CN106651424A - Electric power user figure establishment and analysis method based on big data technology - Google Patents
Electric power user figure establishment and analysis method based on big data technology Download PDFInfo
- Publication number
- CN106651424A CN106651424A CN201610860951.5A CN201610860951A CN106651424A CN 106651424 A CN106651424 A CN 106651424A CN 201610860951 A CN201610860951 A CN 201610860951A CN 106651424 A CN106651424 A CN 106651424A
- Authority
- CN
- China
- Prior art keywords
- user
- sample
- portrait
- data
- power consumer
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F18/24155—Bayesian classification
-
- 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
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
Abstract
The invention discloses an electric power user figure establishment and analysis method based on the big data technology. The method comprises steps that the historical electricity information, basic attributes, the fee-paying information and the appeal information of electric power users are acquired; classification category sets of user figures are determined, an influence factor set of a classification result is determined, and a mapping relationship between the influence factor set and the classification set is determined; random extraction of the acquired data is carried out, one part of the data is taken as a training sample, and other data is taken as prediction sample; normalization processing, discretization processing and attribute reduction for the training sample and the prediction sample are carried out, and an influence factor set after correction is determined; the training sample is trained, ten-fold cross validation is taken as a test mode, an electric power user figure prediction model based on a naive Bayes classifier is established, data classification mining analysis on the prediction sample is carried out through utilizing the prediction model, and electric power user figures are acquired. The method is advantaged in that electric power electric quantity prediction and management can be facilitated.
Description
Technical field
The present invention relates to a kind of power consumer portrait based on big data technology is set up and analysis method.
Background technology
Nowadays increasing industry starts the application for paying attention to user's portrait, but because different industries have different industries
Background, application scenarios and user's request, thus different industries user portrait can not it is same in.Finance, bank's industry do use
Family portrait is because that the consumption habit of present younger generation client there occurs change, and they do not like financial grid point and do business,
But select to carry out financial consumption by smart machine, and nowadays it is difficult have a kind of product while meeting proprietary demand.
Telecommunications industry needs to realize real-time and preciseization marketing such as flow package, telephone expenses set meal by user's portrait, while facing quantity
Huge customer group, accomplishes personal marketing.
With the deep propulsion of informatization and developing rapidly for power business, power grid enterprises are also accumulated from abundant preciousness
Data resource, depth excavates available data and simultaneously makes full use of data results aid decision, so study power network development and
Customer service rule, becomes one of important channel of driving power grid enterprises innovation and development.Therefore, carry out based on big data technology
Power consumer portrait research, formulates the marketing strategy of differentiation and precision, improves the competitiveness of products ＆ services, meets electric power
The increasingly diversified power supply service demand of client, expands occupation rate of the electric energy in social energy-consuming terminal very urgent
Cut.
There is very big value to enterprise from user's portrait of business perspective, user's portrait purpose there are two.One
It is that business scenario sets out, finds target customer.Another is exactly, with reference to the information of user's portrait, be user's deisgn product or
Carry out marketing activity.
Enterprise excavates the ascribed characteristics of population, behavior property, the social network of each user using the potential user group for searching out
The data such as network, psychological characteristics, hobby, through constantly superposition, update, and take out complete information labels, combine and build
Go out user's dummy model of a solid, i.e. user's portrait.And for power grid enterprises, power consumer portrait be then according to
The difference of the base attribute at family, electricity consumption behavior, paying behaviors and demand behavior, carries out tagsort, classification, from each type
Characteristic feature is extracted, the threshold value of label is given, according to final label, with reference to business demand scene, carries out power consumer individual
Portrait and colony's portrait.
The foundation of power consumer portrait improves service satisfaction, prediction visitor for differentiated service is promoted in power industry
Family behavior, reduces enterprises' loss, and correctly estimating for electricity suffers from vital effect.
Existing power consumer portrait, including services client relation management, services channels management, client's outage management, visitor
Family Portrait brand technology application scenarios planning, customer portrait label architectural study etc., there is successful story.Wherein Liaoning Electric Power Anshan
The project " the power system payment channel evaluation method based on big data " of electric company, based on questionnaire data, using K-
Means clustering algorithms carry out portrait analysis to subscriber payment behavior, identify all types of user attribute and all types of user paying behaviors
Corresponding relation.But this scheme has obvious defect from business and technically：From the perspective of in business, paying behaviors are user
Portrait and a part for credit rating, it is impossible to which equivalent is for it；Technically, the determination of K values is key in K-means algorithms,
Clustering Effect is for K values, noise spot and isolated point quite sensitive, and the experiment for generally requiring many times just can determine that optimum k value,
Efficiency is low；Due to each attribute it is different for user's portrait influence degree, it would be desirable to give different weights respectively, but K-
Means algorithms cannot determine specific weight so that classification results lack persuasion.
The content of the invention
The present invention is in order to solve the above problems, it is proposed that a kind of power consumer portrait based on big data technology set up with point
Analysis method, this method is divided to personal portrait and colony's two aspects of portrait to do customer analysis according to user data and business demand, obtains
The user's representation data for arriving is objective, comprehensive, beneficial to the application of electricity needs and electric power analysis.
To achieve these goals, the present invention is adopted the following technical scheme that：
A kind of power consumer portrait based on big data technology is set up and analysis method, comprises the following steps：
(1) history power information, base attribute, payment information and the demand information of power consumer are obtained；
(2) determine the set of user's portrait class categories, and the influence factor collection of classification results, determine influence factor collection
With the mapping relations of classification set；
(3) data of acquisition are randomly selected, used as training sample, it is with data as forecast sample for a part；
(4) training sample and forecast sample are normalized, sliding-model control and attribute reduction, it is determined that correction
Influence factor set afterwards；
(5) training sample is trained, and using ten folding cross validations as test pattern, sets up and be based on naive Bayesian
The power consumer portrait forecast model of grader, using forecast model data classified excavation analysis is carried out to forecast sample, is obtained
Power consumer is drawn a portrait.
In the step (1), the influence factor includes essential information, electricity consumption behavior, payment information, demand information and society
Friendship information, essential information includes sex, electricity consumption type, category of employment, the time limit of registering for a household residence card, supply voltage, the affiliated city class of user
Type and/or load character；User's portrait class categories include high-quality user, general user and low quality user, specifically
Credit rating is divided according to the standard of setting.
In the step (1), the electricity consumption behavior includes charge level, season peak of power consumption, the power supply quality sense of user
Condition in the know, promise breaking electricity consumption degree and stealing degree.
In the step (1), payment information includes meter reading method, cycle of checking meter, bill date of issue, the payment time limit of user
Day, consumption grade and/or payment channel.
In the step (1), the demand information includes demand mood, tolerance, intensity, demand preference, the electricity electricity charge
Preference and/or the guarantee preference that has a power failure.
In the step (1), extract from power system information system file and user's portrait strong correlation information, the degree of correlation
Division given a mark according to expert system and distinguished, search out after key factor with portrait index carry out correlation analysis, look for
Go out really to have in business the user behavior factor of strong correlation, the data source scope of label is determined based on this.
In the step (2), the essential characteristic of user is described by the base attribute label of user, using the electricity consumption of user
Behavior label describes the use electrical feature of user, confirms custom and characteristic during its electricity consumption, is believed using the payment of user
Breath label describes distribution and behavioral characteristic of the user during electricity consumption expense is paid, and is used using demand message reflection electric power
Various demands of the family during electrical power services are enjoyed.
In the step (3), used as training sample, remaining 80% data is used as forecast sample for the data using 20%.
In the step (4), data are normalized：
In formula：xijIt is to normalize front sample, sijIt is sample after normalization；min(xj) it is minimum of a value in original sample；
max(xj) it is maximum in original sample.
In the step (4), sliding-model control is carried out to training sample data：
In formula：zijFor sample after discretization, min (sj) be normalization after sample minimum of a value, max (sj) for normalization after
The maximum of sample, Q is step-length：
In the step (4), the concrete steps of attribute reduction include：After certain attribute is removed, if not found weight
Multiple training sample data, the relation that can not debate in decision table there occurs corresponding change, so this attribute is retained；With
This analogizes, and finally gives the influence factor set of determination.
In the step (5), the concrete grammar for setting up power consumer portrait forecast model is：
(5-1) the power consumer portrait forecast model based on Naive Bayes Classifier, shadow of the model to determine are set up
It is input vector to ring sets of factors, and with user's portrait class categories output vector is combined into；
(5-2) respectively in terms of detailed precision, confusion matrix and node error rate three, inspection power consumer portrait prediction
The accuracy of model.
