CN109993380A - A kind of information processing method, device and computer readable storage medium - Google Patents
A kind of information processing method, device and computer readable storage medium Download PDFInfo
- Publication number
- CN109993380A CN109993380A CN201711472693.4A CN201711472693A CN109993380A CN 109993380 A CN109993380 A CN 109993380A CN 201711472693 A CN201711472693 A CN 201711472693A CN 109993380 A CN109993380 A CN 109993380A
- Authority
- CN
- China
- Prior art keywords
- user
- data
- value
- money
- history
- 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
- 230000010365 information processing Effects 0.000 title claims abstract description 37
- 238000003672 processing method Methods 0.000 title claims abstract description 23
- 238000003860 storage Methods 0.000 title claims abstract description 20
- 230000003993 interaction Effects 0.000 claims abstract description 46
- 238000004891 communication Methods 0.000 claims abstract description 15
- 238000000034 method Methods 0.000 claims description 39
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 claims description 31
- 238000004590 computer program Methods 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 12
- 230000002159 abnormal effect Effects 0.000 claims description 6
- 230000004907 flux Effects 0.000 claims description 6
- 230000000295 complement effect Effects 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 description 15
- 238000010586 diagram Methods 0.000 description 11
- 238000004458 analytical method Methods 0.000 description 10
- 230000006399 behavior Effects 0.000 description 8
- 238000012545 processing Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 6
- 239000011159 matrix material Substances 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 235000013399 edible fruits Nutrition 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000002596 correlated effect Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 235000012054 meals Nutrition 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 239000002253 acid Substances 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000000875 corresponding effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000556 factor analysis Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000011017 operating method Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000000452 restraining effect Effects 0.000 description 1
- 230000011664 signaling Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000004260 weight control Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Theoretical Computer Science (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention discloses a kind of information processing methods, carry out pretreatment to the user data of acquisition and obtain pending data, the pending data includes: user's communication bill data;User's interaction coefficient and user's initial weight are calculated according to the pending data;User's value assessment result is calculated according to user's interaction coefficient and user's initial weight.The embodiment of the invention also discloses a kind of information processing unit and computer readable storage mediums, and the accuracy of user's value assessment result calculating can be improved.
Description
Technical field
The present invention relates to communications field communication data processing technique more particularly to a kind of information processing method, device and meters
Calculation machine readable storage medium storing program for executing.
Background technique
It promotes and executes with the policy that speed-raising drop is taken, the income of operator faces more acid test, and telecommunications industry is just
Seek new income increase point.Under the premise of flow operation increases income major impetus as business, 4G user has become operator
Main business income, as 4G business is nationwide universal, advantage of the operator in terms of developing new networking user into
Enter bottleneck period, and how to make storage user's value preserving and appreciation, it has also become the significant problem that telecommunications industry is paid close attention to jointly.
Client general at present possesses method and mainly concentrates the social networks such as short massage notice, business hall promotional pamphlet, public platform
Network the modes such as propagates and informs the existing favor information of storage client, without differences marketing mode cause cost it is a large amount of export and
It produces little effect.Precision marketing, fine integral method guiding under, the value preserving and appreciation of storage user are needed that differentiation is taken to seek
Pin means, i.e. identification user value and potential value, are oriented according to the point of interest of its consumer behavior habit, social networks circle
Advertisement is launched, and accurate humanistic care pays close attention to user demand conscientiously, is supplied to each user's customized product and set meal is preferential, from
And realize storage user's value preserving and appreciation.The prior art is mainly the research to user's Valuation Method and system, lacks and is based on
The fining for feature user group of the feature constructions such as the distinctive signaling of mobile subscriber, set meal content, call relational network
Valuation Method.
Summary of the invention
In order to solve the above technical problems, the embodiment of the present invention provides a kind of information processing method, device and computer-readable
The accuracy of user's value assessment result calculating can be improved in storage medium.
The technical scheme of the present invention is realized as follows:
The embodiment of the present invention provides a kind of information processing method, which comprises
Pretreatment is carried out to the user data of acquisition and obtains pending data, the pending data includes: user's communication
Billing data;
User's interaction coefficient and user's initial weight are calculated according to the pending data;
User's value assessment result is calculated according to user's interaction coefficient and user's initial weight.
Further, the user data of described pair of acquisition carries out pretreatment and obtains pending data, comprising:
User data is obtained, the data of the group user mark in the user data are rejected, it will be in the user data
Null value mend be 0, the abnormal Value Data in the user data is substituted with mean value, respectively by the flow in the user data
Unit and income unit carry out unification, obtain pending data;
Wherein, the pending data includes: call detailed list of the user identifier to be assessed in the first preset time, described
Summarize the moon of the user identifier to be assessed before the time to be assessed in second preset time using flow, the credit amount of money, increase
The history arrearage amount of money and arrearage number, history of value business total income and home broadband income and the user identifier to be assessed
Terminal purchases machine money and history terminal purchase cost.
It is further, described that user's interaction coefficient and user's initial weight are calculated according to the pending data, comprising:
According in the pending data in call detailed list the duration of call and talk times calculate user interaction system
Number;
It is wide using flow, the credit amount of money, value-added service total income, family according to summarizing the moon in the pending data
Band income, the history arrearage amount of money and arrearage number, history terminal purchase machine money and history terminal purchase cost calculate the user
Initial weight.
