CN107798124A - Search system and method based on prediction modeling technique - Google Patents
Search system and method based on prediction modeling technique Download PDFInfo
- Publication number
- CN107798124A CN107798124A CN201711107547.1A CN201711107547A CN107798124A CN 107798124 A CN107798124 A CN 107798124A CN 201711107547 A CN201711107547 A CN 201711107547A CN 107798124 A CN107798124 A CN 107798124A
- Authority
- CN
- China
- Prior art keywords
- data
- unit
- modeling
- analysis
- result set
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/211—Schema design and management
- G06F16/212—Schema design and management with details for data modelling support
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2453—Query optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Computational Linguistics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to the search system and method based on prediction modeling technique.The system includes database, data cleansing unit, data preparation unit, modeling analysis unit, modeling test cell, result set, the database is connected with the data cleansing unit networks, the data cleansing unit is connected with the data preparation unit networks, the data preparation unit is connected with the modeling analysis unit networks, the modeling analysis unit and the modeling test cell network connection, the modeling test cell is connected by model output unit and the result set network connection, the result set with the data cleansing unit networks.The present invention substantially increases mass data inquiry velocity, the qualitative Query Result degree of accuracy.
Description
Technical field
The invention belongs to model, search technique field, be related to based on prediction modeling technique search system and method.
Background technology
Searching method be in order to realize the search target in search plan used by concrete operation method and means it is total
Claim.Traditional searching method includes auditing up to trail balance, auditing from trail balance, test check, retroactive method, discrete method, browsing, regardless of search
Method, all it is to search out substantial amounts of historical data around problem, gradually distinguish and accept or reject, required for therefrom finds out problem
Data.
This traditional search system and method have following weak point：
Just be to seek out coming first is mass data, and data volume is big, and inquiry velocity is slow；
Data include various incomplete datas, wrong data, duplicate data etc., cause qualitative Query Result poor accuracy；
In order to solve the deficiency of traditional searching method, the present invention intend proposing search system based on prediction modeling technique and
Method.
The content of the invention
Slow in order to solve conventional search system and method inquiry velocity, the problem of Query Result poor accuracy, the present invention is first
A kind of search system based on prediction modeling technique is first proposed, it is whole that the system includes database, data cleansing unit, data
Unit, modeling analysis unit, modeling test cell, result set are managed, wherein：
The database is connected with the data cleansing unit networks；
The data cleansing unit is connected with the data preparation unit networks；
The data preparation unit is connected with the modeling analysis unit networks；
The modeling analysis unit and the modeling test cell network connection；
The modeling test cell passes through model output unit and the result set network connection；
The result set is connected with the data cleansing unit networks.
Further, the data cleansing unit also includes descriptive analysis unit, the descriptive analysis unit and the number
According to finishing unit network connection.
Further, the result set includes modeling log unit, modeling analysis process unit, data summarization unit, people
For factor unit, each unit is all the data part of result set.
Searching method based on prediction modeling technique is carried out according to the system, the described method comprises the following steps：
Step S1, gathered using data acquisition equipment after the first data enter line program parsing and be stored in database, the number
First data are sent to data cleansing unit according to storehouse；
Step S2, the data cleansing unit receive first data and carry out data cleansing, abandon number of non-compliances evidence,
The second qualified data are sent to data preparation unit by network；
Step S3, the data preparation unit receive second data and carry out data preparation analysis, the data that will be obtained
Feature is sent to modeling analysis unit by network；
Step S4, the modeling analysis unit establishes data model according to the data characteristics, according to the data model
The parameter value of model is calculated, the parameter value of the model is passed into modeling test cell by network；
Step S5, the modeling test cell carry out data search according to the parameter value for establishing model, utilize special survey
Die trial type is tested the data of search；The method that modeling analysis is changed according to the degree of accuracy of test, test result mistake
Data return to the modeling analysis unit amendment data model, the correct data of test result are passed through model output unit
It is sent to the result set；
Step S6, the result set store and apply the 3rd data to carry out event prediction.