In the step (5-2), detailed precision includes：TP Rate (real unit ratio), FP Rate (false positive element ratio),
Precision (precision), Recall (recall ratio), F-Measure (harmonic-mean of precision ratio and recall ratio) and ROC
Area。
In the step (5), the concrete grammar of data classified excavation analysis is carried out to forecast sample using forecast model：
(5-a) it is C to count number S of example training sample, classificationiSample number Si, k-th attribute AkEqual to xkAnd
Classification is CiTraining example number of samples Sik；
(5-b) probability of all categories is calculated respectively with middle attribute A of all categorieskEqual to xkProbability；
(5-c) grader is utilized
The ownership classification results of prediction example sample X are drawn, predicting the outcome for user's portrait is contrasted with actual conditions.
In the step (5-b),
Here, ScFor all classification quantity, SkShow k-th attribute value number in training sample.
Beneficial effects of the present invention are：
(1) present invention helps lend some impetus to differentiated service, improves service satisfaction.It is complete in user's portrait label Establishing
Into after, each user can have oneself distinguished tag library, when client provides the bases such as the name of oneself to worker at the production line
After this information, the tag library of the client is just appeared in face of employee, that is, all letters of the user in power system
Breath includes history paying behaviors, with electrographic recording, various demands record, easily communication etc. all can show.This will give
Worker at the production line such as contact staff, business hall person etc. are very helpful in service, and they are understood according to the information in tag library,
Adjustment attitude and strategy, as much as possible go as customer service, the satisfaction of raising client；
(2) present invention can lift marketing success rate, based on user's portrait tag library, which can easily filter out
Which kind of product client is adapted to, and marketing according to different channels to some tagging users of can orienting：Such as, Jing often beats
The user of 95598 customer service hot lines, may more favor in receive make a phone call, marketing of sending short messages；And other Jing often palm electric power,
The client for paid the fees in wechat public number, inquiring about, the mode that may prefer to receive some APP push, micro-signal is pushed is entered
Field headquarters is sold, and the product being adapted to lead referral by the more receptible mode of client, the marketing mode of precision more can be improved
Marketing success rate；
(3) present invention contributes to prediction customer action, reduces enterprises' loss, because information is not got through between all departments,
First-line staff can not in time grasp the bad behavior of client, and such as stealing, arrearage, promise breaking electricity consumption causes very big to enterprise every year
Loss.By user's portrait tag library, first-line staff can pinpoint the problems in time, and to those clients for having " mislead " electricity is taken
The marketing strategy that expense is stopped loss, improves tariff recovery efficiency and effect；
(4) present invention contributes to carrying out user credit grading, there is provided marketing service data supporting, power consumer credit appraisal
The foundation of system, contribute to electric company it is objective, comprehensively and accurately from user basic information, electricity consumption behavior, paying behaviors, tell
Ask behavior to carry out comprehensive assessment to the credit situation of electric service object (family and people), be that provincial company formulates this province user's differentiation
Service Management strategy provided auxiliary decision-making, for 95598 service handling processes at different levels directiveness reference is provided；It is user according to difference
Credit grade enjoy different electric services traffic criterias be provided, provides data for all kinds of electricity consumption APP application services and props up
Support.Meanwhile, it is that next step formulates differentiation credit scoring model for the unit-economy development of different districts and cities and electric power consumption level
System lays the first stone.
Description of the drawings
Fig. 1 is that user's portrait Establishing process diagram of the present invention is intended to；
Fig. 2 is that the power consumer behavior label of the present invention constitutes exemplary plot.
Specific embodiment：
Below in conjunction with the accompanying drawings the invention will be further described with embodiment.
Power consumer portrait based on big data technology is set up and analysis method, comprises the following steps
Step 1, determine user draw a portrait class categories C={ C1,C2,…,Ci, and the influence factor collection of classification results
A={ A1,A2,A3,A4,…,An, determine the mapping relations of two set；
Step 2, collection initial data, used as training sample, remaining 80% data is used as forecast sample for the data using 20%；
Step 3, initial data is pre-processed, including normalized, sliding-model control and attribute reduction,
So that it is determined that influence factor set A={ A1,A2,…,Am, wherein m≤n；
Step 4, training sample is trained, and using ten folding cross validations as test pattern, is established based on simplicity
The power consumer portrait forecast model of Bayes classifier, carries out checking accuracy, so that it is guaranteed that model to institute's established model then
Validity；
Step 5, forecast model of being drawn a portrait using the power consumer based on Naive Bayes Classifier, to forecast sample line number is entered
Analyze according to classified excavation, the power consumer portrait high so as to obtain accuracy.
The step 1 is comprised the following steps that：
Step 1.1, consider after the factors such as 95598 data, electrical network sales department data and power consumer classification, listening
On the basis of having taken electric industry industry technology professional and basic unit's attendant's suggestion, it is determined that user's portrait class categories, i.e.,：
High-quality user, general user and low quality user；
Step 1.2, with reference to Base data platform part valid data and 186 systems (front end) data, through drawing to user
Picture talk out with after analysis, it is determined that user draw a portrait classification results influence factor, i.e.,：Essential information, electricity consumption behavior, pays
Charge information, demand information and social information；
Step 1.3, determine influence factor collection to user draw a portrait classification mapping relations.
The step 2 is comprised the following steps that：
Step 2.1, collect from 95598 data and Base data platform required data, using 20% data as
Training sample, remaining 80% data is used as forecast sample.
The step 3 is comprised the following steps that：
Step 3.1, for the completeness and efficiency of retention data, it would be desirable to place is normalized to sample data
Reason, normalizing formula is：
In formula：xijIt is to normalize front sample, sijIt is sample after normalization；min(xj) it is minimum of a value in original sample；
max(xj) it is maximum in original sample.
Step 3.2, for the discrete data values of higher abstraction hierarchy, training sample data are carried out with sliding-model control, it is public
Formula is as follows：
In formula：zijFor sample after discretization, min (sj) be normalization after sample minimum of a value, max (sj) for normalization after
The maximum of sample, Q is step-length：
Step 3.3, after certain attribute is removed, if not finding the training sample data for having repetition, i.e. ind (C-C1)≠
Ind (C), the relation that can not debate in decision table there occurs corresponding change, so this attribute is retained；By that analogy, most
Influence factor set A={ A are obtained eventually1,A2,…,Am}。
The step 4 is comprised the following steps that：
Step 4.1, establish based on Naive Bayes Classifier power consumer draw a portrait forecast model, the model is with A=
{A1,A2,A3,A4,A5}={ essential information, electricity consumption behavior, payment information, demand information, social information } it is input vector, with C
={ C1,C2,C3}={ high-quality user, general user, low quality user } it is classification output vector；
Step 4.2, respectively in terms of detailed precision, confusion matrix and node error rate three, inspection power consumer portrait
The accuracy of forecast model, wherein detailed precision includes：TP Rate (real unit ratio), FP Rate (false positive element ratio),
Precision (precision), Recall (recall ratio), F-Measure (harmonic-mean of precision ratio and recall ratio) and ROC
Area。
The step 5 is comprised the following steps that：
Step 5.1, number S for counting example training sample, classification are CiSample number Si, k-th attribute AkEqual to xk
And classification is CiTraining example number of samples Sik；
Step 5.2, calculate respectively
Here, ScFor all classification quantity, SkShow k-th attribute value number in training sample.
Step 5.3, utilize grader
Draw the ownership classification results of prediction example sample X.
Step 5.4, to predict the outcome and do comparative analysis with actual conditions what user drew a portrait, excavate deeper data
Value.
The present invention is divided to personal portrait and colony's two aspects of portrait to do customer analysis according to user data and business demand.
Individual's portrait be according to the label in user tag storehouse, to each client in the light of actual conditions stick he her exclusive label.
Colony's portrait is, by known part labels, the personal portrait for meeting selected label simultaneously to be filtered out from custom system, this
A few people's portraits just constitute colony's portrait.Individual's portrait, supports that people associated with it excavates with family portrait, is easy to a line to market
Personnel or contact staff quickly understand user's feature, evade potential risks, save marketing service cost, improve user service
Satisfaction.Colony draw a portrait, can analyze same subscriber group different geographical, different times its portrait constituent difference, be easy to adopt
Take personal marketing strategy and assessment marketing effectiveness.
