score of space size, orientation, temperature, humidity, height, air quality, and final according to linear regression algorithm training
Prediction model.
Preferably, in step (1), the metric data of Reference Environment is handled according to standard value, wherein geographical
Position longitude and latitude, space size use square metre, orientation gate import direction, temperature use degree Celsius are real in humidity air
The percentage of the ratio between saturation vapour pressure under border vapour pressure and at that time temperature indicates and round numbers, and rice, air quality is highly used to use
AQI, registration successful result indicate success with 1, failure are indicated with 0.
Preferably, in step (2), the deep learning model includes deep neural network, convolutional neural networks
And Recognition with Recurrent Neural Network;Deep learning is the branch of machine learning, is that one kind attempting use comprising labyrinth or by multiple non-
Multiple process layers that linear transformation is constituted carry out data the algorithm of higher level of abstraction, and the feature of extraction includes consulting activity spot
Geographical location, space size, orientation, temperature, humidity, height, air quality, these are characterized in that deep learning algorithm is learned automatically
Acquistion is arrived.
Preferably, in step (3), the geographical location of setting consulting activity spot, space size, orientation, temperature,
Humidity, height, air quality are divided into five grades, wherein the corresponding numerical value of five grades is 0,1,2,3,4, it is linear to return
Return algorithmic formula as follows:Y=AX+b, the formula are vector forms, wherein Y is final score, and A and b are prediction model parameters, A
It is vector, b is scalar, and X is the feature vector of extraction.
Preferably, the evaluation and test stage etch is as follows:
(a) metric data of Reference Environment, the i.e. geographical location of consulting activity spot, space size, side are recorded every time
Position, temperature, humidity, height, air quality handle the metric data of Reference Environment according to standard value;
(b) according to the obtained deep learning model of training, to the geographical location of metric data extraction consulting activity spot,
Space size, orientation, temperature, humidity, height, the feature of air quality;
(c) prediction model obtained using training, and probabilistic forecasting is carried out to registration result according to the feature of extraction.
The beneficial effects of the invention are as follows:The mode of measurement is simple, and data are also easy to obtain, and can form automatic Prediction, in conjunction with
Appropriate environmental characteristic is extracted in machine learning, using trained data model, carries out prediction probability, realize to registration at
The automatic Prediction of work(possibility.
Specific implementation mode
The present invention will be further described With reference to embodiment.
A method of the metric analysis prediction registration success rate based on Reference Environment, including training stage and evaluation and test rank
Section, the training stage refer to:By acquiring the conditions of Reference Environment, consulting is trained using deep learning technology
The geographical location of movable spot, the Feature Selection Model of space size, orientation, temperature, humidity, height, air quality, according to
Feature Selection Model extracts feature, and final training obtains prediction model;The evaluation and test stage refers to:By to Reference Environment
Measurement, the geographical location of extraction consulting activity spot, the spy of space size, orientation, temperature, humidity, height, air quality
Sign carries out prediction of result using prediction model.
Wherein:Training stage, steps are as follows:
(1) the metric data collection for establishing Reference Environment, by the site surrounding of passing registration consulting of nearest two naturology years
Situation is recorded and the result of corresponding registration success or not is associated;Geographical location, sky to consulting activity spot
Between size, orientation, temperature, humidity, height, the feature of air quality manually given a mark;Set the ground of consulting activity spot
Reason position, space size, orientation, temperature, humidity, height, air quality are divided into five grades, wherein five grades are right respectively
The numerical value answered is 0,1,2,3,4;The metric data of Reference Environment is handled according to standard value, wherein geographical location longitude and latitude
Degree, space size use square metre, orientation gate import (all directions) direction, temperature use degree Celsius are real in humidity air
The percentage of the ratio between saturation vapour pressure under border vapour pressure and at that time temperature indicates and round numbers, and rice, air quality is highly used to use
AQI, registration successful result indicate success with 1, failure are indicated with 0.Wherein:Success rate of registering is after registration successful result is accumulative
Divided by registration total number of persons.
(2) use step (1) in metric data collection, it includes the geographical location of consulting activity spot, space size,
The artificial marking in orientation, temperature, humidity, height, air quality utilizes the ground of deep learning model training consulting activity spot
Manage position, space size, orientation, temperature, humidity, height, air quality characteristic model;Deep learning model includes depth
Neural network, convolutional neural networks and Recognition with Recurrent Neural Network;Deep learning is the branch of machine learning, is that one kind attempts using packet
The multiple process layers constituted containing labyrinth or by multiple nonlinear transformation carry out data the algorithm of higher level of abstraction, the spy of extraction
Sign includes the geographical location of consulting activity spot, space size, orientation, temperature, humidity, height, air quality, these features
It is that deep learning algorithm is automatically learned.
(3) using the obtained characteristic model of training to registration result extract geographical location in relation to consulting activity spot,
The feature score of space size, orientation, temperature, humidity, height, air quality, and final according to linear regression algorithm training
Prediction model;The geographical location of setting consulting activity spot, space size, orientation, temperature, humidity, height, air quality are equal
It is divided into five grades, wherein the corresponding numerical value of five grades is 0,1,2,3,4, linear regression algorithm formula is as follows:Y=AX
+ b, the formula are vector forms, wherein Y is final score, and A and b are prediction model parameters, and A is vector, and b is scalar, and X is to carry
The feature vector taken.