Further, the duration of call and talk times according in the pending data in call detailed list calculates institute
State user's interaction coefficient, comprising:
User's interaction coefficient Interaction_Degree (i, j) is calculated according to the following formula,
Wherein, n indicates the talk times between user i and user j, calltime(i,j)Indicate call detailed list in user i and
The duration of call between user j.
Further, it is described according to summarize flow in the pending data moon, the credit amount of money, value-added service are always received
Enter, home broadband income, the history arrearage amount of money and arrearage number, history terminal purchase machine money and history terminal purchase cost meter
Calculate user's initial weight, comprising:
User initial weight Initial_weight (i) is calculated according to following formula,
Wherein, n indicates the number of the history terminal purchase machine of user i, and terminal, which purchases machine money (i, j), indicates user i jth time purchase
Terminal pays the expense of operator, and terminal purchases the terminal payment that machine cost (i, j) indicates that operator is user i jth time purchase
Cost, summarizing flow (i) moon indicates the summarizing the moon using flow of user i, and the credit amount of money (i) indicates prestoring for user i
The telephone expenses amount of money, value-added service total income (i) indicate the value-added service total income of user i, and (i) is taken in broadband indicates the family of user i
Front yard broadband income, m indicate that the history arrearage number of user i, the arrearage amount of money (i, k) indicate that the amount of money of user i kth time arrearage, e are
Constant.
Further, described that user's value assessment knot is calculated according to user's interaction coefficient and user's initial weight
Fruit, comprising:
The value assessment result Value (i) of user i is calculated according to the following formula,
Wherein, Interaction_Degree (i, j) is user's interaction coefficient of user i, and Initial_weight (i) is
User's initial weight of user i.
The embodiment of the present invention also provides a kind of information processing unit, and described device includes: transceiver and processor, wherein
The transceiver, for obtaining user data;
The processor obtains pending data, the number to be processed for carrying out pretreatment to the user data of acquisition
According to including: user's communication bill data;It is also used to calculate user's interaction coefficient according to the pending data and user initially weighs
Weight;It is also used to calculate user's value assessment result according to user's interaction coefficient and user's initial weight.
Further, the processor will specifically for rejecting the data of the group user mark in the user data
It is 0 that null value in the user data, which is mended, the abnormal Value Data in the user data is substituted with mean value, respectively by the use
Flux unit and income unit in user data carry out unification, obtain pending data;
Wherein, the pending data includes: call detailed list of the user identifier to be assessed in the first preset time, described
Summarize the moon of the user identifier to be assessed before the time to be assessed in second preset time using flow, the credit amount of money, increase
The history arrearage amount of money and arrearage number, history of value business total income and home broadband income and the user identifier to be assessed
Terminal purchases machine money and history terminal purchase cost.
Further, the processor, is specifically used for:
According in the pending data in call detailed list the duration of call and talk times calculate user interaction system
Number;
It is wide using flow, the credit amount of money, value-added service total income, family according to summarizing the moon in the pending data
Band income, the history arrearage amount of money and arrearage number, history terminal purchase machine money and history terminal purchase cost calculate the user
Initial weight.
Further, the processor, is specifically used for:
User's interaction coefficient Interaction_Degree (i, j) is calculated according to the following formula,
Wherein, n indicates the talk times between user i and user j, calltime(i,j)Indicate call detailed list in user i and
The duration of call between user j.
Further, the processor, is specifically used for:
User initial weight Initial_weight (i) is calculated according to following formula,
Wherein, n indicates the number of the history terminal purchase machine of user i, and terminal, which purchases machine money (i, j), indicates user i jth time purchase
Terminal pays the expense of operator, and terminal purchases the terminal payment that machine cost (i, j) indicates that operator is user i jth time purchase
Cost, summarizing flow (i) moon indicates the summarizing the moon using flow of user i, and the credit amount of money (i) indicates prestoring for user i
The telephone expenses amount of money, value-added service total income (i) indicate the value-added service total income of user i, and (i) is taken in broadband indicates the family of user i
Front yard broadband income, m indicate that the history arrearage number of user i, the arrearage amount of money (i, k) indicate that the amount of money of user i kth time arrearage, e are
Constant.
Further, the processor, is specifically used for:
The value assessment result Value (i) of user i is calculated according to the following formula,
Wherein, Interaction_Degree (i, j) is user's interaction coefficient of user i, and Initial_weight (i) is
User's initial weight of user i.
The embodiment of the present invention also provides a kind of computer readable storage medium, stores in the computer readable storage medium
There is computer program, as above described in any item information processing methods are realized when which is executed by processor.
The embodiment of the present invention also provides a kind of information processing unit, and the system comprises processors and memory;Wherein,
The memory, for storing the computer program that can be run on the processor;
The processor executes the step of the as above any one information processing method when for running the computer program
Suddenly.