User's portrait is related to the latitude of data needs business scenario to combine, should simply it is capable and experienced again with business strong correlation,
Convenient facilitating again should be screened further to operate.User's portrait needs to adhere to three principles, be respectively base attribute and electricity consumption,
Based on payment, demand information；Based on strong correlation information；Based on qualitative data.Just launch respectively below to explain and analyze.
(1) base attribute and electricity consumption, payment, based on demand information
The information of one power consumer of description is a lot, and base attribute is information important in user's portrait, and base attribute is
Consuming capacity information of one people of description in society.It is to find target customer that any enterprise carries out the purpose of user's portrait, its
Must be the user with potential consumption ability.Partial key information in base attribute can directly prove the consumption energy of client
Region, institute's work taken up that power, such as power consumer are lived, and the letter such as income, the house property, the contract capacity that are possessed
Breath.Certainly, the name of user, sex, electricity consumption address, contact method etc. are also what is needed, can contact client with power grid enterprises, will
Products ＆ services are promoted to client.
In addition, except user's base attribute, in addition it is also necessary to understand consumption feelings of the user in electricity commodity process of consumption
Condition (power information), payout status (payment information), consumption feedback (demand information), and consumer in future is each other
Interactive communication situation (social information).
(2) using strong correlation information, weak relevant information is ignored
Strong correlation information is exactly the directly related information of same power marketing business scenario demand, and it can be cause and effect information,
Can also be the very high information of degree of correlation.
If defining the change as coefficient correlation span using 0 to 1, more than 0.6 coefficient correlation just should be defined
For strong correlation information.For example under the premise of other conditions identical, the average salary of one's mid-30s people is 30 higher than average age
The people in year, student's average salary of computer major graduation is higher than philosophy Major, is engaged in the average work of financial industry work
Money exceedes Hainan Province's average salary higher than the average salary for being engaged in textile industry, the average salary in Shanghai.Can be with from these information
Find out that the impact of the age, educational background, occupation, place of messenger to income is larger, be strong correlation relation with income height.For example, it is right
The larger information of electricity consumption, payment, demand behavioral implications is exactly strong correlation information, otherwise is then weak relevant information.
The information such as user's others information, height, body weight, name, the constellation of such as user, it is difficult to analyze from probability
Its custom on electricity consumption, payment, demand affects, and these are weak relevant informations, and these information just should not be put in user's portrait
It is analyzed, not with larger commercial value.
When user draws a portrait with customer analysis, it is required to consider strong correlation information, should not consider weak relevant information, this is user
One principle of portrait.
(3) it is qualitatively information by quantitative information categorization
The purpose of user's portrait is to filter out target customer for electric power marketing strategy, and quantitative information is unfavorable for entering client
Row screening, needs for quantitative information to be converted into qualitative information, and by information category crowd is screened.
For example age bracket can be divided to client, be defined as within -25 years old 18 years old young man, be defined within -35 years old 25 years old
For the young and the middle aged, 36-45 is defined as a middle-aged person etc..Personal income information is may be referred to, crowd is defined as into booming income crowd, it is medium
Income crowd, low-income groups.Can also be high, medium and low rank by client definition with reference to assets information.The classification of qualitative information
And methods, power grid enterprises can be from own service, without fixed pattern.
By all kinds of quantitative informations in power marketing business, concentrate in together, qualitative information is classified, and carry out qualitative
Change, favorably with user is screened, quickly positioning target client, be user portrait another principle.
User's portrait step
Can be divided into three steps from flow process for user's portrait of power industry：Obtain and study user profile, build
Vertical user behavior tag library, development user's portrait (as shown in Figure 1), concrete grammar is as follows：
(1) obtain and study user profile
Power consumer representation data is broadly divided into four classes, base attribute, power information, payment information, demand information.These
Data are all distributed in different information system shelves, such as user's base attribute, payment information in sales service system, user
In power information acquisition system, user's demand information is in 95598 operational support systems for power information.
The latitude information of user's portrait is not The more the better, it is only necessary to found and four big class portrait information strong correlation information,
With business scenario strong correlation information, with product and target customer's strong correlation information.The selecting factors of strong correlation are proposed with
Expert graded reduces the scope, and correlation analysis are carried out with portrait index again according to the key factor that expert estimation is selected, and finds out
Really there is the user behavior factor of strong correlation in business, the data source scope of label is determined based on this.By substantial amounts of reality
Proof is trampled, 360 degree of description user portrait relatively difficult to achieve in a short time for a business, it is also not possible to complete by portrait
Understand client, but can accomplish to approach understanding to user.In addition, the actual effect of data also wants emphasis to consider, for data matter
Amount (accuracy, promptness, integrality) not high factor avoids bringing in label system, in order to avoid affect end user's portrait
Accuracy.
(2) user behavior tag library is set up
Label, is the highly refined signature identification by drawing to user profile analysis, finally by all marks of user
Sign in general, it is possible to sketch the contours of the solid " portrait " of the user.The core work for building user's portrait is pasted to user
" label ", and be partly direct according to the behavioral data of user (time limit of registering for a household residence card, charge level, payment channel etc.) in label
Arrive, obtained (demand tolerance, demand preference etc.) by series of algorithms or rule digging.
Essential information label is some essential characteristics for describing user, can go to recognize client by understanding essential information,
Know that whom client is, including name, sex, age, income, occupation, all kinds of social relationships etc.；Electricity consumption behavior label main presentation
User power utilization feature, understands user using the custom and characteristic during electricity consumption, including power consumption, season peak of power consumption,
Electricity consumption festivals or holidays, power supply quality perception etc..Distribution of the paying behaviors essential record user during electricity consumption expense is paid
And behavioral characteristic, including bill date of issue, payment expiration date, consumption grade, payment channel preference, payment promptness etc..Demand
Behavior emphasis reflects various demands of the power consumer during electrical power services are enjoyed, and further sees clearly user and power supply enterprise is taken
Business quality, the feedback and recommendations of efficiency, promote power industry in operational improvement, mainly include demand mood, demand
Tolerance, demand intensity, demand content-preference etc..
There is to a certain extent limitation in the data of power system, iff by power system internal data use is done
Draw a portrait at family, then figure painting picture out just less enriches.If possessing the data of some economic aspects of user, such as duty
Industry, income, consuming capacity, home background etc., it is possible to make user's portrait information more rich, full.These data are accomplished by beating
Logical external resource, introduces external data, such as introduce the information of Unionpay and electric business to enrich consumption feature information, introduces mobile big
The positional information of data introduces the data of outside firm to enrich social information etc. enriching the hobby information of client.
Realization aspect, the combing of data label derives from the diary record system of the daily accumulation of each system, by Sqoop
Import HDFS, it is also possible to realized with code, such as the JDBC connections traditional database of Spark carries out the Cache of data.Also
A kind of mode, can be by writing data into local file, then by Export of Load or Hive of SparkSQL etc.
Mode imports HDFS.
UDF or HiveQL are write by Hive ETL is spliced according to service logic, mark users different in user's correspondence
Data are signed, corresponding source table data are generated, so as to the data acquisition of following needs user portrait system, is entered by different rules
The generation of row label width table.
(3) user's portrait is carried out
The purpose of user's portrait is to be analyzed to the behavior of user, being provided preferably for customer according to analysis result
Service.Quantitative information is unfavorable for being analyzed client, needs for quantitative information to be converted into qualitative information, by information category
To be analyzed to different customers.
By user tag system, the individual portrait of user and colony's portrait can be carried out.By marketing personnel's input user only
One identity identification information, can from some independent user carry out covering its base attribute, with electrical feature, payment feature, tell
Seek the individual portrait of feature.Instruct a line marketing personnel field operation to take individuation, differentiated service strategy when servicing, drop
Low individual services risk range and degree, improve user service satisfaction rate.In addition selected user basic information, used by marketing personnel
Electric behavior, paying behaviors, the different label characteristics of demand behavior, around selected label, then carry out user group's portrait, weight
Point shows the composition situation of its sub- level label characteristics to this crowd of user portrait, it is intended to observe same subscriber group in different geographical, no
Same time its portrait constituent difference, contribute to the lateral comparison of each power supply unit customer group, or same power supply unit user
Group's analysis of trend, and then take differentiated marketing strategies and assessment marketing effectiveness.