Further,, can be with to there is the data of specific demand during data cleansing unit cleaning data in step S2
Using the descriptive analysis unit of the data cleansing unit, the data analysis rule of artificial disturbance is added, qualified after analysis
Second data are sent to the data preparation unit by network.
Further, the mode of the data cleansing unit cleaning data includes checking data consistency, handles invalid value
And missing values；Remove incomplete data, wrong data, duplicate data；Data normalization；Data are classified.
Further, the mode of the data preparation unit progress data preparation analysis includes data preparation, data correlation.
Further, the method for the modeling analysis Modelon Modeling mainly make use of data analysis method, from substantial amounts of observation
In data, statistical method founding mathematical models are utilized.
Further, the result set storage comprises the following steps using the mode of the 3rd data：
Utilize the daily record of the modeling log unit record whole system course of work；
Modeling analysis process is recorded using the modeling process analytic unit, the specific business point for each service part
Analysis；
The artificial business demand added during being analyzed using the human factor unit record；
Collect last the 3rd data using the data summarization unit.
Further, the result set, obtained the 3rd data is sent to the data cleansing unit, enter line number
According to cleaning again, with correction model.
The application of the present invention obtains obviously benefit：
Coordinate data cleansing, data preparation, modeling analysis, modeling test, mass data data volume is reduced；
According to timed task, such as：Timer, Quartz etc., summary data rule, draw data model；
2 points of the above substantially increases mass data inquiry velocity, the Query Result degree of accuracy.
Brief description of the drawings
Fig. 1 is the working-flow figure of embodiment 1.
Fig. 2 is the working-flow figure of embodiment 2.
Fig. 3 is result set data pie graph.
Embodiment
Below in conjunction with drawings and Examples, the embodiment of the present invention is described in more details, so as to energy
The advantages of enough more fully understanding the solution of the present invention and its various aspects, however, specific embodiments described below is only
The purpose of explanation, rather than limitation of the present invention.
Present invention firstly provides a kind of search system based on prediction modeling technique, as shown in figure 1, the system includes
Database 1, data cleansing unit 2, data preparation unit 3, modeling analysis unit 4, modeling test cell 5, result set 7.Wherein：
The database 1 and the network connection of data cleansing unit 2；The data cleansing unit 2 and the net of data preparation unit 3
Network connects；The data preparation unit 3 and the network connection of modeling analysis unit 4；The modeling analysis unit 4 is built with described
The network connection of mould test cell 5；The modeling test cell 5 is connected by the model output unit 6 and the network of result set 7
Connect；The result set 7 and the network connection of data cleansing unit 2.
As shown in Fig. 2 the data cleansing unit 2 also includes descriptive analysis unit 21, the descriptive analysis unit 21 with
The network connection of data preparation unit 3.The result set 7 includes modeling log unit 71, modeling analysis process unit 72, number
According to collection unit 73, human factor unit 74, each unit is all the data part of result set 7.
Searching method based on prediction modeling technique is carried out according to the system, embodiment is described in detail as follows.
(1) data acquisition
The first data are gathered using various data acquisition equipments, enters line program parsing, is stored in database 1, database 1
First data are sent to the data cleansing unit 2.
Modeling not only needs substantial amounts of data, while data must be reliable, and is adapted to the requirement of modeling.The data of distortion with
And do not meet modeling data must by analysis, be subject to proper treatment.Data are the premise and important evidence of prediction work, in advance
Survey can not be fabrication and fantasy, and the development of anything has certain rule, and conscientiously research is predicted object and fully investigated pre-
The environment residing for object is surveyed, past and present data are summarized in the method for network analysis, therefrom find out rule,
Scientifically infer future.
Data mainly have two effects in prediction：First, for the behavior mould for determining to be made up of some historical limitations points
Type；Second, the future value of independent variable is determined in Causal model prediction.The starting stage of prediction, it is the receipts for being engaged in data first
Collection, arrange, process and analyze, good condition is created for modeling.
