CN105409162B - A kind of method and apparatus obtaining individual consumer's liveness in group - Google Patents
A kind of method and apparatus obtaining individual consumer's liveness in group Download PDFInfo
- Publication number
- CN105409162B CN105409162B CN201480012240.1A CN201480012240A CN105409162B CN 105409162 B CN105409162 B CN 105409162B CN 201480012240 A CN201480012240 A CN 201480012240A CN 105409162 B CN105409162 B CN 105409162B
- Authority
- CN
- China
- Prior art keywords
- group
- probability distribution
- time
- message
- activity
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000004044 response Effects 0.000 claims abstract description 154
- 238000004364 calculation method Methods 0.000 claims abstract description 65
- 230000008859 change Effects 0.000 claims abstract description 18
- 230000000694 effects Effects 0.000 claims description 117
- 238000007619 statistical method Methods 0.000 claims description 6
- 230000002194 synthesizing effect Effects 0.000 claims description 4
- 238000011160 research Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 230000006399 behavior Effects 0.000 description 7
- 230000008569 process Effects 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 230000000977 initiatory effect Effects 0.000 description 4
- 230000001483 mobilizing effect Effects 0.000 description 3
- 230000035484 reaction time Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000001186 cumulative effect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 230000003213 activating effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000003999 initiator Substances 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
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)
- Information Transfer Between Computers (AREA)
- Cosmetics (AREA)
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
Abstract
The invention discloses a kind of method and apparatus of individual consumer's liveness in acquisition group.The method 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;Based on group's response rule directly proportional to the difference between current state and equilibrium state in the adjustment change rate after being disturbed by individual consumer, group's response model is established;According to the group, response model obtains the multistep recurrence calculation pattern of group's response;It is distributed according to the size of message interval time probability distribution or/and size of message Length Pr of the multistep recurrence calculation pattern and the group of group response in the observation time, statistics obtains each liveness of the individual consumer in the observation time in the group.Technical scheme of the present invention, which can easily be got from group, can transfer the active member of group.
Description
Technical Field
The present invention relates to the field of instant messaging technologies, and in particular, to a method and an apparatus for obtaining activity of individual users in a group.
Background
With the rapid development of related research and application related to web2.0, web2.0 focuses more on the interactivity of users, i.e., a group environment participating in expression, creation, communication and sharing is established for users, and the users surf the internet to the user web, so that the users are not only the attendees and browsers initiating group content, but also the producers and propagators initiating group content.
The research on the groups mainly focuses on two aspects, on one hand, the research is a structural research which comprises a social network, an information exchange mode and the like; on the other hand, the active content research is embodied in the reason that the individual driving group actively generates the content, and the part directly influences the attitude, the behavior intention and the actual behavior of the user to the group. Currently, for the research of instant messaging groups, the industry has made many researches from the individual perspective, but the research on group is relatively few, and members capable of activating the group cannot be obtained from the group.
Disclosure of Invention
The embodiment of the invention provides a method and a device for acquiring the activity of an individual user in a group, so that members capable of mobilizing the activity of the group can be conveniently acquired from the group.
In order to achieve the above purpose, the embodiment of the present invention adopts the following technical solutions:
in one aspect, an embodiment of the present invention provides a method for acquiring activity of an individual user in a group, including:
acquiring group data in observation time, and acquiring message quantity interval time probability distribution or/and message quantity length probability distribution of the group in the observation time according to the group data; the message volume interval time is the interval time of two continuous message volume records in the group, and the message volume length is the character length of each record of sending the message volume in the group;
establishing a group response model based on the rule that the adjustment change rate of the group response after being disturbed by the individual user is in direct proportion to the difference between the current state and the balance state;
obtaining a multi-step recursion calculation mode of group response according to the group response model; the multi-step recursion calculation mode is that the calculation result of the previous time period is used as the initial condition of the next time period, and the group response values after a plurality of time periods are obtained by recursion time periods;
and counting to obtain the activity of each individual user in the group in the observation time according to the multi-step recursion calculation mode of the group response and the message quantity interval time probability distribution or/and the message quantity length probability distribution of the group in the observation time.