It is real to play based on the power consumer of big data technology in order to above-mentioned achievement is fully combined with actual sales service
The function and significance of portrait work, should carry out and user power utilization, payment, demand with reference to the major tasks in power grid enterprises' each year
Related typical application scenarios are excavated and marketing strategy is formulated, and are further carried out according to user power utilization behavior, paying behaviors, demand
The precision marketing that behavior is carried out, plays the value of data productivity.
Although the above-mentioned accompanying drawing that combines is described to the specific embodiment of the present invention, not to present invention protection model
The restriction enclosed, one of ordinary skill in the art should be understood that on the basis of technical scheme those skilled in the art are not
Need the various modifications made by paying creative work or deformation still within protection scope of the present invention.
Claims (10)
1. a kind of power consumer portrait based on big data technology is set up and analysis method, it is characterized in that：Comprise the following steps：
(1) history power information, base attribute, payment information and the demand information of power consumer are obtained；
(2) determine the set of user's portrait class categories, and the influence factor collection of classification results, determine influence factor collection and divide
The mapping relations of class set；
(3) data of acquisition are randomly selected, used as training sample, it is with data as forecast sample for a part；
(4) training sample and forecast sample are normalized, sliding-model control and attribute reduction, it is determined that after correction
Influence factor set；
(5) training sample is trained, and using ten folding cross validations as test pattern, sets up and be based on Naive Bayes Classification
The power consumer portrait forecast model of device, using forecast model data classified excavation analysis is carried out to forecast sample, obtains electric power
User draws a portrait.
2. a kind of power consumer portrait based on big data technology as claimed in claim 1 is set up and analysis method, its feature
It is：In the step (1), the influence factor includes essential information, electricity consumption behavior, payment information, demand information and social letter
Breath, user portrait class categories include high-quality user, general user and low quality user, and concrete credit rating is according to setting
Fixed standard is divided.
3. a kind of power consumer portrait based on big data technology as claimed in claim 1 is set up and analysis method, its feature
It is：In the step (3), used as training sample, remaining 80% data is used as forecast sample for the data using 20%.
4. a kind of power consumer portrait based on big data technology as claimed in claim 1 is set up and analysis method, its feature
It is：In the step (4), data are normalized：
In formula：xijIt is to normalize front sample, sijIt is sample after normalization；min(xj) it is minimum of a value in original sample；max
(xj) it is maximum in original sample.
5. a kind of power consumer portrait based on big data technology as claimed in claim 1 is set up and analysis method, its feature
It is：In the step (4), sliding-model control is carried out to training sample data：
In formula：zijFor sample after discretization, min (sj) be normalization after sample minimum of a value, max (sj) it is sample after normalization
Maximum, Q is step-length：
6. a kind of power consumer portrait based on big data technology as claimed in claim 1 is set up and analysis method, its feature
It is：In the step (4), the concrete steps of attribute reduction include：After certain attribute is removed, if not finding to have the instruction of repetition
Practice sample data, the relation that can not debate in decision table there occurs corresponding change, so this attribute is retained；With such
Push away, finally give the influence factor set of determination.
7. a kind of power consumer portrait based on big data technology as claimed in claim 1 is set up and analysis method, its feature
It is：In the step (5), the concrete grammar for setting up power consumer portrait forecast model is：
(5-1) set up the power consumer portrait forecast model based on Naive Bayes Classifier, the model with the impact that determines because
Element collection is combined into input vector, and with user's portrait class categories output vector is combined into；
(5-2) respectively in terms of detailed precision, confusion matrix and node error rate three, inspection power consumer portrait forecast model
Accuracy.
8. a kind of power consumer portrait based on big data technology as claimed in claim 7 is set up and analysis method, its feature
It is：In the step (5-2), detailed precision includes：Real unit's ratio, false positive element ratio, precision, recall ratio, precision ratio and
The harmonic-mean of recall ratio.
9. a kind of power consumer portrait based on big data technology as claimed in claim 1 is set up and analysis method, its feature
It is：In the step (5), the concrete grammar of data classified excavation analysis is carried out to forecast sample using forecast model：
(5-a) it is C to count number S of example training sample, classificationiSample number Si, k-th attribute AkEqual to xkAnd classification is
CiTraining example number of samples Sik；
(5-b) probability of all categories is calculated respectively with middle attribute A of all categorieskEqual to xkProbability；
(5-c) grader is utilized
The ownership classification results of prediction example sample X are drawn, predicting the outcome for user's portrait is contrasted with actual conditions.
10. a kind of power consumer portrait based on big data technology as claimed in claim 9 is set up and analysis method, its feature
It is：In the step (5-b),
Here, ScFor all classification quantity, SkShow k-th attribute value number in training sample.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610860951.5A CN106651424B (en) | 2016-09-28 | 2016-09-28 | Power user portrait establishing and analyzing method based on big data technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610860951.5A CN106651424B (en) | 2016-09-28 | 2016-09-28 | Power user portrait establishing and analyzing method based on big data technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106651424A true CN106651424A (en) | 2017-05-10 |
| CN106651424B CN106651424B (en) | 2020-05-22 |
Family
ID=58854665
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610860951.5A Active CN106651424B (en) | 2016-09-28 | 2016-09-28 | Power user portrait establishing and analyzing method based on big data technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106651424B (en) |
Cited By (73)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107292513A (en) * | 2017-06-21 | 2017-10-24 | 国网辽宁省电力有限公司 | A kind of method that power customer management is realized based on svm classifier algorithm |
| CN107391645A (en) * | 2017-07-12 | 2017-11-24 | 广州市昊链信息科技股份有限公司 | A kind of logistics information automatic push and practical operation specification form system and method |
| CN107464142A (en) * | 2017-08-09 | 2017-12-12 | 南京甄视智能科技有限公司 | Method, the user's portrait system for system of being drawn a portrait based on marketing feedback result back feeding user |
| CN107578277A (en) * | 2017-08-24 | 2018-01-12 | 国网浙江省电力公司电力科学研究院 | Rental housing client's localization method for power marketing |
| CN107730269A (en) * | 2017-07-21 | 2018-02-23 | 南通大学 | A kind of Electricity customers portrait method of Behavior-based control analysis |
| CN107895015A (en) * | 2017-11-14 | 2018-04-10 | 江苏电力信息技术有限公司 | Power consumer feature tag proportional system construction method based on analytic hierarchy process (AHP) |