The embodiment of the invention provides a kind of information processing method, device and computer readable storage mediums, to acquisition
User data carries out pretreatment and obtains pending data, and the pending data includes: user's communication bill data;According to described
Pending data calculates user's interaction coefficient and user's initial weight;It is initially weighed according to user's interaction coefficient and the user
Re-computation user's value assessment result.Information processing method, device and computer-readable storage medium provided in an embodiment of the present invention
Matter can quickly calculate the value assessment of mobile subscriber as a result, analyzing by user's initial weight and user's interaction coefficient
Continuous data calculation method is used in the process, reduces section, to exclude artificial division data set as far as possible and cause to finish
Fruit distortion improves the accuracy that user's value assessment result calculates.
Detailed description of the invention
Fig. 1 is information processing method flow diagram one provided in an embodiment of the present invention;
Fig. 2 is that pending data provided in an embodiment of the present invention processing obtains flow diagram;
Fig. 3 is information processing method flow diagram two provided in an embodiment of the present invention;
Fig. 4 is information processing unit structural schematic diagram provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description.
User in social networks link is detached into individual independent of each other by the prior art, only with linear statistical
Computing rule measure the value of user, have ignored the valuation effects of user's cluster under big data scale.By flow operation
It is converted under the new demand of Digital Services supplier, value of the user for enterprise not only has with the telephone expenses of user income
Association, it is also related to the social status of user, economic conditions etc., and these information can be presented from the social network of user
As a result, therefore, the prior art ignores incidence relation between user.The prior art only passes through the income data in BOSS charge system
The value of user is measured, lacks and the analysis foundation of user behavior is supported, have ignored the potential value of user.It is existing simultaneously
Scheme does not account for the special users such as group customer group, and the consumption data of this kind of user has biggish human factor to influence, no
User's value directly can be measured by consumption data, therefore, the prior art lacks user behavior analysis and ignores the potential of user
Value.The prior art is defeated as user's value judgement by position data, internet behavior data in user's DPI data after parsing
Enter index, have ignored user's customer consumption ability in other main businesses of the operators such as voice communication, dats services with increment,
User's value judgement accuracy is influenced, therefore, prior art input analysis indexes content is low, influences the accuracy of precision marketing.
User's Valuation Method has biggish subjective judgement factor, such as similar technique to the pretreatment mode of data in the prior art
Scheme mentions cell day scoring calculation method, calculation method of the affiliated value interval of terminal etc., and subjectivity divides data interval and mould
The experience of type user is closely bound up, is affected to the objective value judgement of end user;Therefore, the valence that the prior art proposes
Value appraisal procedure is mostly that subjective factor is leading, objective can not be worth and position to user.
To solve above-mentioned prior art drawback, the invention proposes adding for the PageRank webpage sorting calculated based on graph theory
Weigh Fast Convergent Algorithm (Weighted Page Rank for Quick Convergence, WPRQC), using personal user as
Node in social networks constructs network link structure using correspondence as the side of connection network node.It unites with big data
Analysis method is counted, the input field extracted is subjected to index factor and analyzes the weight initial as nodes, will be used
Weight of the call frequency as side between network node between family is calculated by algorithm iteration until convergence, obtains user most
Final value value;In data preprocessing phase, by whether group customer attribute field, filters out the personal user of full dose as model
The input main body of analysis guarantees the objectivity and universality of model application;The comprehensive behavior number of user has been comprehensively considered simultaneously
According to user's calling and called relationship being extracted from call detailed list to construct social network relationships figure, extracts user from BOSS system
Summarize the duration of call moon, the moon summarizes flow, credit is extracted from charge system, arrearage record, value-added service are always received
Enter data.In addition, method provided by the invention has also merged user terminal purchase machine money, purchase cost and home broadband income number
According to measurement user is worth comprehensively;In index factor analytic process, using continuous data calculation method, section is reduced, thus
Artificial division data set is excluded as far as possible and causes result distortion, improves the accuracy of user's value calculation.
Information processing method provided in an embodiment of the present invention, specially user are worth big data appraisal procedure and are based on graph theory
The weighting Fast Convergent Algorithm of calculating can be generalized to other users value analysis field simultaneously, can be according to business scenario difference
Determine the weight control algolithm iteration ranges become between input field index factor and network node and degree, it is final to obtain special project
User's final value in field.Limited in view of length, the embodiment of the present invention will complete skill using motion method with operator's angle
The description of art scheme.
The embodiment of the present invention provides a kind of information processing method, as shown in Figure 1, this method may include:
Step 101 carries out pretreatment acquisition pending data to the user data of acquisition.
The pending data includes: user's communication bill data.
Specifically, information processing method provided in an embodiment of the present invention is that user's value big data based on link structure is commented
Estimate method, the executing subject of this method is information processing unit, i.e. information processing unit locates the user data of acquisition in advance
Reason obtains pending data.
Specifying information processing unit carries out pretreatment to obtain pending data including: information processing to the user data of acquisition
Device obtains user data, the data of the group user mark in the user data is rejected, by the sky in the user data
Value complement is 0, and the abnormal Value Data in the user data is substituted with mean value, respectively by the flux unit in the user data
Unification is carried out with income unit, obtains pending data, is i.e. acquisition user's communication bill data;
Wherein, the pending data includes: call detailed list of the user identifier to be assessed in the first preset time, described
Summarize the moon of the user identifier to be assessed before the time to be assessed in second preset time using flow, the credit amount of money, increase
The history arrearage amount of money and arrearage number, history of value business total income and home broadband income and the user identifier to be assessed
Terminal purchases machine money and history terminal purchase cost.