(2) data cleansing
First data that the database 1 sends are received using the data cleansing unit 2 and carry out data cleansing, are lost
Number of non-compliances evidence is abandoned, the second qualified data are sent to the data preparation unit 3 by network.
The mode that the data cleansing unit 2 cleans data includes：Check data consistency, processing invalid value and missing
Value, remove incomplete data, wrong data, duplicate data；Data normalization；Data are classified.
To there are the data of specific demand, the descriptive analysis unit 21 of the data cleansing unit 2 can be utilized, is added artificial
The second qualified data, the data preparation unit 3 is sent to after analysis by the data analysis rule of interference by network.It is described
The data analysis rule of artificial disturbance is during data cleansing, can add the conclusion that history cleaning data are drawn, so as to
Strengthen the accuracy of data result collection 7.
(3) data preparation
The data preparation unit 3 receives the second data that the data cleansing unit 2 is sent and carries out data preparation analysis,
Including data preparation, data correlation, obtained data characteristics is sent to the modeling analysis unit 4 by network.
Data preparation principle：Accurately, the data after processing can correctly reflect the future trend and situation of things development；And
When, the processing of data is timely；It is applicable, the data of processing can meet the needs of modeling；Economy, data processing is reduced as far as possible
Expense, to reduce forecast cost；Unanimously, the data of processing are whole comparative, it is interior during use must be it is consistent, having can
It is comparative.
Data preparation method：Diagnostic method, by the judgement to historical data, selection can be wherein represented during whole prediction
The data for the pattern being likely occurred are as modeling data.Scalping method, if data volume is bigger, and nonessential continuous data
Amount, at this moment can be rejected in data by the exceptional value of random disturbances.Mean value method, it is fewer or when needing continuous data in data,
Can then mean value method be taken to handle data.Even up method, because condition changes, usually prevent some historical datas from
Reflect current situation, at this moment the method for evening up is a kind of preferable method, and its principle is to add one to fit to the data before turning point
When value, make it consistent with the data trend after break；The change of rule of three, sales terms and environment can usually cause one
The change of enterprise product market sale ratio, it is exactly a kind of relatively effective processing method with rule of three processing data now；Move
Dynamic average and exponential smoothing, if totally trend has certain regularity to initial data, but because being disturbed by enchancement factor, data
Dispersion is very big, is also difficult to handle using mean value method.At this moment it can use once, be secondary, even rolling average and index three times
Smoothly data are carried out smoothly, with smooth data modeling, when decomposing prediction, to handle season data, then to use high
The method of moving average of power, to data smoothing；Calculus of finite differences, if initial data is Non-stationary Data, need to take difference processing.
Difference has three kinds of main Types：Forward difference, backward difference, centered difference.
Data correlation：Data correlation is carried out to the data put in order, the latitude of data is reduced using the correlation of data.
During prediction, because prediction object is different, content is different, and the time limit is different, required data intension and quantity
Also it is different.The data preparation unit 3 makes the second data turn into content by collating sort, data preparation and data correlation
Completely, in order, system, concise unified data in form.
(4) modeling analysis
The data characteristics that the modeling analysis unit 4 transmits according to the data preparation unit 3 establishes data model, according to
The parameter value that data model calculates model passes to the modeling test cell 5 by network.
System modelling is mainly used in three aspects：
Analysis and design real system, such as engineering circles, when analysis designs a new system, usual advanced line number is imitative
True and physical simulation experiment, finally arrives scene and makees full-scale investigation, mathematical simulation is simpler than physical simulation, easy, is imitated with mathematics again
When very to analyze and design a real system, it is necessary to have the model of a description system features, for many complicated industry
Control process, modeling are often most critical and most difficult task, and the qualitative or quantitative research to society and economic system is also
Set about from modeling, such as in population cybernetics, establish various types of demographic models, change some parameters in model,
Influence of the population policy that can analyze and research for population development.
The future developing trend of some states of prediction or forecast real system, prediction or forecast are based on things development process
Continuity, such as the mathematical modeling of meteorological change is established according to conventional measurement data, for forecasting the meteorology in future.