Optionally, the obtained probability distribution of the group message volume interval time and the obtained probability distribution of the group message volume length are subjected to power law distribution in the main part of the probability distribution under the log-log coordinate.
Optionally, after obtaining the message volume interval time probability distribution or/and the message volume length probability distribution of the group in the observation time according to the group data, the method further includes: and determining the liveness of the group according to the message quantity interval time probability distribution or/and the power exponent of the message quantity length probability distribution of the group in the observation time.
Optionally, the group response model is:where y is the current state of the group response, yeFor the equilibrium state of the group response, β is the coefficient, dy/dt is the rate of change of the adjustment of the group response, and t is time.
Optionally, the multi-step recursive computation pattern of the group response is:
where Vt represents an interval time length obtained by dividing the observation time T into n equal time periods, yeiBalance value, y, for group response at i time period0The initial value of the group response, β, is a coefficient.
Optionally, the obtaining, according to the multi-step recursion calculation mode of the group response and the message volume interval time probability distribution or/and the message volume length probability distribution of the group within the observation time, the activity of each individual user in the group within the observation time through statistics includes:
inputting the message quantity interval time probability distribution and/or the message quantity length probability distribution of the group at the initial moment in the observation time into the multi-step recursion calculation mode as initial values to obtain a group response value of each time period;
according to the message volume interval time probability distribution or/and the message volume length probability distribution of the group in the observation time, counting to obtain active individual users and the active frequency of the active individual users in each time period;
and counting the activity of each individual user in the group within the observation time according to the group response value of each time period, the active individual users and the active frequency thereof.
Optionally, the inputting the message quantity interval time probability distribution and/or the message quantity length probability distribution of the group at the initial time within the observation time as the initial value into the multi-step recursion calculation mode includes any one of the following situations:
inputting any one of message quantity interval time probability distribution and message quantity length probability distribution of the group at the initial time in the observation time into the multi-step recursion calculation mode as an initial value; or,
inputting the sum of the message quantity interval time probability distribution and the message quantity length probability distribution of the group at the initial moment in the observation time as an initial value into the multi-step recursion calculation mode; or,
and inputting the product of the message volume interval time probability distribution and the message volume length probability distribution of the group at the initial moment in the observation time as an initial value into the multi-step recursion calculation mode.
Optionally, the counting, according to the group response value of each time period, the active individual users and their active frequencies, the activity of each individual user in the group in the observation time includes:
obtaining the active data of each individual user in each time period according to the group response value of each time period, the active individual users and the active frequency thereof, wherein the active data comprises: the individual user active frequency of each time period and the corresponding group response value;
and respectively carrying out statistical analysis on the activity data of each individual user in the group according to a set mode to obtain the activity of each individual user in the group in the observation time.
Optionally, the method further comprises: acquiring group data in another observation time, and acquiring message quantity interval time probability distribution or/and message quantity length probability distribution of the group in the another observation time according to the group data; and counting the activity of each individual user in the group in the other observation time according to the multi-step recursion calculation mode of the group response and the message quantity interval time probability distribution or/and the message quantity length probability distribution of the group in the other observation time.
Optionally, the method further comprises: and integrating the activity of each individual user in the group in the observation time with the activity of each individual user in at least one other observation time to obtain the integrated activity of each individual user in the group.
In another aspect, an embodiment of the present invention provides an apparatus for acquiring activity of individual users in a group, including:
a probability distribution obtaining unit, configured to obtain group data within observation time, and obtain message quantity interval time probability distribution or/and message quantity length probability distribution of the group within the observation time according to the group data; the message volume interval time is the interval time of two continuous message volume records in the group, and the message volume length is the character length of each record of sending the message volume in the group;
the model establishing unit is used for establishing a group response model based on the rule that the adjustment change rate of the group response after being disturbed by the individual user is in direct proportion to the difference between the current state and the balance state;
the recursive calculation unit is used for obtaining a multi-step recursive calculation mode of the group response according to the group response model; the multi-step recursion calculation mode is that the calculation result of the previous time period is used as the initial condition of the next time period, and the group response values after a plurality of time periods are obtained by recursion time periods;
and the individual activity counting unit is used for counting to obtain the activity of each individual user in the group in the observation time according to the multi-step recursion calculation mode of the group response and the message quantity interval time probability distribution or/and the message quantity length probability distribution of the group in the observation time.