Here, user identifier can be the phone number of user, and the first preset time can be 3 months, 6 months or 1
Year etc., it can also be other values that the second preset time, which can be 1 month, 2 months etc., the first preset time, second it is default when
Between specific value can carry out selection setting with actual conditions, it is not limited in the embodiment of the present invention.
Illustratively, it as shown in Fig. 2, using phone number as user's unique ID, extracts all to value assessment use
Nearly 6 months call detailed lists at family, month to be assessed toward be pushed forward one month summarize flow the moon, credit, value-added service are always received
Enter, home broadband income, history terminal purchases purchase machine money, purchase cost, the arrearage record of machine;Reject group customer, null value or
Null value complement 0 asks distribution to every class field, and abnormal Value Data is substituted with mean value;Flux unit (GB), income unit system (member),
Export data to be processed.
It specifically can be the data for obtaining a collection of user, the field contents required according to the following equation input a collection of user
Data, eventually export the PR value of this crowd of user.
Step 102 calculates user's interaction coefficient and user's initial weight according to the pending data.
Specific information processing unit calculates user's interaction coefficient and user's initial weight, packet according to the pending data
It includes: according to the duration of call and talk times calculating user's interaction coefficient in the pending data in call detailed list;Root
According to summarize the moon in the pending data using flow, the credit amount of money, value-added service total income, home broadband income, go through
The history arrearage amount of money and arrearage number, history terminal purchase machine money and history terminal purchase cost calculate user's initial weight.
The specific initial weight for calculating user node include: gathered information by user month using flow, credit
The amount of money, the history arrearage amount of money and arrearage number, value-added service total income, terminal purchase machine money, terminal purchase cost, home broadband are received
Mathematical transformation and the linear weighted function summation for entering each index, calculate the first knowledge weight of user node.
The specific network edge weight for calculating connection user includes: in user social contact cyberrelationship figure, between Bian Daibiao user
The tightness degree of connection.The present invention program extracts the duration of call in the detailed list of user's communication and talk times as measurement standard,
The duration of call is longer, and the tightness degree connected between user is higher, and talk times are more, and the tightness degree connected between user is got over
It is high.
User social contact cyberrelationship figure is constructed with call detailed list relationship, i.e. building mobile subscriber's network interaction topological diagram.
Specifically, interaction coefficient embody user in a mobile network with the level of interaction of other users.In operator and
Speech, level of interaction mean voice consumption income and potential flow, value-added service income, because having relatively strong for universal
Demand of the user of interaction demand usually to social is stronger, and strong social demand will drive disappearing for voice and flow
Take.
User's interaction coefficient Interaction_Degree (i, j) is calculated according to the following formula,
Wherein, Interaction_Degree (i, j) indicates the tightness degree connected between user i and user j, and n is indicated
Talk times between user i and user j, calltime(i,j)When indicating the call in call detailed list between user i and user j
It is long.Being allowed in the form of product therebetween between talk times and the duration of call is the relationship mutually enhanced, call
Number is more and each duration of call is longer, and the connection between user is closer.Due to differing greatly between numerical value, use is with e
The logarithmic form at bottom can control the numberical range of interaction coefficient in a certain range, avoid subsequent comprehensive multi-index calculate in by
It is excessive in numerical value difference and ignore certain important indicators, meanwhile, in order to guarantee the nonnegativity of interaction coefficient, formula end using plus
1 form.
Specifically, initial weight indicates that individually for the behavioural analysis of some user, the value of user is big
It is small.For operator, summarize that flow is bigger, the credit amount of money is higher, value-added service total income is higher, home broadband the moon
Income gets over Gao Yuegao, and user's value is bigger, and arrearage number is more and the arrearage amount of money is bigger, is worth lower.Meanwhile such as
Fruit user habit buys high value terminal from carrier side, then the value of user is just higher.
User initial weight Initial_weight (i) is calculated according to following formula,
Wherein, n indicates the number of user i history terminal purchase machine, and terminal, which purchases machine money (i, j), indicates user i jth time purchase eventually
The expense of operator is paid at end, and terminal purchases the terminal payment that machine cost (i, j) indicates that operator is user i jth time purchase
Cost, user operator purchase terminal bring operator profit and user operator value be positively correlated;M indicates user
The history arrearage number of i, the arrearage amount of money (i, k) indicate the amount of money of user i kth time arrearage, the history arrearage information and use of user
Family is negatively correlated in the value of operator;Index is standardized using natural logrithm, is inconsistent in order to remove linear module
And cause influence of all kinds of indexs to final result.Summarizing flow (i) moon indicates that user i summarizes the moon using flow, prestores words
Take the credit amount of money of the amount of money (i) expression user i, the value-added service total income of value-added service total income (i) expression user i,
(i) is taken in broadband indicates the home broadband income of user i.
Method provided in an embodiment of the present invention, user have the history for buying terminal at operator, use carrier side
The contribution of family value assessment is very big, therefore is combined in a manner of product between the index and remaining index.At the beginning of guaranteeing user
The nonnegativity of beginning weight, in formula plus e.