Optimum control is carried out to system, the key of controller or optimal control law is designed with control theory or on condition that has
One mathematical modeling that can characterize system features.On the basis of modeling, further according to maximal principle, Dynamic Programming, feedback, solution
The methods of coupling, POLE PLACEMENT USING, self-organizing, adaptive and intelligent control, design various controllers or control law.
System modelling is mainly used in three aspects, for same real system, people can according to different purposes and
Purpose establishes different models.But any model established all is the simplification of real system prototype, because both can not possibly or not have
Necessary all details real system all include.If some essence that can retain system prototype in simplified model are special
Sign, then just it is believed that model to system prototype is similar, may be employed to description original system.Therefore, during actual modeling,
It must be made between the simplification of model and precision of analysis appropriate compromise, this is often the original that modeling follows
Then.
The result that the system modeling calculates is mainly used in predicting or forecasting future developing trend, has directive significance to future.
Conventional modeling method has following several:
Elementary mathematics method, it is mainly used in some static, linear, deterministic models.For example, Modelling for Seats Distribution Problem, student
The comparison of achievement, some simple infectious disease static models.
Data analysis method, from substantial amounts of observation data, using statistical method founding mathematical models, it common are：Return
Analytic approach, time Sequence Analysis Method.
Emulation and other method.Mainly there is computer simulation, be equivalent to sampling test, can be with discrete system simulation and continuous
System simulation, factorial experiment method, mainly does local test in system, carries out constantly analysis modification according to result of the test, asks
Model structure needed for obtaining, artificial reality method, based on the understanding to system and the target to be reached, artificially forming one is
System, analytic hierarchy process (AHP), is mainly used in relevant Economic planning and management, energy decision making and distribution, behavior science, military science, army
The fields such as thing commander, transport, agricultural, education, the talent, medical treatment, environment, to carry out decision-making, evaluation, prediction etc..The party
The crucial step of method is to establish hierarchy Model.
The method that the modeling analysis unit 4 models mainly make use of data analysis method, from substantial amounts of observation data, profit
With statistical method founding mathematical models.Specifically include：The characteristic value of coefficient matrix is obtained according to the methods of Jacobian matrix；According to
The data that previous step is taken out, select important composition, write out the expression formula of main component；According to the data drawn, further unite
Meter analysis.
(5) modeling test
The modeling test cell 5 scans for according to the parameter value for establishing model, using special test model to searching
The data of rope are tested, the method that modeling analysis is changed according to the degree of accuracy of test.The data of test result mistake are returned
Data model is corrected to the modeling analysis unit 4, the correct data of test result are sent to result by model output unit 6
Collection 7.
(6) result set
The result set 7 includes modeling log unit 71, modeling analysis process unit 72, data summarization unit 73, artificial
Factor unit 74, each unit are all the data parts of result set 7.
The result set 7 records the daily record of the whole system course of work using log unit 71 is modeled, and utilizes modeling process
Analytic unit 72 records modeling analysis process, for the specific business diagnosis of each service part, utilizes human factor unit 74
The artificial business demand added during record analysis, the 3rd last data are collected using data summarization unit 73, and together
When the 3rd data are sent to data cleansing unit 2, carry out the cleaning again of data.
The data that result set 7 can be utilized to store are scanned for and predicted：First, the result set 7 can be used as search
The initial data of data, because the data of result set 7 are treated, so as to improve the efficiency of search.Meanwhile summarize
The data rule of the search of user, is added in human factor, improves the accuracy and utilizability of search data；Secondly, lead to
The Rule Summary of result set 7 is crossed, such as：Nearest one week 1 of tourist attractions：00-2：00 flow of the people is very big, then according to this
As a result it is predicted, this period should limit visitor's quantity.
Embodiment 1
As shown in figure 1, Fig. 1 is the working-flow figure of embodiment 1, the system include database 1, data cleansing unit 2,
Data preparation unit 3, modeling analysis unit 4, modeling test cell 5, result set 7.
