CN105409162A - Method and device for acquiring activity level of individual users in group - Google Patents

Method and device for acquiring activity level of individual users in group Download PDF

Info

Publication number
CN105409162A
CN105409162A CN201480012240.1A CN201480012240A CN105409162A CN 105409162 A CN105409162 A CN 105409162A CN 201480012240 A CN201480012240 A CN 201480012240A CN 105409162 A CN105409162 A CN 105409162A
Authority
CN
China
Prior art keywords
group
size
message
time
response
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
Application number
CN201480012240.1A
Other languages
Chinese (zh)
Other versions
CN105409162B (en
Inventor
腾黎明
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Feinno Communication Technology Co Ltd
Original Assignee
Beijing Feinno Communication Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Feinno Communication Technology Co Ltd filed Critical Beijing Feinno Communication Technology Co Ltd
Publication of CN105409162A publication Critical patent/CN105409162A/en
Application granted granted Critical
Publication of CN105409162B publication Critical patent/CN105409162B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/02—Details
    • H04L12/16—Arrangements for providing special services to substations
    • H04L12/18—Arrangements for providing special services to substations for broadcast or conference, e.g. multicast

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Cosmetics (AREA)
  • Information Transfer Between Computers (AREA)
  • Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)

Abstract

Disclosed are a method and device for acquiring the activity level of individual users in a group. The method comprises: acquiring data of a group within an observation period, and acquiring message volume interval probability distribution and/or message volume length probability distribution of the group within the observation period according to the data of the group; creating a group response model based on the rule that the adjusting change rate of group response disturbed by an individual user is in direct proportion to the difference between a current state and an equilibrium state; acquiring a multi-step recursive calculation mode of the group response according to the group response model; and acquiring, by means of statistics, the activity level of each individual user in the group within the observation period according to the multi-step recursive calculation mode of the group response and the message volume interval probability distribution and/or message volume length probability distribution of the group within the observation period. The technical solution of the present invention can conveniently obtain, from a group, members that can activate the group.