It is as follows to evaluate and test stage etch:
(a) metric data of Reference Environment, the i.e. geographical location of consulting activity spot, space size, side are recorded every time
Position, temperature, humidity, height, air quality handle the metric data of Reference Environment according to standard value;
(b) according to the obtained deep learning model of training, to the geographical location of metric data extraction consulting activity spot,
Space size, orientation, temperature, humidity, height, the feature of air quality;
(c) prediction model obtained using training, and probabilistic forecasting is carried out to registration result according to the feature of extraction.
The present invention proposes a kind of method predicted registration success rate, by the factor item for acquiring Reference Environment
Part, using deep learning technology extract geographical location in relation to consulting activity spot, space size, orientation, temperature, humidity,
Highly, the feature of air quality carries out model training and marking, final prediction probability is obtained by model.The present invention is based on
Be objective environmental factor, and measure mode it is simple, data also be easy obtain, automatic Prediction can be formed, in conjunction with machine
Study, extracts appropriate environmental characteristic, using trained data model, carries out prediction probability, realizes and successfully may be used to registering
The automatic Prediction of energy property.
Claims (6)
1. a kind of method of the metric analysis prediction registration success rate based on Reference Environment, characterized in that including the training stage and
Evaluation and test stage, the training stage refer to:By acquiring the conditions of Reference Environment, instructed using deep learning technology
Practice the geographical location of consulting activity spot, the feature extraction mould of space size, orientation, temperature, humidity, height, air quality
Type extracts feature according to Feature Selection Model, and final training obtains prediction model;The evaluation and test stage refers to:By right
The measurement of Reference Environment, the geographical location of extraction consulting activity spot, space size, orientation, temperature, humidity, height, air
The feature of quality carries out prediction of result using prediction model.
2. a kind of method of metric analysis prediction registration success rate based on Reference Environment according to claim 1, special
Sign is that the training stage, steps are as follows:
(1) the metric data collection for establishing Reference Environment, by the site surrounding situation of passing registration consulting of nearest two naturology years
It records and the result of corresponding registration success or not is associated;It is big to geographical location, the space of consulting activity spot
Small, orientation, temperature, humidity, height, the feature of air quality are manually given a mark;Set the geographical position of consulting activity spot
Set, space size, orientation, temperature, humidity, height, air quality are divided into five grades, wherein five grades are corresponding
Numerical value is 0,1,2,3,4;
(2) the metric data collection in step (1) is used, it includes the geographical location of consulting activity spot, space size, sides
The artificial marking of position, temperature, humidity, height, air quality utilizes the geography of deep learning model training consulting activity spot
Position, space size, orientation, temperature, humidity, height, air quality characteristic model;
(3) characteristic model obtained using training extracts geographical location, space in relation to consulting activity spot to registration result
The feature score of size, orientation, temperature, humidity, height, air quality, and final prediction is trained according to linear regression algorithm
Model.
3. a kind of method of metric analysis prediction registration success rate based on Reference Environment according to claim 2, special
Sign is in step (1), the metric data of Reference Environment to be handled according to standard value, wherein geographical location longitude and latitude,
Space size with square metre, orientation gate import direction, temperature with degree Celsius, in humidity air actual water vapor pressure at that time
The percentage of the ratio between saturation vapour pressure under temperature indicates and round numbers, and rice, air quality AQI is highly used to register successfully to tie
Fruit indicates success with 1, and failure is indicated with 0.
4. a kind of method of metric analysis prediction registration success rate based on Reference Environment according to claim 2, special
Sign is, in step (2), the deep learning model includes deep neural network, convolutional neural networks and cycle nerve net
Network;Deep learning is the branch of machine learning, is that one kind attempting use and constituted comprising labyrinth or by multiple nonlinear transformation
Multiple process layers data are carried out with the algorithm of higher level of abstraction, the feature of extraction include the geographical location of consulting activity spot,
Space size, orientation, temperature, humidity, height, air quality, these are characterized in that deep learning algorithm is automatically learned.
5. a kind of method of metric analysis prediction registration success rate based on Reference Environment according to claim 2, special
Sign is, in step (3), the geographical location of setting consulting activity spot, space size, orientation, temperature, humidity, height, sky
Makings amount is divided into five grades, wherein the corresponding numerical value of five grades is 0,1,2,3,4, linear regression algorithm formula is such as
Under:Y=AX+b, the formula are vector forms, wherein Y is final score, and A and b are prediction model parameters, and A is vector, and b is mark
Amount, X are the feature vectors of extraction.