Described according to above-mentioned embodiment, the database 1 stores the mass data that collecting device collection comes and carried out
After program parsing, it is sent to the data cleansing unit 2 and carries out data cleansing, number of non-compliances evidence is abandoned, the second qualified data
The data preparation unit 3 is sent to by network.The mode that the data cleansing unit 2 cleans data includes：Check data
Uniformity, processing invalid value and missing values；Remove incomplete data, wrong data, duplicate data；Data normalization；Data are classified.
The data preparation unit 3 receives second data and carries out data preparation, data correlation, is processed into complete in content
Whole, orderly, system, concise unified data, extraction data characteristics are sent to the modeling analysis unit 4 in form.
The maintenance data analytic approach of modeling analysis unit 4, from substantial amounts of observation data, number is established using statistical method
Model is learned, the parameter value that model is calculated according to data model is sent to the modeling test cell 5 by network.
The modeling test cell 5 scans for according to the parameter value of foundation, using special test model to search
Data are tested, the method that modeling analysis is changed according to the degree of accuracy of test.The data of test result mistake are returned to institute
State modeling analysis unit 4 and correct data model, the correct data of test result are sent to the result by model output unit 6
Collection 7.
Such as Fig. 3, the result set 7 includes modeling log unit 71, modeling analysis process unit 72, human factor unit
74th, data summarization unit 73, each unit are all the data parts of result set 7.
The result set 7 records the daily record of the whole system course of work using the modeling log unit 71, using described
Modeling process analytic unit 72 records modeling analysis process, for the specific business diagnosis of each service part, utilizes the people
For the artificial business demand added during the record analysis of factor unit 74, collected finally using the data summarization unit 73
The 3rd data, and the 3rd data are sent to data cleansing unit 2 simultaneously, carry out the cleaning again of data.
The data that can finally utilize result set 7 to store are scanned for and predicted：First, the result set 7 can conduct
The initial data of data is searched for, because the data of result set 7 are treated, so as to improve the efficiency of search.Meanwhile
The data rule of the search of user is summarized, is added in human factor, improves the accuracy and utilizability of search data；Its
It is secondary, the development trend of things is predicted and handled by the Rule Summary of result set 7.
Embodiment 2
As shown in Fig. 2 Fig. 2 is the working-flow figure of embodiment 2, the system include database 1, data cleansing unit 2,
Data preparation unit 3, modeling analysis unit 4, modeling test cell 5, result set 7.The data cleansing unit 2 also includes description
Analytic unit 21.
Described according to above-mentioned embodiment, the database 1 stores the mass data that collecting device collection comes and carried out
After program parsing, it is sent to the data cleansing unit 2 and carries out data cleansing, number of non-compliances evidence is abandoned, the second qualified data
The data preparation unit 3 is sent to by network.
The mode that the data cleansing unit 2 cleans data includes：Check data consistency, processing invalid value and missing
Value；Remove incomplete data, wrong data, duplicate data；Data normalization；Data are classified.
To there are the data of specific demand, using the descriptive analysis unit 21 of the data cleansing unit 2, artificial disturbance is added
Data analysis rule, that is, add the conclusion that draws of history cleaning data, the second qualified data passed by network after analysis
The data preparation unit 3 is sent to, can so strengthen the accuracy of data result collection 7.
The other each unit embodiments of the present embodiment such as embodiment 1.
Finally it should be noted that：Above-described embodiment is only intended to clearly illustrate example of the present invention, and is not pair
The restriction of embodiment.For making other changes in different forms on the basis of the above description, still in this hair
Among bright protection domain.
Claims (10)
1. a kind of search system based on prediction modeling technique, the system include database, data cleansing unit, data preparation
Unit, modeling analysis unit, modeling test cell, result set, wherein：
The database is connected with the data cleansing unit networks；
The data cleansing unit is connected with the data preparation unit networks；
The data preparation unit is connected with the modeling analysis unit networks；
The modeling analysis unit and the modeling test cell network connection；
The modeling test cell passes through model output unit and the result set network connection；
The result set is connected with the data cleansing unit networks.