Description

Method and device for acquiring activity level of individual users in group
The present invention relates to instant messaging field, more particularly to a kind of method and apparatus for obtaining individual consumer's liveness in group for a kind of method and apparatus technical field for obtaining individual consumer's liveness in group.Background of invention is with the correlative study on Web2.0 and developing rapidly for applying, Web2.0 more focuses on the interactivity of user, as user builds a group environment for participating in expressing, create, link up and sharing, by user's online to user's knitmesh, user is both follower, the viewer for initiating group content, is also the producer and disseminator for initiating group content.
Research to group is concentrated mainly on two aspects, is on the one hand structural research, including community network, information interchange pattern etc.；On the other hand it is active content research, is embodied in the reason for individual driving group actively generates content, this part directly affects attitude, behavior intention and agenda of the user to group.At present, for the research of instant messaging group, industry has done many researchs from individual angle, but relatively fewer for colony's Journal of Sex Research, it is impossible to which the active member of group can be transferred by being got from group.The content of the invention can transfer the active member of group the embodiments of the invention provide a kind of method and apparatus for obtaining individual consumer's liveness in group so as to easily be got from group.
In order to achieve the above object, the technical scheme that the embodiment of the present invention is used is as follows：
On the one hand, the embodiment of the present invention provides a kind of method for obtaining individual consumer's liveness in group, including：The group data in observation time is obtained, obtaining size of message interval time probability distribution or/and size of message Length Pr of the group in the observation time according to the group data is distributed；Wherein, size of message interval time is the interval time of continuous two size of messages record in group, and size of message length is to send the character length that size of message is recorded each time in group；
The rule being directly proportional in the adjustment rate of change after being disturbed by individual consumer to the difference between current state and poised state is responded based on group, group's response model is set up；
The multistep recurrence calculation pattern of group's response is obtained according to group's response model；Wherein described multistep recurrence calculation pattern refers to that the result of calculation of a period obtains group's response after multiple periods as the primary condition of subsequent time period and by period recursion；
Size of message interval time probability distribution or/and size of message Length Pr distribution of the multistep recurrence calculation pattern and the group responded according to the group in the observation time, statistics obtain each liveness of the individual consumer in the observation time in the group.
Alternatively, the probability distribution and group message amount Length Pr of the group message amount interval time of acquisition be distributed in log-log coordinate lower probability distribution main part all obey power-law distribution.
Alternatively, after size of message interval time probability distribution or/and size of message Length Pr distribution of the group in the observation time are obtained according to the group data, methods described also includes：According to the group The power exponent size of size of message interval time probability distribution or/and size of message Length Pr distribution of the group in the observation time, determines the liveness of group.
Alternatively, group's response model is：=[- 3], wherein, j is the current dt that group responds
State, is the poised state of group's response ,/For coefficient, Φ Α are the adjustment rate of change that group responds, and t is the time.
Alternatively, the multistep recurrence calculation pattern of group's response is：
yn= d - e-^ ^y^ i-^ + y^'；Wherein,ΔObservation time Τ is divided into by expression《The interval time lengths of individual equal time section, J ^ are that group responds equilibrium valve in the period, and the initial value of Λ groups response is coefficient.
Alternatively, size of message interval time probability distribution or/and size of message Length Pr of the multistep recurrence calculation pattern and the group responded according to the group in the observation time are distributed, statistics obtain it is described using the group, the size of message interval time probability distribution of initial time and/or the distribution of size of message Length Pr are inputted in the multistep recurrence calculation pattern as initial value in the observation time, obtain group's response of each period；
, just it is blunt according to the group state size of message interval time probability distribution or/and size of message length 14 in observation occupy state each period group ring, should value ^ live and the individual consumer died young and its enliven the frequency, alternatively, described using the group, the size of message interval time probability distribution of initial time and/or the distribution of size of message Length Pr include following any situation to statistics as the initial value input multistep recurrence calculation pattern in the observation time：
Any one of the group in the observation time in the size of message interval time probability distribution of initial time, the distribution of size of message Length Pr is inputted in the multistep recurrence calculation pattern as initial value；Or, using the group, the size of message interval time probability distribution of initial time is inputted in the multistep recurrence calculation pattern with size of message Length Pr distribution sum as initial value in the observation time；Or,
Using the group, the size of message interval time probability distribution of initial time is inputted in the multistep recurrence calculation pattern with the product that size of message Length Pr is distributed as initial value in the observation time.
, alternatively：^ states basis and states group's response of each period, the individual consumer that ^ living dies young and its live according to group's response of each period, active individual consumer and its enliven the frequency, work n jump data of each individual consumer in each period are obtained, the alive data includes：The self-employed worker of each period enlivens the frequency and corresponding, and group rings; 、 、、、 、 ^
, ' can ground methods described also include：, ' group data in another observation time is obtained, size of message interval time probability distribution or/and size of message Length Pr of the group in another observation time are obtained according to the group data and are distributed；The multistep recurrence calculation pattern responded according to the group and the group are described Size of message interval time probability distribution in another observation time or/and the distribution of size of message Length Pr, statistics obtain in institute, group each individual consumer in the separately " liveness in observation time.The liveness of day is integrated with the liveness at least one described another observation time, obtains the synthesis liveness of each individual consumer in the group.On the other hand, the embodiments of the invention provide a kind of device for obtaining individual consumer's liveness in group, including：
Probability distribution acquiring unit, for obtaining the group data in observation time, obtains size of message interval time probability distribution or/and size of message Length Pr of the group in the observation time according to the group data and is distributed；Wherein, size of message interval time is the interval time of continuous two size of messages record in group, and size of message length is to send the character length that size of message is recorded each time in group；
Model sets up unit, for responding the rule being directly proportional in the adjustment rate of change after being disturbed by individual consumer to the difference between current state and poised state based on group, sets up group's response model；