| CN107909288A (en) * | 2017-12-05 | 2018-04-13 | 国网内蒙古东部电力有限公司电力科学研究院 | Paying behaviors analysis method based on SOM neural network clustering algorithms |
| CN108038727A (en) * | 2017-12-11 | 2018-05-15 | 囯网河北省电力有限公司电力科学研究院 | A kind of method, system and terminal device that differentiated service is provided |
| CN108399249A (en) * | 2018-03-02 | 2018-08-14 | 上海精数信息科技有限公司 | Data normalization method, user's portrait providing method, equipment and storage medium |
| CN108520045A (en) * | 2018-04-03 | 2018-09-11 | 平安健康保险股份有限公司 | The service response method and device of data |
| CN108550050A (en) * | 2018-03-27 | 2018-09-18 | 昆山和君纵达数据科技有限公司 | A kind of user's portrait method based on call center data |
| CN108647729A (en) * | 2018-05-11 | 2018-10-12 | 广东省科技基础条件平台中心 | A kind of user's portrait acquisition methods |
| CN108764663A (en) * | 2018-05-15 | 2018-11-06 | 广东电网有限责任公司信息中心 | A kind of power customer portrait generates the method and system of management |
| CN109003127A (en) * | 2018-07-06 | 2018-12-14 | 国网福建省电力有限公司 | A kind of credit rating method based on Electricity customers data |
| CN109063945A (en) * | 2018-06-08 | 2018-12-21 | 广州亦云信息技术股份有限公司 | A kind of 360 degree of customer portrait construction methods of sale of electricity company based on Value accounting system |
| CN109146539A (en) * | 2018-06-28 | 2019-01-04 | 深圳市彬讯科技有限公司 | The update method and device of user's portrait |
| CN109146707A (en) * | 2018-08-27 | 2019-01-04 | 罗孚电气（厦门）有限公司 | Power consumer analysis method, device and electronic equipment based on big data analysis |
| CN109389138A (en) * | 2017-08-09 | 2019-02-26 | 武汉安天信息技术有限责任公司 | A kind of user's portrait method and device |
| CN109412832A (en) * | 2018-09-10 | 2019-03-01 | 中国联合网络通信集团有限公司 | User service providing method and system |
| CN109559146A (en) * | 2018-09-25 | 2019-04-02 | 国家电网有限公司客户服务中心 | Electricity customer service center accesses data center's optimization method based on the provinces and cities of logistic model prediction potential user's quantity |
| CN109598479A (en) * | 2018-10-25 | 2019-04-09 | 北京奇虎科技有限公司 | A kind of bill extracting method, device, electronic equipment and medium |
| CN109615555A (en) * | 2018-08-01 | 2019-04-12 | 美林数据技术股份有限公司 | User's exception electricity consumption Activity recognition method and system based on Ensemble Learning Algorithms |
| CN109614982A (en) * | 2018-10-18 | 2019-04-12 | 平安科技（深圳）有限公司 | Product analysis method, apparatus, computer equipment and storage medium |
| CN109636235A (en) * | 2018-12-26 | 2019-04-16 | 北京汽车研究总院有限公司 | The determination method and processing system of driving behavior portrait model |
| CN109697626A (en) * | 2018-10-19 | 2019-04-30 | 上海吉贝克信息技术有限公司 | Marketing system, method, electric terminal and medium based on JeeSite quick development platform |
| CN109740620A (en) * | 2018-11-12 | 2019-05-10 | 平安科技（深圳）有限公司 | Method for building up, device, equipment and the storage medium of crowd portrayal disaggregated model |
| CN109785034A (en) * | 2018-11-13 | 2019-05-21 | 北京码牛科技有限公司 | User's portrait generation method, device, electronic equipment and computer-readable medium |
| CN109934273A (en) * | 2019-03-01 | 2019-06-25 | 长沙理工大学 | It is a kind of based on the fault characteristic of DML-KNN algorithm and active damage repair technology draw a portrait new method |
| CN109948913A (en) * | 2019-03-01 | 2019-06-28 | 长沙理工大学 | A kind of multi-source feature power consumer composite portrait system based on double-deck xgboost algorithm |
| CN110046657A (en) * | 2019-03-29 | 2019-07-23 | 武汉大学深圳研究院 | A kind of social safety figure painting image space method based on multiple view study |
| CN110119998A (en) * | 2019-03-27 | 2019-08-13 | 广东浩迪创新科技有限公司 | Rental house approaches to IM and system based on intelligent electric meter |
| CN110134860A (en) * | 2019-04-12 | 2019-08-16 | 阿里巴巴集团控股有限公司 | User's portrait generation method, device and equipment |
| CN110191183A (en) * | 2019-05-31 | 2019-08-30 | 深圳市医信科技有限公司 | Accurate intelligent method for pushing, system, device and computer readable storage medium |
| CN110210018A (en) * | 2019-05-14 | 2019-09-06 | 北京百度网讯科技有限公司 | It registers the matching process and device of department |
| CN110263255A (en) * | 2019-06-21 | 2019-09-20 | 北京字节跳动网络技术有限公司 | Acquisition methods, system, server and the storage medium of customer attribute information |
| CN110276647A (en) * | 2019-06-26 | 2019-09-24 | 中国南方电网有限责任公司 | Online service method based on user integral |
| CN110309932A (en) * | 2018-03-21 | 2019-10-08 | 国网浙江省电力公司湖州供电公司 | A kind of power customer resource fine-grained management method based on TEM |
| CN110334463A (en) * | 2019-07-12 | 2019-10-15 | 无锡锐泰节能系统科学有限公司 | Based on the energy management design method with the model that can draw a portrait |
| CN110347923A (en) * | 2019-07-09 | 2019-10-18 | 葛晓滨 | A kind of method for the fast-neutron fission formula building user's portrait recalled |
| CN110472852A (en) * | 2019-08-02 | 2019-11-19 | 上海云扩信息科技有限公司 | A kind of experience assessment implementation management method of electrical power services application |
| CN110516155A (en) * | 2019-08-29 | 2019-11-29 | 深圳市云积分科技有限公司 | Marketing strategy generation method and system |
| CN110599237A (en) * | 2019-08-15 | 2019-12-20 | 国网山东省电力公司临清市供电公司 | Electric power marketing management information acquisition module based on intelligent monitoring |
| CN110689385A (en) * | 2019-10-16 | 2020-01-14 | 国网山东省电力公司信息通信公司 | Power customer service user portrait construction method based on knowledge graph |
| CN110688580A (en) * | 2019-10-14 | 2020-01-14 | 国网山东省电力公司泰安供电公司 | Power supply service data display system based on customer portrait |
| CN110728537A (en) * | 2019-09-24 | 2020-01-24 | 国网河北省电力有限公司信息通信分公司 | Prediction payment method based on power consumer behavior label |
| CN111080367A (en) * | 2019-12-20 | 2020-04-28 | 深圳市国电科技通信有限公司 | Power consumption behavior analysis method for sensing power consumption state of low-voltage user |
| CN111221868A (en) * | 2018-11-26 | 2020-06-02 | 国网上海市电力公司 | Data mining and analyzing method applied to channel preference of power customer |
| CN111401798A (en) * | 2020-06-02 | 2020-07-10 | 南京百敖软件有限公司 | Enterprise waste escaping and debt risk early warning system and construction method |
| CN111506636A (en) * | 2020-05-12 | 2020-08-07 | 上海积成能源科技有限公司 | System and method for analyzing residential electricity consumption behavior based on autoregressive and neighbor algorithm |
| CN111666313A (en) * | 2020-05-25 | 2020-09-15 | 中科星图股份有限公司 | Correlation construction and multi-user data matching method based on multi-source heterogeneous remote sensing data |
| CN111709791A (en) * | 2020-06-19 | 2020-09-25 | 四川中电启明星信息技术有限公司 | Power supply marketing service method based on improved feature word weight algorithm |
| CN111723273A (en) * | 2019-03-18 | 2020-09-29 | 北京中电翔云信息技术有限公司 | Smart cloud retrieval system and method |
| CN111738462A (en) * | 2020-06-08 | 2020-10-02 | 国网江苏省电力有限公司常州供电分公司 | Fault first-aid repair active service early warning method for electric power metering device |
| CN112053100A (en) * | 2020-10-15 | 2020-12-08 | 东北电力大学 | Big data-based analysis method for electricity utilization behavior of user |
| CN112115355A (en) * | 2020-09-02 | 2020-12-22 | 南方电网深圳数字电网研究院有限公司 | Power grid intelligent service system based on user portrait in power industry |
| CN112131475A (en) * | 2020-09-25 | 2020-12-25 | 重庆邮电大学 | Interpretable and interactive user portrait method and device |
| CN112132631A (en) * | 2020-09-29 | 2020-12-25 | 国网上海市电力公司 | Label electricity utilization management method based on power customer portrait |
| CN112184313A (en) * | 2020-09-28 | 2021-01-05 | 国网上海市电力公司 | Power marketing business application system based on client portrait data |
| CN112417308A (en) * | 2020-12-17 | 2021-02-26 | 国网河北省电力有限公司营销服务中心 | User portrait label generation method based on electric power big data |