Optionally, the probability distribution of the interval time of the group message volume and the probability distribution of the length of the group message volume obtained by the probability distribution obtaining unit are subjected to power law distribution in the main part of the probability distribution under the log-log coordinate.
Optionally, the apparatus further comprises:
and the group activity determining unit is suitable for determining the activity of the group according to the power exponent of the message quantity interval time probability distribution or/and the message quantity length probability distribution of the group in the observation time.
Optionally, the group response model established by the model establishing unit is:
where y is the current state of the group response, yeFor the equilibrium state of the group response, β is the coefficient, dy/dt is the rate of change of the adjustment of the group response, and t is time.
Optionally, the multi-step recursive computation mode of the group response obtained by the recursive computation unit is as follows:
where Vt represents an interval time length obtained by dividing the observation time T into n equal time periods, yeiBalance value, y, for group response at i time period0The initial value of the group response, β, is a coefficient.
Optionally, the individual liveness statistics unit comprises:
the group response value calculation module is used for inputting the message quantity interval time probability distribution and/or the message quantity length probability distribution of the group at the initial time in the observation time into the multi-step recursion calculation mode as initial values to obtain a group response value of each time period;
the individual user counting module is used for counting and obtaining active individual users and active frequency thereof in each time period according to message quantity interval time probability distribution or/and message quantity length probability distribution of the group in the observation time;
and the individual activity counting module is used for counting the activity of each individual user in the group within the observation time according to the group response value of each time period, the active individual users and the active frequency thereof.
Optionally, the individual user statistics module includes any one of the following sub-modules:
a first statistical submodule configured to input any one of a message size interval time probability distribution and a message size length probability distribution of the group at an initial time within the observation time as an initial value into the multi-step recursive computation mode;
the second statistic submodule is used for inputting the sum of the message quantity interval time probability distribution and the message quantity length probability distribution of the group at the initial time in the observation time into the multi-step recursion calculation mode as an initial value;
and the third statistical submodule is used for inputting the product of the message quantity interval time probability distribution and the message quantity length probability distribution of the group at the initial moment in the observation time into the multi-step recursion calculation mode as an initial value.
Optionally, the individual liveness statistics module includes:
the active data obtaining sub-module is configured to obtain active data of each individual user in each time period according to the group response value of each time period, the active individual users and the active frequency thereof, where the active data includes: the individual user active frequency of each time period and the corresponding group response value;
and the individual activity degree counting submodule is used for respectively carrying out statistical analysis on the activity data of each individual user in the group according to a set mode to obtain the activity degree of each individual user in the group in the observation time.
Optionally, the apparatus further comprises:
a second probability distribution obtaining unit, configured to obtain group data in another observation time, and obtain, according to the group data, message quantity interval time probability distribution or/and message quantity length probability distribution of the group in the another observation time;
and the second individual activity counting unit is used for counting the activity of each individual user in the group in the other observation time according to the multi-step recursion calculation mode of the group response and the message volume interval time probability distribution or/and the message volume length probability distribution of the group in the other observation time.
Optionally, the apparatus further comprises:
and the comprehensive unit is used for synthesizing the activity of each individual user in the group in the observation time and the activity of each individual user in at least one other observation time to obtain the comprehensive activity of each individual user in the group.