Step 103 calculates user's value assessment result according to user's interaction coefficient and user's initial weight.
Specifically, WPRQC algorithm according to the following formula calculates the value assessment result Value (i) of user i,
Wherein, the value assessment of user i is as a result, there are the weights of the user j of call relationship to be multiplied by with user i by all
Degree of being completely embedded between family i and user j is summed.
Algorithmic statement process depend on control of the user to the condition of convergence, the present invention implement in, algorithm realize on improve
The PageRank algorithm in Spark machine learning library is realized and is added by optimizing and revising maximum number of iterations and resetting probability at random
Weigh Fast Convergent Algorithm WPRQC convergence and tuning.
The core of improved PageRank algorithm is just the WPRQC algorithm of above-mentioned formula.It is building when specific implementation
One huge sparse matrix, original value is 0,1 in matrix, and after WPRQC algorithm weights, the value in matrix is also corresponding
Ground is changed, and by the calculating between matrix, calculates final result as matrix multiple carrys out iteration.
The user's value assessment result calculated in method provided in an embodiment of the present invention is to operator's power-assisted service transformation, essence
Quasi- marketing has directive significance, specifically: mobile subscriber's value assessment result based on link structure can be used to help operator
It realizes key customer's child care work, embodies enterprises service to people, people-oriented customer care objective, realize and existing business is advised
The value preserving extra earning of mould.User's value assessment result realizes precision marketing in combination with existing tag library.Such as combinable interest tags,
Recommend own Digital Services, the value score of user can also be opened in the form of API to industry customer, realize big data
The realization etc. of value analysis.It is not limited only to a certain operator due to analyzing user's value by network of personal connections, is the whole network
Matter, therefore the conversion of high-value user can be realized in conjunction with user's belonging network, promote carrier market competitiveness.
Information processing method provided in an embodiment of the present invention can power-assisted promotion enterprise by the judgement to user's high value
Precision marketing, preserve value extra earning in the case where cost payout is certain.Existing user's value judgement scheme has ignored big data
Under the conditions of, the value association between user, and the data of value analysis are used for, whole communication behavior rails of user cannot be portrayed
Mark.In view of considerations above, as shown in figure 3, the embodiment of the present invention provides a kind of information processing method, this method is advised for big data
Based on user's value judgement of user link structure under mould, this method constructs network topology, net with the call relationship between user
Node on behalf user to be identified in network, the tightness degree contacted between Bian Daibiao user, PageRank is calculated in webpage sorting
It is also high-value user that method, which is specified by the user that a large amount of high-value users are directed toward, by constantly iterating to calculate, until algorithmic statement,
To calculate the value of user.Weighting Fast Convergent Algorithm WPRQC improves the precision of algorithm, accelerates convergence rate, this
The weight that method contacts side to the weight of initial user node, user is redefined, and enables the algorithm to quickly calculate
The value of mobile subscriber out.As described in above-mentioned formula, defining using which parameter is node for the calculating of specific assignment and PR value
Assignment, which parameter is side assignment, and the mode of last product summation calculates the PR value of nodes.
The method provided in an embodiment of the present invention action trail comprehensive at operator in view of user, including call are detailed
It is single, to summarize flow, credit, arrearage record, value-added service total income, terminal purchase machine money, terminal purchase cost, family the moon wide
The personal user's data for most having break-up value are imported into user's value by volume of data cleaning by band income data
In evaluation system, the tightness degree of user's connection is measured with the frequency conversed between user and duration, with user's credit,
Flow service condition, value added data business consumption, terminal purchase machine situation, home broadband service condition are used to measure in network
The initial value of family node, while the initial value of user is influenced by user's history arrearage record.Pass through between each evaluation index
Logical relation is combined using linear algebra and statistical method, forms user's initial value assessment result.By user it
Between degree of being completely embedded and initial value weight imported into the network topology constructed with calling and called relationship, pass through
WPRQC algorithm iteration calculates the final value of user.The PageRank algorithm for optimizing Spark machine learning library, is restraining
Parameter regulation, model result tuning have a convenient and fast operating method, model result to client's child care of related fields, precision marketing,
Market competition has positive directive significance.
The prior art is made using BOSS charge system data as analysis foundation, or with the Internet data that the domain O DPI is parsed
For basic data, lack the parsing to the comprehensive behavior of user;And lacking necessary data prediction step, group customer is this kind of
Special user group does not distinguish processing with ordinary user group, causes value assessment result that will necessarily tend to group customer;It is existing
Analytical plan has ignored the incidence relation between user, the user in social networks link is detached into independent of each other
Individual measures the value of user only with the computing rule of linear statistical, has ignored the valence of user's cluster under big data scale
It is worth effect.In addition, data are rule of thumb artificially carried out section, lacks the support of mathematical reasoning, influence value assessment
Precision.
The embodiment of the present invention also provides a kind of information processing unit 20, as shown in figure 4, described device 20 includes: transceiver
201 and processor 202, wherein
The transceiver 201, for obtaining user data;
The processor 202 obtains pending data for carrying out pretreatment to the user data of acquisition, described to be processed
Data include: user's communication bill data;It is also used to calculate user's interaction coefficient according to the pending data and user is initial
Weight;It is also used to calculate user's value assessment result according to user's interaction coefficient and user's initial weight.