| CN112529630A (en) * | 2020-12-17 | 2021-03-19 | 国网河北省电力有限公司营销服务中心 | Construction method of multi-dimensional informatization label system of power consumer |
| CN112579721A (en) * | 2020-12-22 | 2021-03-30 | 北京市腾河智慧能源科技有限公司 | Method and system for constructing crowd distribution map, terminal device and storage medium |
| CN112862530A (en) * | 2021-02-09 | 2021-05-28 | 江之龙 | Marketing system based on big data |
| CN113159419A (en) * | 2021-04-21 | 2021-07-23 | 成都卫士通信息产业股份有限公司 | Group feature portrait analysis method, device and equipment and readable storage medium |
| CN113313470A (en) * | 2021-06-10 | 2021-08-27 | 郑州科技学院 | Employment type evaluation method and system based on big data |
| CN113344638A (en) * | 2021-06-29 | 2021-09-03 | 云南电网有限责任公司信息中心 | Hypergraph-based power grid user group portrait construction method and device |
| CN113377760A (en) * | 2021-07-06 | 2021-09-10 | 国网江苏省电力有限公司营销服务中心 | Method and system for establishing low-voltage resident feature portrait based on electric power data and multivariate data |
| CN113487448A (en) * | 2021-05-31 | 2021-10-08 | 国网上海市电力公司 | Power credit labeling method and system based on power big data |
| CN113554241A (en) * | 2021-09-02 | 2021-10-26 | 国网山东省电力公司泰安供电公司 | User layering method and prediction method based on user electricity complaint behaviors |
| CN114219241A (en) * | 2021-12-01 | 2022-03-22 | 深圳供电局有限公司 | Customer electricity consumption behavior analysis method and system |
| CN114417955A (en) * | 2021-12-02 | 2022-04-29 | 国网浙江省电力有限公司 | Power index portrait real-time generation method and device suitable for villages and storage medium |
| CN114595323A (en) * | 2022-03-04 | 2022-06-07 | 北京百度网讯科技有限公司 | Portrait construction, recommendation, model training method, apparatus, device and storage medium |
| CN116450952A (en) * | 2023-06-16 | 2023-07-18 | 天津星耀九洲科技有限公司 | Internet user portrait generation method and system based on deep learning technology |
| CN117271905A (en) * | 2023-11-21 | 2023-12-22 | 杭州小策科技有限公司 | Crowd image-based lateral demand analysis method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103325067A (en) * | 2013-05-29 | 2013-09-25 | 广东电网公司佛山供电局 | Service recommendation method and system based on electricity customer segmentation |
| CN104408667A (en) * | 2014-11-20 | 2015-03-11 | 深圳供电局有限公司 | Comprehensive assessment method and system of power quality |
| CN104809473A (en) * | 2015-04-30 | 2015-07-29 | 国网智能电网研究院 | Online collected electric power data classifying method |
| CN105023092A (en) * | 2015-06-24 | 2015-11-04 | 国网山东省电力公司电力科学研究院 | Power load monitoring method based on electric quantity feature analysis |
| CN105260815A (en) * | 2015-09-18 | 2016-01-20 | 国网辽宁省电力有限公司鞍山供电公司 | Electric power system payment channel evaluating method based on large data |
-
2016
- 2016-09-28 CN CN201610860951.5A patent/CN106651424B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103325067A (en) * | 2013-05-29 | 2013-09-25 | 广东电网公司佛山供电局 | Service recommendation method and system based on electricity customer segmentation |
| CN104408667A (en) * | 2014-11-20 | 2015-03-11 | 深圳供电局有限公司 | Comprehensive assessment method and system of power quality |
| CN104809473A (en) * | 2015-04-30 | 2015-07-29 | 国网智能电网研究院 | Online collected electric power data classifying method |
| CN105023092A (en) * | 2015-06-24 | 2015-11-04 | 国网山东省电力公司电力科学研究院 | Power load monitoring method based on electric quantity feature analysis |
| CN105260815A (en) * | 2015-09-18 | 2016-01-20 | 国网辽宁省电力有限公司鞍山供电公司 | Electric power system payment channel evaluating method based on large data |
Cited By (102)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107292513A (en) * | 2017-06-21 | 2017-10-24 | 国网辽宁省电力有限公司 | A kind of method that power customer management is realized based on svm classifier algorithm |
| CN107391645A (en) * | 2017-07-12 | 2017-11-24 | 广州市昊链信息科技股份有限公司 | A kind of logistics information automatic push and practical operation specification form system and method |
| CN107391645B (en) * | 2017-07-12 | 2018-04-10 | 广州市昊链信息科技股份有限公司 | A kind of logistics information automatic push and practical operation specification form system and method |
| CN107730269A (en) * | 2017-07-21 | 2018-02-23 | 南通大学 | A kind of Electricity customers portrait method of Behavior-based control analysis |
| CN107464142B (en) * | 2017-08-09 | 2018-11-09 | 南京甄视智能科技有限公司 | Method, the user's portrait system for system of being drawn a portrait based on marketing feedback result back feeding user |
| CN107464142A (en) * | 2017-08-09 | 2017-12-12 | 南京甄视智能科技有限公司 | Method, the user's portrait system for system of being drawn a portrait based on marketing feedback result back feeding user |
| CN109389138A (en) * | 2017-08-09 | 2019-02-26 | 武汉安天信息技术有限责任公司 | A kind of user's portrait method and device |
| CN107578277A (en) * | 2017-08-24 | 2018-01-12 | 国网浙江省电力公司电力科学研究院 | Rental housing client's localization method for power marketing |
| CN107578277B (en) * | 2017-08-24 | 2021-04-09 | 国网浙江省电力公司电力科学研究院 | Rental house client positioning method for electric power marketing |
| CN107895015A (en) * | 2017-11-14 | 2018-04-10 | 江苏电力信息技术有限公司 | Power consumer feature tag proportional system construction method based on analytic hierarchy process (AHP) |
| CN107909288A (en) * | 2017-12-05 | 2018-04-13 | 国网内蒙古东部电力有限公司电力科学研究院 | Paying behaviors analysis method based on SOM neural network clustering algorithms |
| CN108038727A (en) * | 2017-12-11 | 2018-05-15 | 囯网河北省电力有限公司电力科学研究院 | A kind of method, system and terminal device that differentiated service is provided |
| CN108399249B (en) * | 2018-03-02 | 2020-12-29 | 上海精数信息科技有限公司 | Data normalization method, user portrait providing device and storage medium |
| CN108399249A (en) * | 2018-03-02 | 2018-08-14 | 上海精数信息科技有限公司 | Data normalization method, user's portrait providing method, equipment and storage medium |
| CN110309932A (en) * | 2018-03-21 | 2019-10-08 | 国网浙江省电力公司湖州供电公司 | A kind of power customer resource fine-grained management method based on TEM |
| CN108550050A (en) * | 2018-03-27 | 2018-09-18 | 昆山和君纵达数据科技有限公司 | A kind of user's portrait method based on call center data |
| CN108520045A (en) * | 2018-04-03 | 2018-09-11 | 平安健康保险股份有限公司 | The service response method and device of data |
| CN108520045B (en) * | 2018-04-03 | 2023-05-30 | 平安健康保险股份有限公司 | Data service response method and device |
| CN108647729A (en) * | 2018-05-11 | 2018-10-12 | 广东省科技基础条件平台中心 | A kind of user's portrait acquisition methods |
| CN108647729B (en) * | 2018-05-11 | 2022-03-04 | 广东省科技基础条件平台中心 | User portrait acquisition method |
| CN108764663A (en) * | 2018-05-15 | 2018-11-06 | 广东电网有限责任公司信息中心 | A kind of power customer portrait generates the method and system of management |
| CN108764663B (en) * | 2018-05-15 | 2020-10-16 | 广东电网有限责任公司信息中心 | Method and system for generating and managing power customer portrait |
| CN109063945A (en) * | 2018-06-08 | 2018-12-21 | 广州亦云信息技术股份有限公司 | A kind of 360 degree of customer portrait construction methods of sale of electricity company based on Value accounting system |
| CN109063945B (en) * | 2018-06-08 | 2023-12-29 | 广州云标局网络科技有限公司 | Value evaluation system-based 360-degree customer portrait construction method for electricity selling company |
| CN109146539A (en) * | 2018-06-28 | 2019-01-04 | 深圳市彬讯科技有限公司 | The update method and device of user's portrait |
| CN109003127A (en) * | 2018-07-06 | 2018-12-14 | 国网福建省电力有限公司 | A kind of credit rating method based on Electricity customers data |
| CN109615555A (en) * | 2018-08-01 | 2019-04-12 | 美林数据技术股份有限公司 | User's exception electricity consumption Activity recognition method and system based on Ensemble Learning Algorithms |
| CN109146707A (en) * | 2018-08-27 | 2019-01-04 | 罗孚电气（厦门）有限公司 | Power consumer analysis method, device and electronic equipment based on big data analysis |