The embodiment of the invention has the advantages that by acquiring the group data in the observation time, the message quantity interval time probability distribution or/and the message quantity length probability distribution of the group in the observation time are/is acquired according to the group data; then, based on the rule that the adjustment change rate of the group response after being disturbed by the individual users is in direct proportion to the difference between the current state and the balance state, a group response model is established, then, a multi-step recursion calculation mode of the group response is obtained according to the group response model, and finally, the activity of each individual user in the group in the observation time is obtained through statistics according to the multi-step recursion calculation mode of the group response and the message quantity interval time probability distribution or/and the message quantity length probability distribution of the group in the observation time. The technical scheme of the invention is beneficial to mining the relation between the individual behaviors and the group behaviors, provides a solution for researching the active change process of individual users, and can conveniently acquire the members capable of mobilizing the group activity from the group, so that the members with the mobilized group activity can be used for sending the group information disturbing the group response to stimulate the group activity according to the service requirements.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
FIG. 1 is a flowchart illustrating a method for obtaining activity of individual users in a group according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating obtaining group information for a group over a period of time according to an embodiment of the invention;
FIG. 3 shows a schematic diagram of a group message quantity interval probability distribution of a group in a log-log coordinate according to an embodiment of the invention;
FIG. 4 illustrates an example of a group message size probability distribution of a group in log-log coordinates according to an embodiment of the invention;
FIG. 5 is a schematic diagram illustrating a method for counting the activity of each individual user in the group during the observation time according to an embodiment of the present invention;
FIG. 6 is a schematic diagram illustrating four phases of a group response curve according to an embodiment of the invention;
fig. 7 is a schematic structural diagram of an apparatus for acquiring activity of individual users in a group according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
Referring to fig. 1, an embodiment of the present invention provides a method for obtaining activity of individual users in a group, including the following steps:
s11, acquiring group information in observation time, and acquiring message quantity interval time probability distribution or/and message quantity length probability distribution of the group in the observation time according to the group information.
The message volume interval time is the interval time of two continuous message volume records in the group, and the message volume length is the character length of each record of the message volume sent in the group.
Fig. 2 is a diagram illustrating group information obtained for a group over a period of time. The group information includes basic information and an observation table of the group. The group basic information comprises a group name, the number of people in the group, the purpose of establishing the group, an observation time interval, whether to voluntarily join, a group type, a group brief introduction, a group notice and the like, wherein the group type can comprise a primary type and a secondary type; the observation table includes observation date, conversation initiating time, conversation ending time, the number of people participating in the conversation, the initiator, the initiating reason, chat topics and the like.
Because the instant messaging group is a chat group, the heterogeneity of message sending quantity of individual users, such as typing speed, internet surfing frequency, message sending positivity and the like, can be greatly reduced, so that the more representative the obtained data is, the more the characteristics of the individual users in the group can be reflected. Therefore, according to the obtained group information, it is also necessary to obtain a group message amount interval time probability distribution or/and a group message amount length probability distribution within the observation time T.
In one embodiment of the invention, by researching the probability distribution of the message volume interval time or/and the message volume length probability distribution obtained by different groups, the probability distribution of the message volume interval time or/and the message volume length probability distribution in a log-log coordinate is consistent, and the main body part follows power law distribution.
Fig. 3 is an example of a group message volume interval time probability distribution of a group in a log-log coordinate. Although the records of each group are different, the composition personnel of the individual users of the group are different, the probability distribution under the log-log coordinate shows consistency, and the probability distribution shows obvious fat tail distribution, namely, most of the message sending quantity has short interval time, a small quantity of the message sending quantity has long interval time, the drooping head and the long tail are removed, the main body of the probability distribution follows power law distribution, and the power exponent is concentrated in the range of about 1.85-2.11. The distribution characteristics are in accordance with the potential rules of group interactive behaviors, group message propagation is often caused by a topic, and a person initiates the topic to cause the mutual participation of members in a group, so that when the topic occurs, the disturbance is much and the interval time is short, and after the topic is ended, the silence is very long.