Further, the processor 202, specifically for rejecting the number of the group user mark in the user data
According to mending the null value in the user data is 0, and the abnormal Value Data in the user data is substituted with mean value, respectively will
Flux unit and income unit in the user data carry out unification, obtain pending data;
Wherein, the pending data includes: call detailed list of the user identifier to be assessed in the first preset time, described
Summarize the moon of the user identifier to be assessed before the time to be assessed in second preset time using flow, the credit amount of money, increase
The history arrearage amount of money and arrearage number, history of value business total income and home broadband income and the user identifier to be assessed
Terminal purchases machine money and history terminal purchase cost.
Further, the processor 202, is specifically used for:
According in the pending data in call detailed list the duration of call and talk times calculate user interaction system
Number;
It is wide using flow, the credit amount of money, value-added service total income, family according to summarizing the moon in the pending data
Band income, the history arrearage amount of money and arrearage number, history terminal purchase machine money and history terminal purchase cost calculate the user
Initial weight.
Further, the processor 202, is specifically used for:
User's interaction coefficient Interaction_Degree (i, j) is calculated according to the following formula,
Wherein, n indicates the talk times between user i and user j, calltime(i,j)Indicate call detailed list in user i and
The duration of call between user j.
Further, the processor 202, is specifically used for:
User initial weight Initial_weight (i) is calculated according to following formula,
Wherein, n indicates the number of the history terminal purchase machine of user i, and terminal, which purchases machine money (i, j), indicates user i jth time purchase
Terminal pays the expense of operator, and terminal purchases the terminal payment that machine cost (i, j) indicates that operator is user i jth time purchase
Cost, summarizing flow (i) moon indicates the summarizing the moon using flow of user i, and the credit amount of money (i) indicates prestoring for user i
The telephone expenses amount of money, value-added service total income (i) indicate the value-added service total income of user i, and (i) is taken in broadband indicates the family of user i
Front yard broadband income, m indicate that the history arrearage number of user i, the arrearage amount of money (i, k) indicate that the amount of money of user i kth time arrearage, e are
Constant.
Further, the processor 202, is specifically used for:
The value assessment result Value (i) of user i is calculated according to the following formula,
Wherein, Interaction_Degree (i, j) is user's interaction coefficient of user i, and Initial_weight (i) is
User's initial weight of user i.
The embodiment of the present invention also provides a kind of information processing unit, and the system comprises processors and memory;Wherein,
The memory, for storing the computer program that can be run on the processor;
The processor executes the step of the as above any one information processing method when for running the computer program
Suddenly.
Specifically, the understanding of information processing unit provided in an embodiment of the present invention can be real with reference to above- mentioned information processing method
The explanation of example is applied, details are not described herein for the embodiment of the present invention.
The embodiment of the present invention also provides a kind of computer readable storage medium, stores in the computer readable storage medium
There is computer program, as above described in any item information processing methods are realized when which is executed by processor.The meter
Calculation machine readable storage medium storing program for executing includes effumability random access memory (RAM), read-only memory (ROM), electric erazable programmable
Read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM (CD-ROM), digital versatile disc (DVD) or its
His medium that he is accessed.
By implementing technical solution provided in an embodiment of the present invention, have the advantages that
(1) the value relevance for having comprehensively considered user group, using based on the figure for improving PageRank webpage sorting
Calculation method constructs network using the relationship of call as the side of connection network node using personal user as the node in social networks
Link structure.The input field extracted is subjected to index factor analysis with the method for big data statistical analysis, as net
The initial weight of network interior joint, using between user, call frequency, the duration of call are as the weight on side between network node, by repeatedly
In generation, calculates, and obtains the final value of user;
(2) data handling procedure is comprehensive, it is contemplated that influence of the special user group to value assessment result: in data prediction
Stage guarantees by whether group customer attribute field, filters out input main body of the personal user as model analysis of full dose
The objectivity and universality of model application;
(3) mode input index is analyzed and is excavated using educible mathematical statistics method, reduced to model
Human intervention improves the accuracy of value assessment: in index factor analytic process, using continuous data calculation method, reducing
The accuracy of user's value calculation is improved to exclude artificial division data set as far as possible and cause result distortion in section.
(4) modeling data is more comprehensive, can portray the communication behavior track of user, the call detailed list including user, into
And summarize and summarize flow, the moon duration of call moon out, also include credit, the arrearage record, value-added service in charge system
Total income, terminal purchase machine money, terminal purchase cost, home broadband income data.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer program
Product.Therefore, the shape of hardware embodiment, software implementation or embodiment combining software and hardware aspects can be used in the present invention
Formula.Moreover, the present invention, which can be used, can use storage in the computer that one or more wherein includes computer usable program code
The form for the computer program product implemented on medium (including but not limited to magnetic disk storage and optical memory etc.).
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
The foregoing is only a preferred embodiment of the present invention, is not intended to limit the scope of the present invention.
Claims (11)
1. a kind of information processing method characterized by comprising
Pretreatment is carried out to the user data of acquisition and obtains pending data, the pending data includes: user's communication bill
Data;
User's interaction coefficient and user's initial weight are calculated according to the pending data;
User's value assessment result is calculated according to user's interaction coefficient and user's initial weight.