Recurrence calculation unit, the multistep recurrence calculation pattern for obtaining group's response according to group's response model；Wherein described multistep recurrence calculation pattern refers to that the result of calculation of a period obtains group's response after multiple periods as the primary condition of subsequent time period and by period recursion；
Individual liveness statistic unit, for the size of message interval time probability point in the sight time described in the multistep recurrence calculation pattern responded according to the group and the group or/and the distribution of size of message Length Pr, alternatively, the probability distribution and group message amount Length Pr for the group message amount interval time that the general ^ distributed acquisitions unit is obtained are distributed in the main part of log-log coordinate lower probability distribution and all obey power-law distribution.
Alternatively, described device also includes：
Group's liveness determining unit, suitable for the power exponent size being distributed according to size of message interval time probability distribution of the group in the observation time or/and size of message Length Pr, determines the liveness of group.
Alternatively, the model set up unit foundation group's response model be：
-=β ^ ~] ^ wherein, be group response current state, be group response poised state, dt
It is the adjustment rate of change of group's response, t is the time for coefficient.
Alternatively, the multistep recurrence calculation pattern of the recurrence calculation unit is obtained group response is： yn= (l - e-^' ^y^ i-^-^ + y^；Wherein, ^ represents observation time T being divided into " equilibrium valve that the interval time lengths of individual equal time section are group's response in the period,；.The initial value of group's response, is coefficient.
Alternatively, the individual liveness statistic unit includes：
Group's response computing module, for the size of message interval time probability distribution of initial time and/or the distribution of size of message Length Pr to be inputted in the multistep recurrence calculation pattern as initial value in the observation time using the group, group's response of each period is obtained；
Individual consumer's statistical module, during for according to size of message interval of the group in the observation time Between probability distribution or/and the distribution of size of message Length Pr, statistics obtains the active individual consumer of each period and its enlivens the frequency；
Individual liveness counts mould,, the work of ^ values rings in the group that ^ roots ^ states each period1^ individual liveness.
Alternatively, individual consumer's statistical module includes following appoint --- individual submodule：
First statistic submodule, for any one of the group in the observation time in the size of message interval time probability distribution of initial time, the distribution of size of message Length Pr to be inputted in the multistep recurrence calculation pattern as initial value；
Second statistic submodule, for the size of message interval time probability distribution of initial time to be inputted in the multistep recurrence calculation pattern with size of message Length Pr distribution sum as initial value in the observation time using the group；
3rd statistic submodule, for the size of message interval time probability distribution of initial time to be inputted in the multistep recurrence calculation pattern with the product that size of message Length Pr is distributed as initial value in the observation time using the group.
Alternatively, the individual liveness statistical module includes：
Alive data acquisition submodule, for group's response according to each period, active individual consumer and its enlivens the frequency, obtains alive data of each individual consumer in each period, the alive data includes：The individual consumer of each period enlivens the frequency and corresponding group's response；
Individual liveness statistic submodule, statistical analysis is done for the alive data according to certain mode respectively to each individual consumer in group, obtains each liveness of the individual consumer in the observation time in the group.
Alternatively, described device also includes：
Second probability distribution acquiring unit, for obtaining the group data in another observation time, obtains size of message interval time probability distribution or/and size of message Length Pr of the group in another observation time according to the group data and is distributed；
Second individual liveness statistic unit, for size of message interval time probability distribution in another observation time of the multistep recurrence calculation pattern and the group that are responded according to the group or/and the distribution of size of message Length Pr, statistics obtains liveness of each individual consumer in another observation time in the group.
Alternatively, it is described described another at least one
The synthesis liveness of user.The beneficial effect of the embodiment of the present invention is, by obtaining the group data in observation time, and obtaining size of message interval time probability distribution or/and size of message Length Pr of the group in the observation time according to the group data is distributed；It is then based on the rule that group's response is directly proportional in the adjustment rate of change after being disturbed by individual consumer to the difference between current state and poised state, set up group's response model, then the multistep recurrence calculation pattern of group's response is obtained further according to group's response model, the size of message interval time probability distribution in the observation time of multistep recurrence calculation pattern and the group that is responded finally according to the group or/ With the distribution of size of message Length Pr, statistics obtains each liveness of the individual consumer in the observation time in the group.Technical scheme helps to excavate the relation between individual behavior and group behavior, solution is provided for research individual consumer's active changing process, easily can be got from group can transfer the active member of group, so that the group information responded with the member's transmission disturbance group for transferring group's liveness can be utilized to encourage group to enliven according to business demand.
Described above is only the general introduction of technical solution of the present invention, in order to better understand the technological means of the present invention, and can be practiced according to the content of specification, and in order to which above and other objects of the present invention, feature and advantage can be become apparent, below especially exemplified by the embodiment of the present invention.Brief Description Of Drawings are by reading the detailed description of hereafter preferred embodiment, and various other advantages and benefit will be clear understanding for those of ordinary skill in the art.Accompanying drawing is only used for showing the purpose of preferred embodiment, and is not considered as limitation of the present invention.And in whole accompanying drawing, identical part is denoted by the same reference numerals.In the accompanying drawings：
Fig. 1 shows the method flow diagram of individual consumer's liveness in a kind of acquisition group provided according to embodiments of the present invention；
Fig. 2 show according to it is of the invention real, apply example obtained within a period of time ^^ and scheme group information signal day;, the signal of probability distribution；^, rate distribution example；
Fig. 5 shows that method schematic diagram ^, Fig. 7 of each liveness of the individual consumer in the observation time in statistics group according to embodiments of the present invention shows individual consumer's liveness apparatus structure schematic diagram in a kind of acquisition group according to embodiments of the present invention.The exemplary embodiment of the disclosure is more fully described in the mode for implementing the present invention below with reference to accompanying drawings.Although showing the exemplary embodiment of the disclosure in accompanying drawing, it being understood, however, that may be realized in various forms the disclosure without that should be limited by embodiments set forth here.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can by the scope of the present disclosure completely convey to those skilled in the art.