Fig. 4 is an example of a group message size probability distribution of a group in a log-log coordinate. The group is used as an instant communication tool, the interaction has instantaneity, and the length of each message is not too long in general. The cumulative distribution of the character number has certain volatility, the main body part can be approximately fitted by a straight line, and therefore the cumulative distribution of the character number can be considered to be in a power law, and the power exponent is between 1.55 and 2.64. The span is large relative to the interval time, which may be due to the different composition topics for each group, and thus the content of the sent messages is differentiated. Colleagues such as group B, in which messages are often short, messages with few characters account for a greater proportion of the total number of messages, and therefore have a greater slope. Interest group a tends to share content for interest, with longer ones in the message being relatively more numerous, resulting in a difference in slope.
In one embodiment of the invention, the liveness of the group is determined according to the message volume interval time probability distribution or/and the power exponent size of the message volume length probability distribution of the group in the observation time.
The way of determining the activity of the group can be determined according to only one of the power exponent size of the message volume interval time probability distribution or the power exponent size of the message volume length probability distribution of the group in the observation time, or according to the sum or product of the power exponent sizes of the two.
Generally speaking, the larger the power exponent, the higher the liveness of a group, so the liveness of different groups can be sorted according to the magnitude of the power exponent of the message quantity interval time probability distribution or/and the message quantity length probability distribution of the group. And selecting a certain number of groups for research according to the activity ranking of different groups.
S12, establishing a group response model based on the rule that the adjustment change rate of the group response after being disturbed by the individual user is in direct proportion to the difference between the current state and the balance state.
The established group response model can be described by a first order ordinary differential equation:
where y is the current state of the group response, yeDy/dt is the equilibrium state of the group responseThe rate of change of the adjustment of the group response, T being time, 0 ≦ T ≦ observed time T, β being a coefficient, β may in principle be time-variable but is assumed here to be constant for solution convenience.
The above formula is a basic model of individual user disturbance group response, which can be used to describe the time course of group adjustment and group balance, and the group response model has general applicability.
And S13, obtaining a multi-step recursion calculation mode of the group response according to the group response model.
The multi-step recursion calculation mode is that the calculation result of the previous time period is used as the initial condition of the next time period, and the group response values after a plurality of time periods are obtained by recursion time periods.
The process of solving the first order ordinary differential equation (1) is as follows:
due to the fact thatIt is complicated to bring specific values, and equation (1) can be rewritten to the following general form for easy solution, and obtained:
it is apparent that equation (2) is a first-order non-homogeneous linear equation whose general solution is:
y=e-∫βdt[∫βyee-∫βdtdt+C1](3)
wherein, C1Where is the integration constant.
When t is equal to 0, y is equal to y0Substituting the formula (3) to obtain a special solution of the formula (2):
wherein y is the group response, yeA balance value for the group response; y is0Is an initial condition or value, t is time, β is a coefficient in principle β is time variable, assumed to be constant.
The integral term on the right side of the formula (4) is directly solved to obtain
y=ye+(y0-ye)e-βt(5)
Y is y when t is 00
Y is y when t is infinitye
Or
y=(1-e-βt)ye+y0e-βt(6)
Analytical formula:
because the result of the group adjustment stage by the activity of the individual user is used as the initial condition of the next time interval no matter whether the state is the balance state or not, the group response is influenced, and therefore the message volume length and the interval time condition in the previous period influence the group response in the later period, therefore, the calculation result in the previous time interval can be used as the initial condition of the next time interval, and the response state values after a plurality of time intervals can be obtained by recursion from time interval to time interval.
For this purpose, the observation time T is equally divided into n time segments, each time segment is denoted as Vt, and equation (6) can be denoted as Vt in the first time segment
y1=(1-e-βVt)ye1+y0e-βVt(7)
Wherein, ye1Is the equilibrium value of the group during the first time period.
Correspondingly, the 2 nd time period also has
y2=(1-e-βVt)ye2+y1e-βVt(8)
Combining (7) and (8) to obtain
y2=(1-e-βVt)[ye2+y1e-βVt]+y0e-2βVt(9)
Thereby stepping to the nth time period
Equation (10), a multi-step recursive computation scheme called group response, where yeiFor the equilibrium value of the group in the i-th time segment, Vt represents the division of the observation time T into n equal time segments, y0Initial value of group response.