| CN109412832B (en) * | 2018-09-10 | 2022-04-15 | 中国联合网络通信集团有限公司 | User service providing method and system |
| CN109412832A (en) * | 2018-09-10 | 2019-03-01 | 中国联合网络通信集团有限公司 | User service providing method and system |
| CN109559146B (en) * | 2018-09-25 | 2022-11-04 | 国家电网有限公司客户服务中心 | Provincial and municipal access data center optimization method for predicting number of potential users by electric power customer service center based on logistic model |
| CN109559146A (en) * | 2018-09-25 | 2019-04-02 | 国家电网有限公司客户服务中心 | Electricity customer service center accesses data center's optimization method based on the provinces and cities of logistic model prediction potential user's quantity |
| CN109614982A (en) * | 2018-10-18 | 2019-04-12 | 平安科技（深圳）有限公司 | Product analysis method, apparatus, computer equipment and storage medium |
| CN109697626A (en) * | 2018-10-19 | 2019-04-30 | 上海吉贝克信息技术有限公司 | Marketing system, method, electric terminal and medium based on JeeSite quick development platform |
| CN109598479B (en) * | 2018-10-25 | 2024-04-05 | 三六零科技集团有限公司 | Bill extraction method and device, electronic equipment and medium |
| CN109598479A (en) * | 2018-10-25 | 2019-04-09 | 北京奇虎科技有限公司 | A kind of bill extracting method, device, electronic equipment and medium |
| CN109740620B (en) * | 2018-11-12 | 2023-09-26 | 平安科技（深圳）有限公司 | Method, device, equipment and storage medium for establishing crowd figure classification model |
| WO2020098308A1 (en) * | 2018-11-12 | 2020-05-22 | 平安科技（深圳）有限公司 | Method, device and equipment for establishing crowd portrait classification medel and storage medium |
| CN109740620A (en) * | 2018-11-12 | 2019-05-10 | 平安科技（深圳）有限公司 | Method for building up, device, equipment and the storage medium of crowd portrayal disaggregated model |
| CN109785034A (en) * | 2018-11-13 | 2019-05-21 | 北京码牛科技有限公司 | User's portrait generation method, device, electronic equipment and computer-readable medium |
| CN111221868A (en) * | 2018-11-26 | 2020-06-02 | 国网上海市电力公司 | Data mining and analyzing method applied to channel preference of power customer |
| CN109636235A (en) * | 2018-12-26 | 2019-04-16 | 北京汽车研究总院有限公司 | The determination method and processing system of driving behavior portrait model |
| CN109948913A (en) * | 2019-03-01 | 2019-06-28 | 长沙理工大学 | A kind of multi-source feature power consumer composite portrait system based on double-deck xgboost algorithm |
| CN109934273A (en) * | 2019-03-01 | 2019-06-25 | 长沙理工大学 | It is a kind of based on the fault characteristic of DML-KNN algorithm and active damage repair technology draw a portrait new method |
| CN111723273A (en) * | 2019-03-18 | 2020-09-29 | 北京中电翔云信息技术有限公司 | Smart cloud retrieval system and method |
| CN110119998B (en) * | 2019-03-27 | 2023-04-28 | 广东浩迪创新科技有限公司 | Renting house information management method and system based on intelligent ammeter |
| CN110119998A (en) * | 2019-03-27 | 2019-08-13 | 广东浩迪创新科技有限公司 | Rental house approaches to IM and system based on intelligent electric meter |
| CN110046657A (en) * | 2019-03-29 | 2019-07-23 | 武汉大学深圳研究院 | A kind of social safety figure painting image space method based on multiple view study |
| CN110046657B (en) * | 2019-03-29 | 2022-06-14 | 武汉大学深圳研究院 | Social security figure image method based on multi-view learning |
| CN110134860B (en) * | 2019-04-12 | 2023-04-07 | 创新先进技术有限公司 | User portrait generation method, device and equipment |
| CN110134860A (en) * | 2019-04-12 | 2019-08-16 | 阿里巴巴集团控股有限公司 | User's portrait generation method, device and equipment |
| CN110210018A (en) * | 2019-05-14 | 2019-09-06 | 北京百度网讯科技有限公司 | It registers the matching process and device of department |
| CN110210018B (en) * | 2019-05-14 | 2023-07-11 | 北京百度网讯科技有限公司 | Matching method and device for registration department |
| CN110191183A (en) * | 2019-05-31 | 2019-08-30 | 深圳市医信科技有限公司 | Accurate intelligent method for pushing, system, device and computer readable storage medium |
| CN110263255A (en) * | 2019-06-21 | 2019-09-20 | 北京字节跳动网络技术有限公司 | Acquisition methods, system, server and the storage medium of customer attribute information |
| CN110276647A (en) * | 2019-06-26 | 2019-09-24 | 中国南方电网有限责任公司 | Online service method based on user integral |
| CN110347923A (en) * | 2019-07-09 | 2019-10-18 | 葛晓滨 | A kind of method for the fast-neutron fission formula building user's portrait recalled |
| CN110334463B (en) * | 2019-07-12 | 2021-06-18 | 无锡锐泰节能系统科学有限公司 | Energy management design method based on energy consumption portrait model |
| CN110334463A (en) * | 2019-07-12 | 2019-10-15 | 无锡锐泰节能系统科学有限公司 | Based on the energy management design method with the model that can draw a portrait |
| CN110472852A (en) * | 2019-08-02 | 2019-11-19 | 上海云扩信息科技有限公司 | A kind of experience assessment implementation management method of electrical power services application |
| CN110472852B (en) * | 2019-08-02 | 2023-03-03 | 上海云扩信息科技有限公司 | Experience evaluation implementation management method for power service application |
| CN110599237A (en) * | 2019-08-15 | 2019-12-20 | 国网山东省电力公司临清市供电公司 | Electric power marketing management information acquisition module based on intelligent monitoring |
| CN110516155A (en) * | 2019-08-29 | 2019-11-29 | 深圳市云积分科技有限公司 | Marketing strategy generation method and system |
| CN110728537A (en) * | 2019-09-24 | 2020-01-24 | 国网河北省电力有限公司信息通信分公司 | Prediction payment method based on power consumer behavior label |
| CN110688580A (en) * | 2019-10-14 | 2020-01-14 | 国网山东省电力公司泰安供电公司 | Power supply service data display system based on customer portrait |
| CN110689385A (en) * | 2019-10-16 | 2020-01-14 | 国网山东省电力公司信息通信公司 | Power customer service user portrait construction method based on knowledge graph |
| CN111080367A (en) * | 2019-12-20 | 2020-04-28 | 深圳市国电科技通信有限公司 | Power consumption behavior analysis method for sensing power consumption state of low-voltage user |
| CN111080367B (en) * | 2019-12-20 | 2023-11-28 | 深圳市国电科技通信有限公司 | Power consumption behavior analysis method for sensing power consumption state of low-voltage user |
| CN111506636A (en) * | 2020-05-12 | 2020-08-07 | 上海积成能源科技有限公司 | System and method for analyzing residential electricity consumption behavior based on autoregressive and neighbor algorithm |
| CN111666313A (en) * | 2020-05-25 | 2020-09-15 | 中科星图股份有限公司 | Correlation construction and multi-user data matching method based on multi-source heterogeneous remote sensing data |
| CN111666313B (en) * | 2020-05-25 | 2023-02-07 | 中科星图股份有限公司 | Correlation construction and multi-user data matching method based on multi-source heterogeneous remote sensing data |
| CN111401798A (en) * | 2020-06-02 | 2020-07-10 | 南京百敖软件有限公司 | Enterprise waste escaping and debt risk early warning system and construction method |
| CN111738462B (en) * | 2020-06-08 | 2022-08-30 | 国网江苏省电力有限公司常州供电分公司 | Fault first-aid repair active service early warning method for electric power metering device |
| CN111738462A (en) * | 2020-06-08 | 2020-10-02 | 国网江苏省电力有限公司常州供电分公司 | Fault first-aid repair active service early warning method for electric power metering device |
| CN111709791A (en) * | 2020-06-19 | 2020-09-25 | 四川中电启明星信息技术有限公司 | Power supply marketing service method based on improved feature word weight algorithm |
| CN112115355A (en) * | 2020-09-02 | 2020-12-22 | 南方电网深圳数字电网研究院有限公司 | Power grid intelligent service system based on user portrait in power industry |
| CN112131475B (en) * | 2020-09-25 | 2023-10-10 | 重庆邮电大学 | Interpretable and interactive user portrayal method and device |
| CN112131475A (en) * | 2020-09-25 | 2020-12-25 | 重庆邮电大学 | Interpretable and interactive user portrait method and device |
| CN112184313A (en) * | 2020-09-28 | 2021-01-05 | 国网上海市电力公司 | Power marketing business application system based on client portrait data |
| CN112132631A (en) * | 2020-09-29 | 2020-12-25 | 国网上海市电力公司 | Label electricity utilization management method based on power customer portrait |
| CN112053100A (en) * | 2020-10-15 | 2020-12-08 | 东北电力大学 | Big data-based analysis method for electricity utilization behavior of user |
