CN106599249A - Method for carrying out micro-blog forwarding prediction based on cluster gravitation modeling - Google Patents
Method for carrying out micro-blog forwarding prediction based on cluster gravitation modeling Download PDFInfo
- Publication number
- CN106599249A CN106599249A CN201611184741.5A CN201611184741A CN106599249A CN 106599249 A CN106599249 A CN 106599249A CN 201611184741 A CN201611184741 A CN 201611184741A CN 106599249 A CN106599249 A CN 106599249A
- Authority
- CN
- China
- Prior art keywords
- node
- microblogging
- forwarding
- weight
- gravitation
- 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
- 238000000034 method Methods 0.000 title claims abstract description 21
- 230000000694 effects Effects 0.000 claims description 7
- 238000012360 testing method Methods 0.000 claims description 3
- 238000005065 mining Methods 0.000 abstract 1
- 238000012546 transfer Methods 0.000 description 12
- 230000005484 gravity Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001717 pathogenic effect Effects 0.000 description 1
- 230000001902 propagating effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; 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
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Engineering & Computer Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Databases & Information Systems (AREA)
- Game Theory and Decision Science (AREA)
- Computing Systems (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 invention discloses a method for carrying out micro-blog forwarding prediction based on cluster gravitation modeling. The method comprises the following steps of carrying out deep mining on a corresponding user relationship and building a micro-blog user relationship network; calculating a weight of each micro-blog user according to the micro-blog user relationship and calculating a weight of each edge by using a statistical approach; and carrying out micro-blog forwarding prediction on a to-be-detected micro-blog based on a cluster gravitation model. A node relation between any two points is built through the cluster gravitation model, and meanwhile, the forwarding probabilities of different nodes are described by using the weights, so that the local prediction accuracy is improved. Meanwhile, the forwarding conditions of the Kth batch of followers can be predicted through the cluster gravitation model with the weights; and compared with the scheme of predicting the forwarding conditions of the Kth batch of followers through one-by-one and layer-by-layer iterative calculation, the method has the advantage that the prediction efficiency is improved.
Description
Technical field
The present invention relates to microblogging prediction field, specifically a kind of side that microblogging forwarding prediction is carried out based on the modeling of class gravitation
Method.
Background technology
Microblogging is the social platform that a kind of real time information based on customer relationship is exchanged, shared, propagating, with Facebook,
The social networkies such as Twitter equally have impact on the life exchange way of the mankind.In microblog, as number of users is with hundreds of millions
What the behind of the magnanimity informations such as increase of the level for unit quantity, substantial amounts of picture, text was reflected is life idea, the knowledge of people
With interesting thing.The appearance of microblogging also brings many problems except producing beneficial effect, such as bad speech without constraint
The heavy damage social life pathogenic wind such as propagate.So, the active state of microblog users is predicted, for government, enterprise's thing
Industry unit, individual have important meaning.
In existing microblogging forwarding prediction solution, the Chinese patent of Publication No. CN103984701A discloses one kind
Microblogging transfer amount forecast model generation method and microblogging transfer amount Forecasting Methodology.Microblogging transfer amount forecast model generation method bag
Include：Training data is obtained, training data includes the microblogging of a plurality of known transfer amount；With the transfer amount of microblogging as foundation, by microblogging
It is divided into more than 3 transfer amount classifications；Extract the basic feature of every microblogging；Set up many between basic feature and transfer amount classification
Disaggregated model；For each transfer amount classification, the regression model set up between basic feature and microblogging transfer amount.Microblogging is forwarded
Amount Forecasting Methodology includes：Extract the basic feature of microblogging to be predicted；According to many disaggregated models and basic feature, judge to be predicted micro-
Win affiliated transfer amount classification；Obtain the corresponding regression model of transfer amount classification；According to regression model and basic feature, prediction is treated
The transfer amount of prediction microblogging.Extract simple using the method feature of the offer of the present invention and be adapted to used in large-scale data.
The patent is in identical field with the present invention, but the problem for solving is different, and existing solution can not solve the present invention
Problem to be solved.
For the microblogging forwarding network of personal connections of various grades of layer relation, level layer relation can only be calculated to predict certain by successive iteration
Plant the forwarding situation of specific level layer, it is impossible to which arbitrarily prediction K criticizes the forwarding situation of follower, predictive efficiency is relatively low.
The content of the invention
It is an object of the invention to overcome the deficiencies in the prior art, there is provided a kind of to carry out microblogging forwarding based on the modeling of class gravitation
The method of prediction, at least to realize arbitrarily predicting the forwarding situation of K crowd of microblogging follower, improve the effect of predictive efficiency.