2. system according to claim 1, it is characterised in that the data cleansing unit also includes descriptive analysis unit,
The descriptive analysis unit is connected with the data preparation unit networks.
3. system according to claim 1, it is characterised in that the result set includes modeling log unit, modeling analysis
Process unit, data summarization unit, human factor unit, each unit are all the data parts of result set.
4. a kind of searching method based on prediction modeling technique, comprises the following steps：
Step S1, gathered using data acquisition equipment after the first data enter line program parsing and be stored in database, the database
First data are sent to data cleansing unit；
Step S2, the data cleansing unit receive first data and carry out data cleansing, number of non-compliances evidence are abandoned, qualified
The second data data preparation unit is sent to by network；
Step S3, the data preparation unit receive second data and carry out data preparation analysis, the data characteristics that will be obtained
Modeling analysis unit is sent to by network；
Step S4, the modeling analysis unit establish data model according to the data characteristics, are calculated according to the data model
Go out the parameter value of model, the parameter value is passed into modeling test cell by network；
Step S5, the modeling test cell carry out data search according to the parameter value for establishing model, utilize special test mould
Type is tested the data of search, the method that modeling analysis is changed according to the degree of accuracy of test, the number of test result mistake
According to the modeling analysis unit amendment data model is returned to, by model output unit send the correct data of test result to
The result set；
Step S6, the result set store and apply the 3rd data to carry out event prediction.
5. according to the method for claim 4, it is characterised in that in step S2, when the data cleansing unit cleans data,
To there are the data of specific demand, the descriptive analysis unit of the data cleansing unit can be utilized, adds the data of artificial disturbance
Analysis rule, the second qualified data are sent to the data preparation unit by network after analysis.
6. according to the method for claim 4, it is characterised in that in step S2, the data cleansing unit cleaning data
Mode includes checking data consistency, processing invalid value and missing values；Remove incomplete data, wrong data, duplicate data；Data
Standardization；Data are classified.
7. according to the method for claim 4, it is characterised in that in step S3, it is whole that the data preparation unit carries out data
The mode of reason analysis includes data preparation, data correlation.
8. according to the method for claim 4, it is characterised in that in step S4, the method for the modeling analysis Modelon Modeling
Data analysis method mainly is make use of, from substantial amounts of observation data, utilizes statistical method founding mathematical models.
9. according to the method for claim 4, it is characterised in that in step S5, the result set stores and applies described the
The mode of three data comprises the following steps：
Utilize the daily record of the modeling log unit record whole system course of work；
Modeling analysis process, the specific business diagnosis for each service part are recorded using the modeling process analytic unit；
The artificial business demand added during being analyzed using the human factor unit record；
Collect last the 3rd data using the data summarization unit.
10. according to the method for claim 4, it is characterised in that the result set, obtained the 3rd data are passed through
Network is sent to the data cleansing unit, the cleaning again of data is carried out, with correction model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711107547.1A CN107798124A (en) | 2017-11-10 | 2017-11-10 | Search system and method based on prediction modeling technique |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711107547.1A CN107798124A (en) | 2017-11-10 | 2017-11-10 | Search system and method based on prediction modeling technique |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107798124A true CN107798124A (en) | 2018-03-13 |
Family
ID=61534813
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711107547.1A Pending CN107798124A (en) | 2017-11-10 | 2017-11-10 | Search system and method based on prediction modeling technique |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107798124A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109886843A (en) * | 2019-03-13 | 2019-06-14 | 山东浪潮云信息技术有限公司 | The processing method of multi-destination data in a kind of tourism data |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101141456A (en) * | 2007-10-09 | 2008-03-12 | 南京财经大学 | Vertical search based network data excavation method |