| CN112529630A (en) * | 2020-12-17 | 2021-03-19 | 国网河北省电力有限公司营销服务中心 | Construction method of multi-dimensional informatization label system of power consumer |
| CN112417308A (en) * | 2020-12-17 | 2021-02-26 | 国网河北省电力有限公司营销服务中心 | User portrait label generation method based on electric power big data |
| CN112579721B (en) * | 2020-12-22 | 2023-10-27 | 北京市腾河智慧能源科技有限公司 | Method and system for constructing crowd distribution map, terminal equipment and storage medium |
| CN112579721A (en) * | 2020-12-22 | 2021-03-30 | 北京市腾河智慧能源科技有限公司 | Method and system for constructing crowd distribution map, terminal device and storage medium |
| CN112862530A (en) * | 2021-02-09 | 2021-05-28 | 江之龙 | Marketing system based on big data |
| CN113159419A (en) * | 2021-04-21 | 2021-07-23 | 成都卫士通信息产业股份有限公司 | Group feature portrait analysis method, device and equipment and readable storage medium |
| CN113487448A (en) * | 2021-05-31 | 2021-10-08 | 国网上海市电力公司 | Power credit labeling method and system based on power big data |
| CN113313470B (en) * | 2021-06-10 | 2023-06-09 | 郑州科技学院 | Employment type assessment method and system based on big data |
| CN113313470A (en) * | 2021-06-10 | 2021-08-27 | 郑州科技学院 | Employment type evaluation method and system based on big data |
| CN113344638A (en) * | 2021-06-29 | 2021-09-03 | 云南电网有限责任公司信息中心 | Hypergraph-based power grid user group portrait construction method and device |
| CN113344638B (en) * | 2021-06-29 | 2022-05-24 | 云南电网有限责任公司信息中心 | Power grid user group portrait construction method and device based on hypergraph |
| CN113377760A (en) * | 2021-07-06 | 2021-09-10 | 国网江苏省电力有限公司营销服务中心 | Method and system for establishing low-voltage resident feature portrait based on electric power data and multivariate data |
| CN113554241A (en) * | 2021-09-02 | 2021-10-26 | 国网山东省电力公司泰安供电公司 | User layering method and prediction method based on user electricity complaint behaviors |
| CN114219241A (en) * | 2021-12-01 | 2022-03-22 | 深圳供电局有限公司 | Customer electricity consumption behavior analysis method and system |
| CN114417955A (en) * | 2021-12-02 | 2022-04-29 | 国网浙江省电力有限公司 | Power index portrait real-time generation method and device suitable for villages and storage medium |
| CN114417955B (en) * | 2021-12-02 | 2023-10-31 | 国网浙江省电力有限公司 | Method, device and storage medium for generating power index image suitable for village in real time |
| CN114595323B (en) * | 2022-03-04 | 2023-03-10 | 北京百度网讯科技有限公司 | Portrait construction, recommendation, model training method, apparatus, device and storage medium |
| CN114595323A (en) * | 2022-03-04 | 2022-06-07 | 北京百度网讯科技有限公司 | Portrait construction, recommendation, model training method, apparatus, device and storage medium |
| CN116450952A (en) * | 2023-06-16 | 2023-07-18 | 天津星耀九洲科技有限公司 | Internet user portrait generation method and system based on deep learning technology |
| CN117271905B (en) * | 2023-11-21 | 2024-02-09 | 杭州小策科技有限公司 | Crowd image-based lateral demand analysis method and system |
| CN117271905A (en) * | 2023-11-21 | 2023-12-22 | 杭州小策科技有限公司 | Crowd image-based lateral demand analysis method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106651424B (en) | 2020-05-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106651424A (en) | Electric power user figure establishment and analysis method based on big data technology | |
| CN103714139B (en) | Parallel data mining method for identifying a mass of mobile client bases | |
| CN109919652A (en) | User group's classification method, device, equipment and storage medium | |
| US8255392B2 (en) | Real time data collection system and method | |
| Zhao et al. | Clustering analysis of ridership patterns at subway stations: A case in Nanjing, China | |
| US20060218169A1 (en) | Constrained tree structure method and system | |
| CN109447364A (en) | Power customer based on label complains prediction technique | |
| CN107730269A (en) | A kind of Electricity customers portrait method of Behavior-based control analysis | |
| CN110119948A (en) | Based on when variable weight dynamic combined power consumer credit assessment method and system | |
| CN111178721A (en) | Intelligent tourism system | |
| CN113516500A (en) | Implementation method and system based on big data business and travel operation platform | |
| CN111221868A (en) | Data mining and analyzing method applied to channel preference of power customer | |
| Wang et al. | DeepFM-Based taxi pick-up area recommendation | |
| US20070118416A1 (en) | Method and system for planning | |
| Green | Local action on labour market integration of new arrivals: issues and dilemmas for policy | |
| Ryan | What are the longer-term outcomes for individuals completing vocational education and training qualifications? | |
| CN114611622A (en) | Method for identifying cross-city commuting crowd by utilizing mobile phone data | |
| CN107908620A (en) | A kind of method and apparatus based on job documentation anticipation user's occupation | |
| US20220036460A1 (en) | Systems and Methods for Asset Analysis | |
| Liang et al. | Human capital and the re‐employment of retrenchment labor in urban China | |
| CN109919811B (en) | Insurance agent culture scheme generation method based on big data and related equipment | |
| Thurgood et al. | Using online job vacancies to understand the UK labour market from the bottom-up | |
| CN109800942A (en) | Computer room operation management method, electronic device and storage medium | |
| CN110264244A (en) | A kind of advertising user track following management system and method | |
| Alvarez | Evaluating electric vehicle user mobility data using neural network-based language models |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 250002 Wang Yue Road, Ji'nan City, Shandong Province, No. 2000 Patentee after: ELECTRIC POWER RESEARCH INSTITUTE OF STATE GRID SHANDONG ELECTRIC POWER Co. Patentee after: SHANDONG LUNENG SOFTWARE TECHNOLOGY Co.,Ltd. Patentee after: STATE GRID CORPORATION OF CHINA Address before: 250002 Wang Yue Road, Ji'nan City, Shandong Province, No. 2000 Patentee before: ELECTRIC POWER RESEARCH INSTITUTE OF STATE GRID SHANDONG ELECTRIC POWER Co. Patentee before: SHANDONG LUNENG SOFTWARE TECHNOLOGY Co.,Ltd. Patentee before: State Grid Corporation of China |
|
| CP01 | Change in the name or title of a patent holder | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210407 Address after: No. 150, Jinger Road, Daguanyuan, Shizhong District, Jinan City, Shandong Province Patentee after: Shandong Electric Power Marketing Center Patentee after: ELECTRIC POWER RESEARCH INSTITUTE OF STATE GRID SHANDONG ELECTRIC POWER Co. Patentee after: SHANDONG LUNENG SOFTWARE TECHNOLOGY Co.,Ltd. Patentee after: STATE GRID CORPORATION OF CHINA Address before: 250002 Wang Yue Road, Ji'nan City, Shandong Province, No. 2000 Patentee before: ELECTRIC POWER RESEARCH INSTITUTE OF STATE GRID SHANDONG ELECTRIC POWER Co. Patentee before: SHANDONG LUNENG SOFTWARE TECHNOLOGY Co.,Ltd. Patentee before: STATE GRID CORPORATION OF CHINA |
|
| TR01 | Transfer of patent right | ||
| CP01 | Change in the name or title of a patent holder |
Address after: No. 150, Jinger Road, Daguanyuan, Shizhong District, Jinan City, Shandong Province Patentee after: Shandong Electric Power Marketing Center Patentee after: ELECTRIC POWER RESEARCH INSTITUTE OF STATE GRID SHANDONG ELECTRIC POWER Co. Patentee after: Shandong luruan Digital Technology Co.,Ltd. Patentee after: STATE GRID CORPORATION OF CHINA Address before: No. 150, Jinger Road, Daguanyuan, Shizhong District, Jinan City, Shandong Province Patentee before: Shandong Electric Power Marketing Center Patentee before: ELECTRIC POWER RESEARCH INSTITUTE OF STATE GRID SHANDONG ELECTRIC POWER Co. Patentee before: SHANDONG LUNENG SOFTWARE TECHNOLOGY Co.,Ltd. Patentee before: STATE GRID CORPORATION OF CHINA |
|
| CP01 | Change in the name or title of a patent holder |