Referring to Fig. 1, the embodiments of the invention provide a kind of method for obtaining individual consumer's liveness in group, comprise the following steps：
Sll, the group information obtained in observation time, obtain size of message interval time probability distribution or/and size of message Length Pr of the group in the observation time according to the group information and are distributed.
Wherein, size of message interval time is the interval time of continuous two size of messages record in group, and size of message length is to send the character length that size of message is recorded each time in group.
Fig. 2 shows the group information schematic diagram of group in a period of time of acquisition.Group information includes group The essential information and observation table of group.Wherein, group's essential information include in group name, group number, build group purpose, observation period, whether voluntarily addition, group's classification, group's brief introduction and group announce etc., wherein group's classification can include one-level class and two grades of classes；Observation table includes the observation date, initiates talk time, terminate talk time, participate in dialogue number, promoter, initiate reason and chat topic etc..
Because instant messaging group is a chat collective, so individual consumer sends the heterogeneity of size of message, such as typing speed, frequency of surfing the Net, the enthusiasm of hair message can be reduced significantly, so as to which the data obtained are more representative, the individual consumer's feature that can more reflect in group.Therefore according to the group information of acquisition, in addition it is also necessary to obtain the group message amount interval time probability distribution in observation time T or/and the distribution of group message amount Length Pr.
In an embodiment of the invention, studied by probability distribution or/and size of message the Length Pr distribution of the size of message interval time obtained to different groups, it was found that uniformity is presented in their probability distribution under log-log coordinate, main part all obeys power-law distribution.
Fig. 3 is group message amount interval time probability distribution example of the group under log-log coordinate.Although each group records number is different, the individual consumer member of group is different, but uniformity is but presented in the probability distribution under its log-log coordinate, obvious fat clients are all presented, i.e. most of transmission size of message interval times are shorter, and a small amount of interval time is very long, removes sagging head and long afterbody, the main body of its probability distribution obeys power-law distribution, and power exponent is concentrated in the range of about 1.85-2.11.Such distribution characteristics is the potential rule for meeting group mutual-action behavior, group message is propagated Chang Youyi topic and triggered, one people initiates topic and causes the common participation of member in group, cause when topic occurs, disturbance is more and interval time is short, and the silence of some time is had after topic terminates.
Fig. 4 is group message amount Length Pr distribution example of the group under log-log coordinate.Group is as an immediate communication tool, and interaction has instantaneity, wherein the length of every message typically will not be long.The cumulative distribution of number of characters has certain fluctuation, and main part can use straight line approximate fits, it can be considered that the cumulative distribution of number of characters also obeys power law, power exponent is between 1.55-2.64.Span scope is larger for interval time, and its reason may is that the composition theme of each group is different, therefore the content of transmission message has been distinguished.Work together group such as group B, and wherein message is often relatively more brief, and the few message of character accounts for the larger specific gravity of total message count, therefore slope is larger.It is often longer relatively more in message and interest group A is then because the sharing contents of interesting property, so as to cause the difference on slope.
In one embodiment of the invention, the power exponent size being distributed according to size of message interval time probability distribution of the group in the observation time or/and size of message Length Pr, determines the liveness of group.
Determine the mode of the liveness of group, can be determined according only to one of power exponent size of power exponent size or size of message the Length Pr distribution of size of message interval time probability distribution of the group in the observation time, can also according to the two the big d of power exponent, and/or product determine.
In general, the liveness of power exponent Yue great groups is higher, therefore the power exponent size that can be distributed according to the size of message interval time probability distribution or/and size of message Length Pr of group, the liveness of different groups is ranked up.Sorted according to the liveness of different groups, preferably go out a number of group and studied.
S12, the rule being directly proportional based on group's response in the adjustment rate of change after being disturbed by individual consumer to the difference between current state and poised state, set up group's response model.
Group's response model of foundation, can be described with One first-order ordinary differential equation： ( i )
at
Wherein, j is the current state that group responds,ΛThe poised state responded for group, is the adjustment rate of change of group's response, t is time, 0 t observation times T;It is essentially to change over time for coefficient, but in order to solve convenient here it is assumed that being constant.
Above formula is the basic model that individual consumer disturbs group's response, and can change with time process for the adjustment of description group and group's balance, and group's response model has universal applicability.
S13, the multistep recurrence calculation pattern for obtaining according to group's response model group's response.
Wherein, the multistep recurrence calculation pattern refers to that the result of calculation of a period obtains group's response after multiple periods as the primary condition of subsequent time period and by period recursion.
Solve above-mentioned One first-order ordinary differential equation（1) process is as follows：
Due to ^ bring into it is more complicated after concrete numerical value, can be by formula for the ease of solving（1) it is rewritten as following dt
General type, obtain：
^ y= ye (2) at
Obvious formula（2) it is First Order Nonhomogeneous linear equation, its general solution is：
y = e-^yee-^dt + C1] (3) wherein, be integral constant in formula.
Will= 0When, j= JQSubstitute into above-mentioned formula（3) formula, is obtained（2) particular solution： y = y0e^+e-^yee^dt] (4)
0
In formula, y responds for group, is the equilibrium valve of group's response； y.It is primary condition or initial value, t is the time；/For coefficient.In principle/It can change over time, it is assumed that be constant.
To formula（4) thus the integral term direct solution on the right of obtains
y = νβ + (ν0 - ν -βί(5) as t=0y = y。
Y=y as t=∞e
Or
y = i\ - e-pt)ye+ (6) parse formula：
Due to the active result for group's adjusting stage of individual consumer, regardless of whether will all be used as the primary condition of next period for poised state, group is responded and produces influence and so that the size of message length and interval time condition of early stage respond on the group in later stage and produces influence, so as to, and the responsive state value after multiple periods just can be obtained by period recursion using the result of calculation of a upper period as the primary condition of subsequent period.
Therefore, observation time T is divided into n period, each period is designated as At, formula（6) it can be designated as in first period
y, ={\-e ^)ye,+y0e ^ (7) Wherein, it is the equilibrium valve of the group within first period.