S14, according to the multi-step recursion calculation mode of the group response and the message quantity interval time probability distribution or/and the message quantity length probability distribution of the group in the observation time, counting to obtain the activity of each individual user in the group in the observation time.
In an embodiment of the present invention, referring to fig. 5, step S14 specifically includes:
and S51, inputting the message quantity interval time probability distribution and/or the message quantity length probability distribution of the group at the initial time in the observation time into the multi-step recursion calculation mode as initial values to obtain the group response value of each time period.
In one embodiment, the message amount of the group at the initial time t equal to 0 in the observation time may be separated by the time probability distribution W0Or message size probability distribution R0As an initial value input into the multi-step recursive calculation mode, i.e. y0=W0Or y is0=R0。
An embodiment, groups may be formedMessage size interval time probability distribution W of group with initial time t being 0 in observation time0And message size probability distribution R0The sum is input as an initial value into the multi-step recursive computation mode, i.e. y0=W0+R0。
In still another embodiment, the group may be separated by a message amount interval probability distribution W with an initial time t equal to 0 within the observation time0And message size probability distribution R0The product of (a) and (b) is input as an initial value into the multi-step recursive computation model, i.e. y0=W0*R0。
After the initial value is input into the multi-step recursion calculation mode, the group response value y of each time period can be obtainedn,ynAs discrete points, for ynAnd performing curve fitting to obtain a response curve of the individual user activity to the group disturbance.
Considering the diversity and complexity of the individual user activity to the group space-time change, the response curve of the individual user activity to the group disturbance has a series of different shapes, and the corresponding reaction time and the adjustment time are different; the reaction time and the adjustment time are collectively called response time, and the response process of the individual user activity to the group disturbance is called an active response phenomenon of the individual user to the group evolution.
Therefore, according to the characteristics of the response curve, a reflection stage, an adjustment stage and a balance stage of the group response y are obtained; wherein: the reflecting stage is a reaction time period required by the activity of the individual user to the group disturbance; the adjusting stage is a time period when the individual user is active and the group is adjusted to a balance state; the balance phase is the period of time that an individual user is active to maintain a balanced state for the group.
Referring to FIG. 6, FIG. 6 is a diagram showing four phases of the group response curve, the response curve of the group in the observation time T can be divided into four parts, ① before disturbance, ② reflection phase, ③ adjustment phase and ④ balance phase.
S52, according to the message quantity interval time probability distribution or/and the message quantity length probability distribution of the group in the observation time, obtaining the active individual users and the active frequency of each time period in a statistical manner.
For example, on the group message amount interval probability distribution diagram, it can be determined what the message amount interval probability of each transmission is, what the total number of messages to be transmitted is, what individual user has transmitted the message amount (active), and what number of message amounts (frequencies) the active individual users have transmitted, respectively, in any time period Vt, so that the active individual users and their active frequencies in each time period Vt can be obtained statistically.
Similarly, on the group message volume length probability distribution map, it can be determined what message volume length probability is sent each time, what number of messages to be sent in total, which individual user sent the message volume (active) and which individual user sent the message volume (frequency) several times, in any time slot Vt, so that the active individual users and their active frequencies in each time slot Vt can be obtained statistically.
Of course, the active individual users and their active frequencies in each time period Vt may be obtained through statistics in combination with the group message volume interval probability distribution and the group message volume length probability distribution.
S53, according to the group response value of each time period, the active individual users and the active frequency thereof, counting the activity of each individual user in the group in the observation time.
This step comprises two substeps:
step one, obtaining active data of each individual user in each time period according to the group response value of each time period, the active individual users and the active frequency thereof, wherein the active data comprises: individual user active frequency per time period and corresponding group response value.
And secondly, respectively carrying out statistical analysis on the activity data of each individual user in the group according to a set mode to obtain the activity of each individual user in the group in the observation time.
An example of a statistical approach: in each time period, the active frequency of the individual user corresponds to the group response value of the time period, and the active data of the individual user in each time period is obtained; and then accumulating or accumulating the activity data of the individual users in each time period in the observation time to obtain the activity of each individual user.