The purpose of the present invention is achieved through the following technical solutions：It is a kind of that microblogging forwarding is carried out based on the modeling of class gravitation
The method of prediction, it comprises the following steps：
S1：According to the forwarding relation between microblog users, directed graph network is set up based on directed edge；
S2：Node number M in statistics directed graph network, and give the weight of each node 1/M；
S3：Count the total N that the microblogging that microblog users to be measured issue is forwarded, and microblog users to be measured it is corresponding each
Quantity n forwarded by follower1,n2,n3…ni, the initial weight for calculating the corresponding every directed edge of each follower is：
S4：User first with issuing microblog arrives the weight of present node as node according to the weight distribution of directed edge
Pay close attention on all nodes of the node, to the weight for updating each node for paying close attention to the node；
S5：According to the weight for updating the later corresponding directed edge of node weights calculating；
S6：Bad execution S5～S6 steps are followed, to the weight for updating each node, until the weight of each node restrains；
S7：Node weights k of the K degree followers of microblog users to be measured are obtained as needed1,k2,…kn；
S8：Microblog users to be measured are calculated to the gravitational index of the K degree follower for selecting：
Wherein, M is the node weights of microblog users to be measured, and m is the node weights of a selected K degree followers, and r is M
To the inverse of the weight sum of all directed edges of a route of m, G is set according to actual needs；
S9：Threshold value Q1 is set as needed, judges whether gravitational index F exceedes threshold value Q1, if it exceeds Q1, then
Retain F, if not above Q1, removing F, then carry out the forwarding prediction in the stage；If gravitational index F is no more than the threshold
Value Q1, then can not carry out the forwarding prediction of K degree.
Directed edge in step S1 is the unidirectional side that follower is pointed to by the person of being concerned.
Node in step S2 is the follower for being related to forward microblogging, and node number M is to be related to forward microblogging
Follower number.
Described K degree follower is the user of K batch of concern forwarding microblogging, and K crowd of user is by paying close attention to K-1 crowd of user
It is concerned about the forwarding microblogging.
In step S4, it is micro- that the microblogging number issued according to the person of being concerned of follower's forwarding accounts for that the person of being concerned is forwarded
The ratio for winning sum is allocated weight.
By setting threshold value Q2 in step S6, judge the rate of change of weight of each node whether less than the threshold
Value Q2, if so, then stops iteration, otherwise continues iteration until the weight of each node restrains.
, from the beginning of a setting value, the prediction effect that passes a test is until finding the G-value under optimum prediction effect for described G.
The invention has the beneficial effects as follows：The present invention forwards relation directed graph network by setting up microblogging, counts all nodes
Number after giving each node identical initial weight, it is corresponding with the calculating of the forwarding situation of node based on identical initial weight
Each edge initial weight, update all node weights using the initial weight and corresponding node weights of each edge, so
The node contacts set up between any two points by class gravity model afterwards, while portraying difference using the node weights with weight
Node forwarding probability, improve the precision of local prediction, while by can be arbitrarily pre- with the class gravity model of weight
The forwarding situation of K crowd followers is surveyed, is compared by the iterative calculation of an a level layer then level layer predicting K batch of concern
The forwarding situation of person, improves predictive efficiency.
Description of the drawings
Fig. 1 is the flow chart of the present invention.
Specific embodiment
Technical scheme is described in further detail below in conjunction with the accompanying drawings, but protection scope of the present invention is not limited to
Described below.
A kind of method for carrying out microblogging forwarding prediction based on the modeling of class gravitation, it comprises the following steps：
S1：According to the forwarding relation between microblog users, directed graph network is set up based on directed edge；
S2：Node number M in statistics directed graph network, and give the weight of each node 1/M；
S3：Count the total N that the microblogging that microblog users to be measured issue is forwarded, and microblog users to be measured it is corresponding each
Quantity n forwarded by follower1,n2,n3…ni, the initial weight for calculating the corresponding every directed edge of each follower is：
S4：User first with issuing microblog arrives the weight of present node as node according to the weight distribution of directed edge
Pay close attention on all nodes of the node, to the weight for updating each node for paying close attention to the node；
S5：According to the weight for updating the later corresponding directed edge of node weights calculating；
S6：Bad execution S5～S6 steps are followed, to the weight for updating each node, until the weight of each node restrains；
S7：Node weights k of the K degree followers of microblog users to be measured are obtained as needed1,k2,…kn；
S8：Microblog users to be measured are calculated to the gravitational index of the K degree follower for selecting：
Wherein, M is the node weights of microblog users to be measured, and m is the node weights of a selected K degree followers, and r is M
To the inverse of the weight sum of all directed edges of a route of m, G is set according to actual needs；
S9：Threshold value Q1 is set as needed, judges whether gravitational index F exceedes threshold value Q1, if it exceeds Q1, then
Retain F, if not above Q1, removing F, then carry out the forwarding prediction in the stage；If gravitational index F is no more than the threshold
Value Q1, then can not carry out the forwarding prediction of K degree.