Accordingly, the 2nd period equally has
Merge（7) and（8) obtain
y2 = \ - e-^)[ye2 + y^] + y0Thus recursion was obtained to n-th of period e-^ (9)
=(1-e-') (" -)+" (1 0) formula（1 0) be referred to as group's response multistep recurrence calculation pattern, wherein, ^ is the equilibrium valve of the group within i-th of period, and At represents observation time T being divided into " the individual impartial period,；.The initial value of group's response.
The size of message interval time probability distribution or/and size of message Length Pr of S 14, the multistep recurrence calculation pattern and the group responded according to the group in the observation time are distributed, and statistics obtains each liveness of the individual consumer in the observation time in the group.
In one embodiment of the invention, referring to Fig. 5, step S 14 is specifically included：
S51, the size of message interval time ^ of group initial time in the observation time both rate distributions and/or the distribution of size of message Length Pr inputted in the multistep recurrence calculation pattern as initial value, obtain group's response of each period.
A kind of embodiment, can be only by the size of message interval time probability distribution W of group initial time t=0 in the observation time.Or size of message Length Pr distribution R.Inputted as initial value in the multistep recurrence calculation pattern, i.e. y.= W., or y.= Ro.
A kind of embodiment, can be by the size of message interval time probability distribution W of group initial time t=0 in the observation time.R is distributed with size of message Length Pr.Sum is inputted in the multistep recurrence calculation pattern as initial value, i.e. y.= W.+ R0。
Another embodiment, can be by the size of message interval time probability distribution W of group initial time t=0 in the observation time.R is distributed with size of message Length Pr.Product inputted as initial value in the multistep recurrence calculation pattern, i.e. y.=W。*R。。
After above-mentioned initial value is input in the multistep recurrence calculation pattern, group's response of each period can be obtainedΛ , ΛIt is right for discrete point；Carry out curve fitting, individual consumer can be obtained active to group 4 especially dynamic response curve.
In view of the active diversity and complexity for group's change in time and space of individual consumer, individual consumer is active, and the response curve disturbed to group will have a series of different shapes, and corresponding reaction time and adjustment time are also different；Reaction time and adjustment time are referred to as the response time by we, and the active response process to group's disturbance of individual consumer is called the active response phenomenon that individual consumer develops to group.
For this characteristic according to response curve, reflection stage, adjusting stage and equilibrium stage that group responds J are obtained；Wherein：The reflection stage, which is that individual consumer is active, disturbs required reaction time section for group；Adjusting stage is that individual consumer is active to be adjusted to the period of poised state for group；Equilibrium stage is individual consumer's active period for group's maintenance poised state.
Referring to Fig. 6, Fig. 6 is the schematic diagram of group's response curve four-stage.Group is in observation time T Response curve can be divided into four parts：1. before disturbing, stage, 3. adjusting stage and 4. equilibrium stage are 2. reflected.
' S52, just it is blunt according to the group described in ^ observe in size of message interval time probability distribution or/and message for example, in group message amount interval time probability distribution graph, it can determine in any one period At, the size of message interval time probability sent every time is how many, the size of message quantity sent altogether is how many, is which individual consumer have sent size of message respectively（It is active）, active individual consumer have sent size of message several times respectively（Frequency）, obtain individual consumer active in each period At it is possible thereby to count and its enliven frequency.
Equally, on group message amount Length Pr distribution map, it may be determined that go out in any one period At, the size of message Length Pr sent every time is how many, and the size of message quantity sent altogether is how many, is which individual consumer have sent size of message respectively（It is active）, active individual consumer have sent size of message several times respectively（Frequency）, thus can also count and obtain individual consumer active in each period At and its enliven frequency.
Group message amount interval time probability distribution and the distribution of group message amount Length Pr can certainly be combined, common statistical obtains individual consumer active in each period At and its enlivens frequency.
, S53, basis, group's response of each period, the individual consumer of day jump living and its enliven the frequency, the step includes two sub-steps：
The first step, according to group's response of each period, active individual consumer and its enlivens the frequency, obtains alive data of each individual consumer in each period, the alive data includes：The frequency is enlivened at the individual ^ families of each period and corresponding group rings. 、 、 、
Meter mode shows row：Each period, by the group's response for enlivening the frequency * correspondence periods of individual consumer, obtain individual consumer's alive data of each period；Then individual consumer's alive data of each period in observation time is carried out cumulative or accumulates and obtain the liveness of each individual consumer.In another embodiment of the invention, the method for the embodiment of the present invention also includes：
The group data in another observation time is obtained, obtaining size of message interval time probability distribution or/and size of message Length Pr of the group in another observation time according to the group data is distributed；
Size of message interval time probability distribution or/and size of message Length Pr distribution of the multistep recurrence calculation pattern and the group responded according to the group in another observation time, statistics obtain liveness of each individual consumer in another observation time in the group.
The process for obtaining each liveness of the individual consumer in another observation time in group is same as above-mentioned, is not repeating herein.The embodiment can obtain the liveness of each individual consumer in the group in different observation times, can analyze accordingly each individual consumer's liveness it is high when period for concentrating.
The group data that can be chosen in one or more another observation times is analyzed.
In the present invention, another embodiment in, present invention method also includes：Liveness in ,-another observation time is integrated, and obtains the synthesis liveness of each individual consumer in the group. Degrees of data row it is folded or average or other can mode integrated,5The present invention is not limited.Liveness in observation time obtains the synthesis liveness of each individual consumer in group after being integrated, the data more have universal significance compared with the degrees of data of enlivening in an observation time.
Scheme statistics according to embodiments of the present invention is obtained in group after the liveness of each individual consumer, user can just be distinguished according to user activity, and and then the group information responded with the member's transmission disturbance group for transferring group's liveness can be utilized to encourage group to enliven according to business demand.
User in group can be for example divided into leader's type, concord person's type, inquirer's type, participator's type；Wherein, leader's type is contribution frequency and the high people of contribution quantity, creates a large amount of contents, attracts numerous member's sympathetic responses, and the formation and development for group plays very important effect；Concord person's type for often other members are contributed the approval of topic content representation, follow or oppose, do not provide depth content, also not actively contributing content；Inquirer's type is, as the place of study and consulting, often to herein propose group problem, seek answer；Participator's type plays an active part in the building-up work of group not to create topic content actively, including it is earnest answer Information Inquirer and the problem of learner proposes, and exchange related experience and gains in depth of comprehension.Based on above method embodiment, present invention also offers a kind of device for obtaining individual consumer's liveness in group, as shown in Figure 70 0, the device 700 of individual consumer's liveness includes in the acquisition group of the embodiment of the present invention：