| US20100257133A1 (en) * | 2005-05-09 | 2010-10-07 | Crowe Keith E | Computer-Implemented System And Method For Storing Data Analysis Models |
| CN105894336A (en) * | 2016-05-25 | 2016-08-24 | 北京比邻弘科科技有限公司 | Mobile Internet-based big data mining method and system |
| CN106815338A (en) * | 2016-12-25 | 2017-06-09 | 北京中海投资管理有限公司 | A kind of real-time storage of big data, treatment and inquiry system |
| CN106933956A (en) * | 2017-01-22 | 2017-07-07 | 深圳市华成峰科技有限公司 | Data digging method and device |
-
2017
- 2017-11-10 CN CN201711107547.1A patent/CN107798124A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100257133A1 (en) * | 2005-05-09 | 2010-10-07 | Crowe Keith E | Computer-Implemented System And Method For Storing Data Analysis Models |
| CN101141456A (en) * | 2007-10-09 | 2008-03-12 | 南京财经大学 | Vertical search based network data excavation method |
| CN105894336A (en) * | 2016-05-25 | 2016-08-24 | 北京比邻弘科科技有限公司 | Mobile Internet-based big data mining method and system |
| CN106815338A (en) * | 2016-12-25 | 2017-06-09 | 北京中海投资管理有限公司 | A kind of real-time storage of big data, treatment and inquiry system |
| CN106933956A (en) * | 2017-01-22 | 2017-07-07 | 深圳市华成峰科技有限公司 | Data digging method and device |
Non-Patent Citations (4)
| Title |
|---|
| EVILLIST: "数据挖掘的一般过程", 《CSDN》 * |
| 付金会: "《统计学》", 30 April 2017, 中国矿业大学出版社 * |
| 西电校草: "数据挖掘一般流程", 《CSDN》 * |
| 黄丽霞等: "《信息检索教程》", 31 July 2014, 知识产权出版社 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109886843A (en) * | 2019-03-13 | 2019-06-14 | 山东浪潮云信息技术有限公司 | The processing method of multi-destination data in a kind of tourism data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Song et al. | Environmental performance evaluation with big data: Theories and methods | |
| Wang et al. | Insight of a fuzzy regression model | |
| CN110866528B (en) | Model training method, energy consumption use efficiency prediction method, device and medium | |
| CN106453293A (en) | Network security situation prediction method based on improved BPNN (back propagation neural network) | |
| CN104091216A (en) | Traffic information predication method based on fruit fly optimization least-squares support vector machine | |
| WO2017071369A1 (en) | Method and device for predicting user unsubscription | |
| CN110690701A (en) | Analysis method for influence factors of abnormal line loss | |
| CN106296315A (en) | Context aware systems based on user power utilization data | |
| CN109934469A (en) | Based on the heterologous power failure susceptibility method for early warning and device for intersecting regression analysis | |
| Yu et al. | An active construction dynamic schedule management model: using the fuzzy earned value management and BP neural network | |
| CN111199493A (en) | Arrearage risk identification method based on customer payment information and credit investigation information | |
| CN112419711A (en) | Closed parking lot parking demand prediction method based on improved GMDH algorithm | |
| Aghimien et al. | A review of the application of data mining for sustainable construction in Nigeria | |
| CN114548494A (en) | Visual cost data prediction intelligent analysis system | |
| CN109117352B (en) | Server performance prediction method and device | |
| CN117370766A (en) | Satellite mission planning scheme evaluation method based on deep learning | |
| CN112860973A (en) | Organization health assessment method based on social network analysis | |
| CN107798124A (en) | Search system and method based on prediction modeling technique | |
| Narasimhan | A Fuzzy Subset Characterization Of A Site‐Selection Problem | |
| CN111597176A (en) | Teaching simulation training method and system for delaying supply chain generation | |
| CN109978138A (en) | The structural reliability methods of sampling based on deeply study | |
| CN114155072B (en) | Financial prediction model construction method and system based on big data analysis | |
| Rojek | Neural networks as prediction models for water intake in water supply system | |
| CN102096692A (en) | System and method for evaluating values of website visitors | |
| Sun et al. | An automated warehouse sorting system for small manufacturing enterprise applying discrete event simulation |
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: 20180313 |
|
| RJ01 | Rejection of invention patent application after publication |