Directed edge in step S1 is the unidirectional side that follower is pointed to by the person of being concerned.
Node in step S2 is the follower for being related to forward microblogging, and node number M is to be related to forward microblogging
Follower number.
Described K degree follower is the user of K batch of concern forwarding microblogging, and K crowd of user is by paying close attention to K-1 crowd of user
It is concerned about the forwarding microblogging.
In step S4, it is micro- that the microblogging number issued according to the person of being concerned of follower's forwarding accounts for that the person of being concerned is forwarded
The ratio for winning sum is allocated weight.
By setting threshold value Q2 in step S6, judge the rate of change of weight of each node whether less than the threshold
Value Q2, if so, then stops iteration, otherwise continues iteration until the weight of each node restrains.
, from the beginning of a setting value, the prediction effect that passes a test is until finding the G-value under optimum prediction effect for described G.
Gravity factor G is used to distinguish different domain predictions, improves generalization ability, for example：G starts value from -50, always
To 50 until finding optimal G, if this optimum G=25, this G be determined as calculate gravitational index F G.
The above is only the preferred embodiment of the present invention, it should be understood that the present invention is not limited to described herein
Form, is not to be taken as the exclusion to other embodiment, and can be used for various other combinations, modification and environment, and can be at this
In the text contemplated scope, it is modified by the technology or knowledge of above-mentioned teaching or association area.And those skilled in the art are entered
Capable change and change, then all should be in the protection domains of claims of the present invention without departing from the spirit and scope of the present invention
It is interior.
Claims (7)
1. it is a kind of that the method that microblogging forwarding is predicted is carried out based on the modeling of class gravitation, it is characterised in that it comprises the following steps：
S1：According to the forwarding relation between microblog users, directed graph network is set up based on directed edge；
S2：Node number M in statistics directed graph network, and give the weight of each node 1/M；
S3：Count the total N that the microblogging of microblog users issue to be measured is forwarded, and corresponding each concern of microblog users to be measured
Quantity n forwarded by person1,n2,n3…ni, the initial weight for calculating the corresponding every directed edge of each follower is：
S4：User first with issuing microblog as node, the weight of present node according to the weight distribution of directed edge to concern
On all nodes of the node, to the weight for updating each node for paying close attention to the node；
S5：According to the weight for updating the later corresponding directed edge of node weights calculating；
S6：Bad execution S5～S6 steps are followed, to the weight for updating each node, until the weight of each node restrains；
S7：Node weights k of the K degree followers of microblog users to be measured are obtained as needed1,k2,…kn；
S8：Microblog users to be measured are calculated to the gravitational index of the K degree follower for selecting：
Wherein, M is the node weights of microblog users to be measured, and m is the node weights of a selected K degree followers, and r is M to m's
The inverse of the weight sum of all directed edges of one route, G are set according to actual needs；
S9：Threshold value Q1 is set as needed, judges whether gravitational index F exceedes threshold value Q1, if it exceeds Q1, then retain
F, if not above Q1, removing F, then carries out the forwarding prediction in the stage；If gravitational index F is no more than the threshold value
Q1, then can not carry out the forwarding prediction of K degree.
2. it is according to claim 1 it is a kind of based on class gravitation modeling carry out microblogging forwarding prediction method, it is characterised in that：
Directed edge in step S1 is the unidirectional side that follower is pointed to by the person of being concerned.
3. it is according to claim 1 it is a kind of based on class gravitation modeling carry out microblogging forwarding prediction method, it is characterised in that：
Node in step S2 is the follower for being related to forward microblogging, and node number M is the follower for being related to forward microblogging
Number.
4. it is according to claim 1 it is a kind of based on class gravitation modeling carry out microblogging forwarding prediction method, it is characterised in that：
Described K degree follower is K batch and pays close attention to the user for forwarding microblogging, and K crowd of user is concerned about this by paying close attention to K-1 crowd of user
Forwarding microblogging.