Probability distribution acquiring unit 710, for obtaining the group data in observation time, obtains size of message interval time probability distribution or/and size of message Length Pr of the group in the observation time according to the group data and is distributed；Wherein, size of message interval time is the interval time of continuous two size of messages record in group, and size of message length is to send the character length that size of message is recorded each time in group；
Model sets up unit 720, for responding the rule being directly proportional in the adjustment rate of change after being disturbed by individual consumer to the difference between current state and poised state based on group, sets up group's response model；
Recurrence calculation unit 730, the multistep recurrence calculation pattern for obtaining group's response according to group's response model；Wherein described multistep recurrence calculation pattern refers to that the result of calculation of a period obtains group's response after multiple periods as the primary condition of subsequent time period and by period recursion；
Individual liveness statistic unit 740, for the multistep recurrence calculation pattern responded according to the group and the group, stating the size of message interval time probability distribution in observation time/and size of message Length Pr point wherein, the main part that the probability distribution and group message amount Length Pr of the group message amount interval time that the probability distribution acquiring unit 710 is obtained are distributed in the distribution of log-log coordinate lower probability all obeys power-law distribution.
In an embodiment of the invention, device 700 also includes：Group's liveness determining unit, suitable for the power exponent size being distributed according to size of message interval time probability distribution of the group in the observation time or/and size of message Length Pr, determines the liveness of group.
Wherein, the model set up unit 720 foundation group's response model be：
-=β ^ ~] ^ wherein, be group response current state, be group response poised state, dt It is the adjustment rate of change of group's response, t is the time for coefficient.
Wherein, the multistep recurrence calculation pattern of the recurrence calculation unit 730 is obtained group response is： yn= d - e-^ ^y^ i-^ + y^'；Wherein,ΔRepresent observation time Τ being divided into that " interval time lengths of individual equal time section, J ^ are that group responds equilibrium valve in the period, and the initial value of Λ groups response is coefficient.
In an embodiment of the invention, the individual liveness statistic unit 740 includes：
Group's response computing module, for the size of message interval time probability distribution of initial time and/or the distribution of size of message Length Pr to be inputted in the multistep recurrence calculation pattern as initial value in the observation time using the group, group's response of each period is obtained；
Individual consumer's statistical module, for being distributed according to size of message interval time probability distribution of the group in the observation time or/and size of message Length Pr, statistics obtains the active individual consumer of each period and its enlivens the frequency；
Individual liveness counts mould,, the work of ^ values rings in the group that ^ roots ^ states each period1^ individual liveness.
One embodiment, individual consumer's statistical module includes following appoint --- individual submodule：
First statistic submodule, for any one of the group in the observation time in the size of message interval time probability distribution of initial time, the distribution of size of message Length Pr to be inputted in the multistep recurrence calculation pattern as initial value；
Second statistic submodule, for the size of message interval time probability distribution of initial time to be inputted in the multistep recurrence calculation pattern with size of message Length Pr distribution sum as initial value in the observation time using the group；
3rd statistic submodule, for the size of message interval time probability distribution of initial time to be inputted in the multistep recurrence calculation pattern with the product that size of message Length Pr is distributed as initial value in the observation time using the group.
Another embodiment, the individual liveness statistical module can include：
Alive data acquisition submodule, for group's response according to each period, active individual consumer and its enlivens the frequency, obtains alive data of each individual consumer in each period, the alive data includes：The individual consumer of each period enlivens the frequency and corresponding group's response；
Individual liveness statistic submodule, statistical analysis is done for the alive data according to certain mode respectively to each individual consumer in group, obtains each liveness of the individual consumer in the observation time in the group.
In another embodiment, device 700 also includes：
Second probability distribution acquiring unit, for obtaining the group data in another observation time, obtains size of message interval time probability distribution or/and size of message Length Pr of the group in another observation time according to the group data and is distributed；
Second individual liveness statistic unit, it is general for the size of message interval time probability distribution or/and size of message length of the multistep recurrence calculation pattern and the group that are responded according to the group in another observation time Rate is distributed, and statistics obtains liveness of each individual consumer in another observation time in the group.In this sends out another embodiment, device 700 also includes：, day, the liveness at least one described another observation time is integrated, and obtains the synthesis liveness of each individual consumer in the group.
The specific works mode of each module in said apparatus embodiment of the present invention, may refer to the related content in the invention described above embodiment of the method, will not be repeated here.In summary, the method and apparatus provided in an embodiment of the present invention for obtaining individual consumer's liveness in group, by obtaining the group data in observation time, size of message interval time probability distribution or/and size of message Length Pr of the group in the observation time are obtained according to the group data and are distributed；It is then based on the rule that group's response is directly proportional in the adjustment rate of change after being disturbed by individual consumer to the difference between current state and poised state, set up group's response model, then the multistep recurrence calculation pattern of group's response is obtained further according to group's response model, the size of message interval time probability distribution in the observation time of multistep recurrence calculation pattern and the group responded finally according to the group or/and the distribution of size of message Length Pr, statistics obtains the relation between excavation individual behavior and group behavior, solution is provided for research individual consumer's active changing process, easily can be got from group can transfer the active member of group, so as to which the group information responded with the member's transmission disturbance group for transferring group's liveness can be utilized to encourage group to enliven according to business demand.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the scope of the present invention.Any modification, equivalent substitution and improvements made within the spirit and principles of the invention etc., are all contained in protection scope of the present invention.