2. the method according to claim 1, wherein described pair acquisition user data carry out pretreatment obtain to
Handle data, comprising:
User data is obtained, the data of the group user mark in the user data are rejected, by the sky in the user data
Value complement is 0, and the abnormal Value Data in the user data is substituted with mean value, respectively by the flux unit in the user data
Unification is carried out with income unit, obtains pending data;
Wherein, the pending data includes: call detailed list of the user identifier to be assessed in the first preset time, described to be evaluated
Estimate and summarizes the moon of the user identifier before the time to be assessed in second preset time using flow, the credit amount of money, increment industry
Total income of being engaged in and home broadband income and the history arrearage amount of money and arrearage number of the user identifier to be assessed, history terminal
Purchase machine money and history terminal purchase cost.
3. according to the method described in claim 2, it is characterized in that, described calculate user's interaction system according to the pending data
Several and user's initial weight, comprising:
According to the duration of call and talk times calculating user's interaction coefficient in the pending data in call detailed list;
It is received according to summarizing the moon in the pending data using flow, the credit amount of money, value-added service total income, home broadband
Enter, the history arrearage amount of money and arrearage number, that history terminal purchase machine money and history terminal purchase cost calculate the user is initial
Weight.
4. according to the method described in claim 3, it is characterized in that, described according in the pending data in call detailed list
The duration of call and talk times calculate user's interaction coefficient, comprising:
User's interaction coefficient Interaction_Degree (i, j) is calculated according to the following formula,
Wherein, n indicates the talk times between user i and user j, calltime(i,j)Indicate user i and user j in call detailed list
Between the duration of call.
5. according to the method described in claim 3, it is characterized in that, it is described according to summarize the moon in the pending data flow,
The credit amount of money, value-added service total income, home broadband income, the history arrearage amount of money and arrearage number, history terminal purchase machine
Money and history terminal purchase cost calculate user's initial weight, comprising:
User initial weight Initial_weight (i) is calculated according to following formula,
Wherein, n indicates the number of the history terminal purchase machine of user i, and terminal, which purchases machine money (i, j), indicates user i jth time purchase terminal
Pay the expense of operator, terminal purchase the terminal payment that machine cost (i, j) indicates that operator is user i jth time purchase at
This, summarizing flow (i) moon indicates the summarizing the moon using flow of user i, and the credit amount of money (i) indicates the credit of user i
The amount of money, value-added service total income (i) indicate the value-added service total income of user i, and (i) is taken in broadband indicates that the family of user i is wide
Band income, m indicate that the history arrearage number of user i, the arrearage amount of money (i, k) indicate the amount of money of user i kth time arrearage, and e is normal
Number.
6. according to the method described in claim 5, it is characterized in that, described according at the beginning of user's interaction coefficient and the user
Beginning weight calculation user value assessment result, comprising:
The value assessment result Value (i) of user i is calculated according to the following formula,
Wherein, Interaction_Degree (i, j) is user's interaction coefficient of user i, and Initial_weight (i) is user
User's initial weight of i.
7. a kind of information processing unit, which is characterized in that described device includes: transceiver and processor, wherein
The transceiver, for obtaining user data;
The processor obtains pending data, the pending data packet for carrying out pretreatment to the user data of acquisition
It includes: user's communication bill data;It is also used to calculate user's interaction coefficient and user's initial weight according to the pending data;Also
For calculating user's value assessment result according to user's interaction coefficient and user's initial weight.
8. device according to claim 7, which is characterized in that the processor is specifically used for rejecting the user data
In group user mark data, by the user data null value mend be 0, by the exceptional value number in the user data
Mean value substitutes accordingly, and the flux unit in the user data is carried out unification with income unit respectively, obtains pending data;
Wherein, the pending data includes: call detailed list of the user identifier to be assessed in the first preset time, described to be evaluated
Estimate and summarizes the moon of the user identifier before the time to be assessed in second preset time using flow, the credit amount of money, increment industry
Total income of being engaged in and home broadband income and the history arrearage amount of money and arrearage number of the user identifier to be assessed, history terminal
Purchase machine money and history terminal purchase cost.
9. device according to claim 8, which is characterized in that the processor is specifically used for:
According to the duration of call and talk times calculating user's interaction coefficient in the pending data in call detailed list;
It is received according to summarizing the moon in the pending data using flow, the credit amount of money, value-added service total income, home broadband
Enter, the history arrearage amount of money and arrearage number, that history terminal purchase machine money and history terminal purchase cost calculate the user is initial
Weight.
10. a kind of computer readable storage medium, which is characterized in that be stored with computer in the computer readable storage medium
Program, the computer program realize information processing method as claimed in any one of claims 1 to 6 when being executed by processor.