In another embodiment of the present invention, the method of the embodiment of the present invention further comprises:
acquiring group data in another observation time, and acquiring message quantity interval time probability distribution or/and message quantity length probability distribution of the group in the another observation time according to the group data;
and counting the activity of each individual user in the group in the other observation time according to the multi-step recursion calculation mode of the group response and the message quantity interval time probability distribution or/and the message quantity length probability distribution of the group in the other observation time.
The process of obtaining the activity of each individual user in the group in the other observation time is the same as described above, and is not described herein again. According to the embodiment, the activity of each individual user in the group within different observation times can be obtained, and accordingly, the time period when the activity of each individual user is high and concentrated can be analyzed.
The cohort data for one or more further observation times may be selected for analysis.
In yet another embodiment of the present invention, the method of the embodiment of the present invention further includes:
and integrating the activity of each individual user in the group in the observation time with the activity of each individual user in at least one other observation time to obtain the integrated activity of each individual user in the group.
The integration manner may be, for example, to superimpose or average the activity data of each individual user in the group in different observation times or to integrate the activity data in other feasible manners, which is not limited in the present invention. And synthesizing the activity of each individual user in the group in the observation time and the activity of each individual user in at least one other observation time to obtain the comprehensive activity of each individual user in the group, wherein the data has more general significance than the activity data in one observation time.
According to the scheme of the embodiment of the invention, after the activity of each individual user in the group is obtained through statistics, the users can be distinguished according to the activity of the users, and further, the group information of the group response is sent by using the member with the activity of the mobilized group according to the service requirement so as to stimulate the activity of the group.
For example, users in a group can be classified into a collar-sleeve type, a responder type, an inquirer type, and a sharer type; the leader type is a person with extremely high contribution frequency and contribution quantity, creates a large amount of contents, attracts a plurality of members to resonate, and plays an important role in the formation and development of groups; the type of the responder represents identity, follow or object for topic contents which often contribute to other members, does not provide deep contents and does not actively contribute to the contents; the inquirer type is that a group is used as a place for learning and consultation, and frequently questions are put forward and answers are sought; the sharer type does not actively create topic contents, but actively participates in the group construction work, including the steps of enthusiastically answering questions posed by information inquirers and learners and communicating related experiences and hearts.
Based on the above method embodiment, the present invention further provides a device for acquiring the activity of the individual users in the group, as shown in fig. 7, the device 700 for acquiring the activity of the individual users in the group according to the embodiment of the present invention includes:
a probability distribution obtaining unit 710, configured to obtain group data within observation time, and obtain, according to the group data, message quantity interval time probability distribution or/and message quantity length probability distribution of the group within the observation time; the message volume interval time is the interval time of two continuous message volume records in the group, and the message volume length is the character length of each record of sending the message volume in the group;
a model establishing unit 720, configured to establish a group response model based on a rule that an adjustment change rate of the group response after being disturbed by the individual user is proportional to a difference between the current state and the balanced state;
the recursive computation unit 730 is configured to obtain a multi-step recursive computation mode of the group response according to the group response model; the multi-step recursion calculation mode is that the calculation result of the previous time period is used as the initial condition of the next time period, and the group response values after a plurality of time periods are obtained by recursion time periods;
and the individual activity counting unit 740 is configured to count the activity of each individual user in the group within the observation time according to the multi-step recursion calculation mode of the group response and the message quantity interval time probability distribution or/and the message quantity length probability distribution of the group within the observation time.
The probability distribution of the interval time of the group message volume and the probability distribution of the length of the group message volume obtained by the probability distribution obtaining unit 710 are subjected to power law distribution in the main part of the probability distribution under the log-log coordinate.
In one embodiment of the present invention, the apparatus 700 further comprises: and the group activity determining unit is suitable for determining the activity of the group according to the power exponent of the message quantity interval time probability distribution or/and the message quantity length probability distribution of the group in the observation time.