Claims (1)

  1. Claims
    1st, a kind of method for obtaining individual consumer's liveness in group, it is characterised in that methods described includes：The group data in observation time is obtained, obtaining size of message interval time probability distribution or/and size of message Length Pr of the group in the observation time according to the group data is distributed；Wherein, size of message interval time is the interval time of continuous two size of messages record in group, and size of message length is to send the character length that size of message is recorded each time in group；
    The rule being directly proportional in the adjustment rate of change after being disturbed by individual consumer to the difference between current state and poised state is responded based on group, group's response model is set up；
    The multistep recurrence calculation pattern of group's response is obtained according to group's response model；Wherein described multistep recurrence calculation pattern refers to that the result of calculation of a period obtains group's response after multiple periods as the primary condition of subsequent time period and by period recursion；
    Size of message interval time probability distribution or/and size of message Length Pr distribution of the multistep recurrence calculation pattern and the group responded according to the group in the observation time, statistics obtain each liveness of the individual consumer in the observation time in the group.
    2nd, according to the method described in claim 1, it is characterised in that
    The main part that the probability distribution and group message amount Length Pr of the group message amount interval time of acquisition is distributed in the distribution of log-log coordinate lower probability all obeys power-law distribution.
    3rd, method according to claim 2, it is characterised in that after size of message interval time probability distribution or/and size of message Length Pr distribution of the group in the observation time are obtained according to the group data, methods described also includes：
    The power exponent size being distributed according to size of message interval time probability distribution of the group in the observation time or/and size of message Length Pr, determines the liveness of group.
    4th, according to the method described in claim 1, it is characterised in that
    Group's response model is：
    — = [ye- y] ^ wherein,>The current state responded for group, is the poised state of group's response, dt
    /It is the adjustment rate of change of group's response, t is the time for coefficient.
    5th, method according to claim 4, it is characterised in that
    The multistep recurrence calculation pattern of group response is：
    yn= d - e-^ ^y^ i-^ + y^'；Wherein,ΔRepresent observation time Τ being divided into that " interval time lengths of individual equal time section, J ^ are that group responds equilibrium valve in the period, and the initial value of Λ groups response is coefficient.
    6th, according to the method described in claim 1, it is characterized in that, size of message interval time probability distribution or/and size of message Length Pr of the multistep recurrence calculation pattern and the group responded according to the group in the observation time are distributed, and statistics, which obtains each liveness of the individual consumer in the observation time in the group, to be included：
    The group size of message interval time probability distribution of initial time and/or is disappeared in the observation time The distribution of breath amount Length Pr is inputted in the multistep recurrence calculation pattern as initial value, obtains group's response of each period；
    , just blunt according to the group, stating the size of message interval time ^ in observation, both rate distribution or/and size of message length were according to having stated group's response of each period, the individual consumer died young living and its enlivened the frequency, statistics
    7th, method according to claim 6, characterized in that, described, using the group, the size of message interval time probability distribution of initial time and/or the distribution of size of message Length Pr include following any situation as the initial value input multistep recurrence calculation pattern in the observation time：
    Any one of the group in the observation time in the size of message interval time probability distribution of initial time, the distribution of size of message Length Pr is inputted in the multistep recurrence calculation pattern as initial value；Or, using the group, the size of message interval time probability distribution of initial time is inputted in the multistep recurrence calculation pattern with size of message Length Pr distribution sum as initial value in the observation time；Or,
    Using the group, the size of message interval time probability distribution of initial time is inputted in the multistep recurrence calculation pattern with the product that size of message Length Pr is distributed as initial value in the observation time.
    8th, method according to claim 6, it is characterized in that, described according to group's response of each period, active individual consumer and its to enliven the frequency, statistics, which obtains each liveness of the individual consumer in the observation time in the group, to be included：
    According to group's response of each period, active individual consumer and its frequency is enlivened, obtain alive data of each individual consumer in each period, the alive data includes：Each period
    9th, the method according to claim any one of 1-8, it is characterised in that methods described also includes：The group data in another observation time is obtained, obtaining size of message interval time probability distribution or/and size of message Length Pr of the group in another observation time according to the group data is distributed；
    Size of message interval time probability distribution or/and size of message Length Pr distribution of the multistep recurrence calculation pattern and the group responded according to the group in another observation time, statistics obtain liveness of each individual consumer in another observation time in the group.
    10th, the method described in root ^ claims 9, its feature is in methods described also includes：Liveness in another observation time is integrated, and obtains the synthesis liveness of each individual consumer in the group.
    11st, a kind of device for obtaining individual consumer's liveness in group, it is characterised in that described device includes：Probability distribution acquiring unit, for obtaining the group data in observation time, obtains size of message interval time probability distribution or/and size of message Length Pr of the group in the observation time according to the group data and is distributed；Wherein, size of message interval time is the interval time of continuous two size of messages record in group, and size of message length is to send the character length that size of message is recorded each time in group；
    Model sets up unit, for based on group respond adjustment rate of change after being disturbed by individual consumer with The rule that difference between current state and poised state is directly proportional, sets up group's response model；Recurrence calculation unit, the multistep recurrence calculation pattern for obtaining group's response according to group's response model；Wherein described multistep recurrence calculation pattern refers to that the result of calculation of a period obtains group's response after multiple periods as the primary condition of subsequent time period and by period recursion；
    Individual liveness statistic unit, for the size of message interval time probability point in the sight time described in the multistep recurrence calculation pattern responded according to the group and the group or/and the distribution of size of message Length Pr,
    12nd, device according to claim 11, characterized in that, the main part that the probability distribution and group message amount Length Pr of the group message amount interval time that the probability distribution acquiring unit is obtained are distributed in the distribution of log-log coordinate lower probability all obeys power-law distribution.
    13rd, device according to claim 12, it is characterised in that described device also includes：Group's liveness determining unit, suitable for the power exponent size being distributed according to size of message interval time probability distribution of the group in the observation time or/and size of message Length Pr, determines the liveness of group.
    14th, device according to claim 11, it is characterised in that
    The model set up unit foundation group's response model be：
    -=β ^ ~] ^ wherein, be group response current state, be group response poised state, dt
    It is the adjustment rate of change of group's response, t is the time for coefficient.
    15th, device according to claim 14, it is characterised in that
    The multistep recurrence calculation pattern of group that the recurrence calculation unit is obtained response is：
    yn= (l - e-^' ^y^ i-^-^ + y^；Wherein, ^ represents observation time T being divided into, and " interval time lengths of individual equal time section, j ^ are that group responds equilibrium valve in the period, y.The initial value of group's response ,/For coefficient.
    16th, device according to claim 11, it is characterised in that the individual liveness statistic unit includes：
    Group's response computing module, for the size of message interval time probability distribution of initial time and/or the distribution of size of message Length Pr to be inputted in the multistep recurrence calculation pattern as initial value in the observation time using the group, group's response of each period is obtained；
    Individual consumer's statistical module, for being distributed according to size of message interval time probability distribution of the group in the observation time or/and size of message Length Pr, statistics obtains the active individual consumer of each period and its enlivens the frequency；
    Individual liveness counts mould,, the work of ^ values rings in the group that ^ roots ^ states each period1^ individual liveness.
    17th, device according to claim 16, it is characterised in that individual consumer's statistical module includes following appoint --- individual submodule：
    First statistic submodule, for any one of the group in the observation time in the size of message interval time probability distribution of initial time, the distribution of size of message Length Pr to be inputted in the multistep recurrence calculation pattern as initial value； Second statistic submodule, for the size of message interval time probability distribution of initial time to be inputted in the multistep recurrence calculation pattern with size of message Length Pr distribution sum as initial value in the observation time using the group；
    3rd statistic submodule, for the size of message interval time probability distribution of initial time to be inputted in the multistep recurrence calculation pattern with the product that size of message Length Pr is distributed as initial value in the observation time using the group.
    18th, device according to claim 16, it is characterised in that the individual liveness statistical module includes：
    Alive data acquisition submodule, for group's response according to each period, active individual consumer and its enlivens the frequency, obtains alive data of each individual consumer in each period, the alive data includes：The individual consumer of each period enlivens the frequency and corresponding group's response；
    Individual liveness statistic submodule, statistical analysis is done for the alive data according to certain mode respectively to each individual consumer in group, obtains each liveness of the individual consumer in the observation time in the group.
    19th, the device according to claim any one of 11-18, it is characterised in that described device also includes：Second probability distribution acquiring unit, for obtaining the group data in another observation time, obtains size of message interval time probability distribution or/and size of message Length Pr of the group in another observation time according to the group data and is distributed；
    Second individual liveness statistic unit, for size of message interval time probability distribution in another observation time of the multistep recurrence calculation pattern and the group that are responded according to the group or/and the distribution of size of message Length Pr, statistics obtains liveness of each individual consumer in another observation time in the group.
    20th, the device described in a claim 19, it is characterised in that described device, including：' the liveness at least one described another observation time integrated, and obtains the synthesis liveness of each individual consumer in the group.
CN201480012240.1A 2014-06-30 2014-06-30 A kind of method and apparatus obtaining individual consumer's liveness in group Active CN105409162B (en)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2014/081248 WO2016000164A1 (en) 2014-06-30 2014-06-30 Method and device for acquiring activity level of individual users in group