5. it is according to claim 1 it is a kind of based on class gravitation modeling carry out microblogging forwarding prediction method, it is characterised in that：
In step S5, the microblogging number issued according to the person of being concerned of follower's forwarding accounts for the microblogging sum that the person of being concerned is forwarded
Ratio is allocated weight.
6. it is according to claim 1 it is a kind of based on class gravitation modeling carry out microblogging forwarding prediction method, it is characterised in that：
By setting threshold value Q2 in step S6, whether the rate of change of weight of each node is judged less than threshold value Q2, if
It is then to stop iteration, otherwise continues iteration until the weight of each node restrains.
7. it is according to claim 1 it is a kind of based on class gravitation modeling carry out microblogging forwarding prediction method, it is characterised in that：
, from the beginning of a setting value, the prediction effect that passes a test is until finding the G-value under optimum prediction effect for described G.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611184741.5A CN106599249A (en) | 2016-12-20 | 2016-12-20 | Method for carrying out micro-blog forwarding prediction based on cluster gravitation modeling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611184741.5A CN106599249A (en) | 2016-12-20 | 2016-12-20 | Method for carrying out micro-blog forwarding prediction based on cluster gravitation modeling |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106599249A true CN106599249A (en) | 2017-04-26 |
Family
ID=58600310
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611184741.5A Pending CN106599249A (en) | 2016-12-20 | 2016-12-20 | Method for carrying out micro-blog forwarding prediction based on cluster gravitation modeling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106599249A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110222297A (en) * | 2019-06-19 | 2019-09-10 | 武汉斗鱼网络科技有限公司 | A kind of recognition methods of tagging user and relevant device |
-
2016
- 2016-12-20 CN CN201611184741.5A patent/CN106599249A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110222297A (en) * | 2019-06-19 | 2019-09-10 | 武汉斗鱼网络科技有限公司 | A kind of recognition methods of tagging user and relevant device |
| CN110222297B (en) * | 2019-06-19 | 2021-07-23 | 武汉斗鱼网络科技有限公司 | Identification method of tag user and related equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| He et al. | Mining transition rules of cellular automata for simulating urban expansion by using the deep learning techniques | |
| Lin et al. | Storm surge return levels induced by mid-to-late-twenty-first-century extratropical cyclones in the Northeastern United States | |
| Cigizoglu | Application of generalized regression neural networks to intermittent flow forecasting and estimation | |
| CN109886444A (en) | A kind of traffic passenger flow forecasting, device, equipment and storage medium in short-term | |
| Minglei et al. | Classified real-time flood forecasting by coupling fuzzy clustering and neural network | |
| CN110263280A (en) | A kind of dynamic link predetermined depth model and application based on multiple view | |
| Li et al. | Identifying explicit formulation of operating rules for multi-reservoir systems using genetic programming | |
| Xu et al. | A multiobjective short‐term optimal operation model for a cascade system of reservoirs considering the impact on long‐term energy production | |
| CN106448151A (en) | Short-time traffic flow prediction method | |
| CN105096614A (en) | Newly established crossing traffic flow prediction method based on generating type deep belief network | |
| CN102567391A (en) | Method and device for building classification forecasting mixed model | |
| CN106126615A (en) | The method and system that a kind of point of interest is recommended | |
| CN108564228A (en) | A method of based on the temporal aspect predicted orbit traffic OD volumes of the flow of passengers | |
| Hemati et al. | Water allocation using game theory under climate change impact (case study: Zarinehrood) | |
| CN115270506B (en) | Method and system for predicting passing time of crowd ascending along stairs | |
| CN110443422B (en) | OD attraction degree-based urban rail transit OD passenger flow prediction method | |
| Nozari et al. | Simulation and optimization of control system operation and surface water allocation based on system dynamics modeling | |
| Liu et al. | Using a Bayesian probabilistic forecasting model to analyze the uncertainty in real-time dynamic control of the flood limiting water level for reservoir operation | |
| Jones | Spatial bias in LUTI models | |
| CN106599249A (en) | Method for carrying out micro-blog forwarding prediction based on cluster gravitation modeling | |
| CN107341346A (en) | A kind of hydrologic forecasting method | |
| CN106777162A (en) | A kind of high accuracy microblogging forwards Forecasting Methodology | |
| CN103838964A (en) | Social relationship network generation method and device based on artificial transportation system | |
| Chang et al. | Development of a real-time forecasting model for turbidity current arrival time to improve reservoir desilting operation | |
| Gerey et al. | Groundwater single-and multiobjective optimization using Harris Hawks and Multiobjective Billiards-inspired algorithm |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170426 |