11. a kind of information processing unit, which is characterized in that the system comprises processors and memory;Wherein,
The memory, for storing the computer program that can be run on the processor;
The processor, perform claim requires any one of 1 to 6 information processing method when for running the computer program
The step of.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711472693.4A CN109993380A (en) | 2017-12-29 | 2017-12-29 | A kind of information processing method, device and computer readable storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711472693.4A CN109993380A (en) | 2017-12-29 | 2017-12-29 | A kind of information processing method, device and computer readable storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109993380A true CN109993380A (en) | 2019-07-09 |
Family
ID=67108470
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711472693.4A Pending CN109993380A (en) | 2017-12-29 | 2017-12-29 | A kind of information processing method, device and computer readable storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109993380A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110596595A (en) * | 2019-09-25 | 2019-12-20 | 广东毓秀科技有限公司 | Method for predicting RUL of rail-traffic lithium battery through big data |
CN111915156A (en) * | 2020-07-14 | 2020-11-10 | 中国联合网络通信集团有限公司 | Service pushing method based on user value, electronic equipment and storage medium |
CN113407827A (en) * | 2021-06-11 | 2021-09-17 | 广州三七极创网络科技有限公司 | Information recommendation method, device, equipment and medium based on user value classification |
CN113591018A (en) * | 2021-07-30 | 2021-11-02 | 中国联合网络通信集团有限公司 | Communication client classification management method, system, electronic device and storage medium |
CN113727332A (en) * | 2021-08-12 | 2021-11-30 | 中国联合网络通信集团有限公司 | Method and device for recovering internet surfing |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103024207A (en) * | 2011-09-26 | 2013-04-03 | 中国电信股份有限公司 | Method and system for query processing of post-paid package expense and service decomposition server |
-
2017
- 2017-12-29 CN CN201711472693.4A patent/CN109993380A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103024207A (en) * | 2011-09-26 | 2013-04-03 | 中国电信股份有限公司 | Method and system for query processing of post-paid package expense and service decomposition server |
Non-Patent Citations (1)
Title |
---|
迟准: ""电信运营企业客户流失预测与评价研究"", 《中国博士学位论文全文数据库 经济与管理科学辑》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110596595A (en) * | 2019-09-25 | 2019-12-20 | 广东毓秀科技有限公司 | Method for predicting RUL of rail-traffic lithium battery through big data |
CN111915156A (en) * | 2020-07-14 | 2020-11-10 | 中国联合网络通信集团有限公司 | Service pushing method based on user value, electronic equipment and storage medium |
CN111915156B (en) * | 2020-07-14 | 2024-02-27 | 中国联合网络通信集团有限公司 | Service pushing method based on user value, electronic equipment and storage medium |
CN113407827A (en) * | 2021-06-11 | 2021-09-17 | 广州三七极创网络科技有限公司 | Information recommendation method, device, equipment and medium based on user value classification |
CN113591018A (en) * | 2021-07-30 | 2021-11-02 | 中国联合网络通信集团有限公司 | Communication client classification management method, system, electronic device and storage medium |
CN113727332A (en) * | 2021-08-12 | 2021-11-30 | 中国联合网络通信集团有限公司 | Method and device for recovering internet surfing |
CN113727332B (en) * | 2021-08-12 | 2022-09-02 | 中国联合网络通信集团有限公司 | Method and device for recovering internet surfing |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109993380A (en) | A kind of information processing method, device and computer readable storage medium | |
Han et al. | Segmentation of telecom customers based on customer value by decision tree model | |
Dobson et al. | Heuristics for pricing and positioning a product-line using conjoint and cost data | |
EP2506207A1 (en) | Computer system, computer-implemented method and computer program product for managing operations of a system | |
CN109325640A (en) | User's Value Prediction Methods, device, storage medium and equipment | |
CN107103548A (en) | The monitoring method and system and risk monitoring and control method and system of network behavior data | |
CN112734559B (en) | Enterprise credit risk evaluation method and device and electronic equipment | |
CN109859052A (en) | A kind of intelligent recommendation method, apparatus, storage medium and the server of investment tactics | |
CN106203773A (en) | A kind of method and device of automatic management employee | |
CN106970953B (en) | Network construction management-control method and its platform based on big data analysis | |
CN108388974A (en) | Top-tier customer Optimum Identification Method and device based on random forest and decision tree | |
CN103310353B (en) | The data filtering of a kind of attack resistance optimizes system and method | |
Dullaghan et al. | Integration of machine learning techniques to evaluate dynamic customer segmentation analysis for mobile customers | |
CN111915156B (en) | Service pushing method based on user value, electronic equipment and storage medium | |
CN110097455B (en) | Bank customer optimal configuration method and device | |
CN106408325A (en) | User consumption behavior prediction analysis method based on user payment information and system | |
CN106991577A (en) | A kind of method and device for determining targeted customer | |
KR20170049328A (en) | Apparatus and method for evaluating a value of technology or intellectual property | |
CN110415106A (en) | A kind of performance analysis forecast assessment system of management body | |
CN115147144A (en) | Data processing method and electronic equipment | |
Noorizadeh et al. | Suppliers ranking in the presence of undesirable outputs | |
CN109858947A (en) | Retail user value analysis system and method | |
Dockner et al. | Pricing constant maturity floaters with embeeded options using Monte Carlo simulation | |
KR101856550B1 (en) | Host country evaluation method and system for international construction enterprises | |
CN114119107A (en) | Steel trade enterprise transaction evaluation method, device, equipment and storage medium |
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: 20190709 |
|
RJ01 | Rejection of invention patent application after publication |