Publications (2)

Publication Number Publication Date
CN105409162A true CN105409162A (en) 2016-03-16
CN105409162B CN105409162B (en) 2018-07-27

Family

ID=55018250

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201480012240.1A Active CN105409162B (en) 2014-06-30 2014-06-30 A kind of method and apparatus obtaining individual consumer's liveness in group

Country Status (2)

Country Link
CN (1) CN105409162B (en)
WO (1) WO2016000164A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110971673A (en) * 2019-11-15 2020-04-07 广东浪潮大数据研究有限公司 Computer device and method for acquiring user activity of deep learning platform
CN112818244A (en) * 2021-02-24 2021-05-18 深圳市网联安瑞网络科技有限公司 Method, system and terminal for judging activity of channel, group and group user

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7940702B1 (en) * 2005-09-23 2011-05-10 Avaya Inc. Method and apparatus for allowing communication within a group
CN1980219A (en) * 2005-12-09 2007-06-13 腾讯科技（深圳）有限公司 Method and system for realizing grouped member cleaning
CN101478450B (en) * 2009-02-12 2011-09-28 阿里巴巴集团控股有限公司 Monitoring method and server for network user active degree
CN103684807A (en) * 2013-12-10 2014-03-26 小米科技有限责任公司 Method, device, terminal device and server for identifying group member information

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110971673A (en) * 2019-11-15 2020-04-07 广东浪潮大数据研究有限公司 Computer device and method for acquiring user activity of deep learning platform
CN110971673B (en) * 2019-11-15 2022-08-12 广东浪潮大数据研究有限公司 Computer device and method for acquiring user activity of deep learning platform
CN112818244A (en) * 2021-02-24 2021-05-18 深圳市网联安瑞网络科技有限公司 Method, system and terminal for judging activity of channel, group and group user

Also Published As

Publication number Publication date
WO2016000164A1 (en) 2016-01-07
CN105409162B (en) 2018-07-27

Similar Documents

Publication Publication Date Title
Ferschke et al. Positive impact of collaborative chat participation in an edX MOOC
Bell Toward a Definition of
Ergün et al. An analysis of density and degree-centrality according to the social networking structure formed in an online learning environment
Ebner et al. Getting granular on twitter: tweets from a conference and their limited usefulness for non-participants
Gautier et al. Social networks and acquisition of sociolinguistic variation in a study abroad context: A preliminary study
Corti et al. The researcher as experimental subject: Using self-experimentation to access experiences, understand social phenomena, and stimulate reflexivity
König Narratives 2.0: A multi-dimensional approach to semi-public storytelling in WhatsApp voice messages
CN105409162A (en) Method and device for acquiring activity level of individual users in group
Tychsen et al. Player-Character Dynamics in Multi-Player Role Playing Games.
Yalcinalp et al. Higher education Student's behaviors as avatars in a web based course in second life
Selmbacherova et al. The impact of visual realism on the authenticity of educational simulation: a comparative study
Aleffi et al. Research, education and co-creation: the university in place/Ricerca, formazione e co-creazione: l'università sul campo
Pürcher et al. Individual versus collaborative learning in a virtual world
Price Consumer culture theory: A front-row seat at the sidelines
Marley The role of online communication in raising awareness of bilingual identity
Abend et al. Introduction: The boundaries of play
Ennis et al. Youth arts and mental health: exploring connections in the top end
du Plessis Using Q methodology to test Perspectives and Attitudes: Experiences from a Study about Content Marketing
Brozen The value of technological change
Watson et al. Licorice production and manufacturing: All-sorts of practical applications for statistics
Sharma A Study of Entrepreneurs Background and their Perception of Tourism in Goa
Kristiansen et al. Is gaming always fun? Social conflicts in online consumption communities
Fitriyanti et al. Empowerment of Emporing Family Welfare Members of Sumber Kidul Village through Household Waste-Based Craft during COVID-19
Manasia The impact of social software on developing communities of practice to enhance adult learning
Pithouse-Morgan et al. A transcontinental tapestry: Co-creativity in polyvocal self-study research

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CP02 Change in the address of a patent holder
CP02 Change in the address of a patent holder

Address after: Room 810, 8 / F, 34 Haidian Street, Haidian District, Beijing 100080

Patentee after: BEIJING D-MEDIA COMMUNICATION TECHNOLOGY Co.,Ltd.

Address before: 100089 Beijing city Haidian District wanquanzhuang Road No. 28 Wanliu new building 6 storey block A room 602

Patentee before: BEIJING D-MEDIA COMMUNICATION TECHNOLOGY Co.,Ltd.