6. a kind of side of metric analysis prediction registration success rate based on Reference Environment according to Claims 2 or 3 or 4
Method, characterized in that the evaluation and test stage etch is as follows:
(a) metric data of Reference Environment, the i.e. geographical location of consulting activity spot, space size, orientation, temperature are recorded every time
Degree, humidity, height, air quality handle the metric data of Reference Environment according to standard value;
(b) the deep learning model obtained according to training, to geographical location, the space of metric data extraction consulting activity spot
Size, orientation, temperature, humidity, height, the feature of air quality;
(c) prediction model obtained using training, and probabilistic forecasting is carried out to registration result according to the feature of extraction.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810303689.3A CN108596370A (en) | 2018-03-30 | 2018-03-30 | A method of the metric analysis prediction registration success rate based on Reference Environment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810303689.3A CN108596370A (en) | 2018-03-30 | 2018-03-30 | A method of the metric analysis prediction registration success rate based on Reference Environment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108596370A true CN108596370A (en) | 2018-09-28 |
Family
ID=63625459
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810303689.3A Pending CN108596370A (en) | 2018-03-30 | 2018-03-30 | A method of the metric analysis prediction registration success rate based on Reference Environment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108596370A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101118620A (en) * | 2007-09-18 | 2008-02-06 | 吉林大学 | Vehicle gear shifting quality evaluation method based on nerval net |
CN103218669A (en) * | 2013-03-19 | 2013-07-24 | 中山大学 | Intelligent live fish cultivation water quality comprehensive forecasting method |
CN106127333A (en) * | 2016-06-21 | 2016-11-16 | 北京大学 | Movie attendance Forecasting Methodology and system |
US20170083937A1 (en) * | 2015-09-18 | 2017-03-23 | Mms Usa Holdings Inc. | Micro-moment analysis |
CN106844178A (en) * | 2017-01-22 | 2017-06-13 | 腾云天宇科技(北京)有限公司 | Prediction is presented method, computing device, server and the system of information transferring rate |
-
2018
- 2018-03-30 CN CN201810303689.3A patent/CN108596370A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101118620A (en) * | 2007-09-18 | 2008-02-06 | 吉林大学 | Vehicle gear shifting quality evaluation method based on nerval net |
CN103218669A (en) * | 2013-03-19 | 2013-07-24 | 中山大学 | Intelligent live fish cultivation water quality comprehensive forecasting method |
US20170083937A1 (en) * | 2015-09-18 | 2017-03-23 | Mms Usa Holdings Inc. | Micro-moment analysis |
CN106127333A (en) * | 2016-06-21 | 2016-11-16 | 北京大学 | Movie attendance Forecasting Methodology and system |
CN106844178A (en) * | 2017-01-22 | 2017-06-13 | 腾云天宇科技(北京)有限公司 | Prediction is presented method, computing device, server and the system of information transferring rate |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Batary et al. | Evidence of an edge effect on avian nest success | |
Carey et al. | Relationships between dimensions of supervisors' influence and counselor trainees' performance. | |
CN107705033A (en) | A kind of quality of instruction evaluation system based on big data | |
Vondrák et al. | Methods for obtaining more complete species lists in surveys of lichen biodiversity | |
CN106997475A (en) | A kind of insect image-recognizing method based on parallel-convolution neutral net | |
US11694005B2 (en) | Method, system and equipment for vegetation restoration or rehabilitation based on machine learning | |
Waddle | Use of amphibians as ecosystem indicator species | |
CN107819810A (en) | Adaptive planning formulates system | |
Vanclay | Site productivity assessment in rainforests: an objective approach using indicator species | |
CN108596370A (en) | A method of the metric analysis prediction registration success rate based on Reference Environment | |
CN106373135A (en) | Color-based pest identifying and counting method | |
CN111563443A (en) | Continuous motion action accuracy evaluation method | |
CN108573051A (en) | Knowledge point collection of illustrative plates based on big data analysis | |
Saner et al. | Reduced soil respiration in gaps in logged lowland dipterocarp forests | |
Iverson et al. | Adapting oak management in an age of ongoing mesophication but warming climate | |
Delany et al. | Sources of variation in the abundance and detection of the endangered Florida grasshopper sparrow | |
Rana et al. | Measuring canopy anomaly influence on golf putt kinematics: does annual bluegrass influence ball roll behavior? | |
Martin et al. | Shorebird abundance estimates in interior Alaska | |
Shen et al. | Discrete fourier transformation for seasonal-factor pattern classification and assignment | |
Ramya et al. | A Customisable AI Deck for Pitch Reports and Automated III Umpire Decision Review System DRS | |
CN115798676B (en) | Interactive experience analysis management method and system based on VR technology | |
CN113191650B (en) | Student offline teaching management method based on online education software platform data feature matching | |
CN109919375B (en) | Method and device for predicting examination key students based on big data | |
Rotenberg et al. | An integrated community-based harpy eagle and avian conservation program for the maya mountains massif, Belize | |
Mahdavivafa et al. | INVESTIGATING THE FACTORS OF COMMUNICATION AND INFORMATION RESOURCES AFFECTING THE TENDENCY OF LOCAL COMMUNITIES TO PARTICIPATE IN THE COMPREHENSIVE MANAGEMENT OF HABLEHROUD WATERSHED IN TEHRAN PROVINCE |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | 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 |
Application publication date: 20180928 |
|
RJ01 | Rejection of invention patent application after publication |