The group response model established by the model establishing unit 720 is:
where y is the current state of the group response, yeFor the equilibrium state of the group response, β is the coefficient, dy/dt is the rate of change of the adjustment of the group response, and t is time.
The multi-step recursive computation mode of the group response obtained by the recursive computation unit 730 is as follows:
where Vt represents an interval time length obtained by dividing the observation time T into n equal time periods, yeiBalance value, y, for group response at i time period0The initial value of the group response, β, is a coefficient.
In an embodiment of the present invention, the individual activity statistic unit 740 includes:
the group response value calculation module is used for inputting the message quantity interval time probability distribution and/or the message quantity length probability distribution of the group at the initial time in the observation time into the multi-step recursion calculation mode as initial values to obtain a group response value of each time period;
the individual user counting module is used for counting and obtaining active individual users and active frequency thereof in each time period according to message quantity interval time probability distribution or/and message quantity length probability distribution of the group in the observation time;
and the individual activity counting module is used for counting the activity of each individual user in the group within the observation time according to the group response value of each time period, the active individual users and the active frequency thereof.
In one embodiment, the individual user statistic module includes any one of the following sub-modules:
a first statistical submodule configured to input any one of a message size interval time probability distribution and a message size length probability distribution of the group at an initial time within the observation time as an initial value into the multi-step recursive computation mode;
the second statistic submodule is used for inputting the sum of the message quantity interval time probability distribution and the message quantity length probability distribution of the group at the initial time in the observation time into the multi-step recursion calculation mode as an initial value;
and the third statistical submodule is used for inputting the product of the message quantity interval time probability distribution and the message quantity length probability distribution of the group at the initial moment in the observation time into the multi-step recursion calculation mode as an initial value.
In another embodiment, the individual liveness statistics module may include:
the active data obtaining sub-module is configured to obtain active data of each individual user in each time period according to the group response value of each time period, the active individual users and the active frequency thereof, where the active data includes: the individual user active frequency of each time period and the corresponding group response value;
and the individual activity degree counting submodule is used for respectively carrying out statistical analysis on the activity data of each individual user in the group according to a set mode to obtain the activity degree of each individual user in the group in the observation time.
In another embodiment of the present invention, the apparatus 700 further comprises:
a second probability distribution obtaining unit, configured to obtain group data in another observation time, and obtain, according to the group data, message quantity interval time probability distribution or/and message quantity length probability distribution of the group in the another observation time;
and the second individual activity counting unit is used for counting the activity of each individual user in the group in the other observation time according to the multi-step recursion calculation mode of the group response and the message volume interval time probability distribution or/and the message volume length probability distribution of the group in the other observation time.
In yet another embodiment of the present invention, the apparatus 700 further comprises:
and the comprehensive unit is used for synthesizing the activity of each individual user in the group in the observation time and the activity of each individual user in at least one other observation time to obtain the comprehensive activity of each individual user in the group.
For the specific working modes of the modules in the above device embodiment of the present invention, reference may be made to the related contents in the above method embodiment of the present invention, and details are not described herein again.
In summary, according to the method and apparatus for acquiring activity of individual users in a group provided by the embodiments of the present invention, by acquiring group data within an observation time, a message volume interval time probability distribution or/and a message volume length probability distribution of the group within the observation time is acquired according to the group data; then, based on the rule that the adjustment change rate of the group response after being disturbed by the individual users is in direct proportion to the difference between the current state and the balance state, a group response model is established, then, a multi-step recursion calculation mode of the group response is obtained according to the group response model, and finally, the activity of each individual user in the group in the observation time is obtained through statistics according to the multi-step recursion calculation mode of the group response and the message quantity interval time probability distribution or/and the message quantity length probability distribution of the group in the observation time. The technical scheme of the invention is beneficial to mining the relation between the individual behaviors and the group behaviors, provides a solution for researching the active change process of individual users, and can conveniently acquire the members capable of mobilizing the group activity from the group, so that the members with the mobilized group activity can be used for sending the group information disturbing the group response to stimulate the group activity according to the service requirements.
The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (20)
1. A method for obtaining activity of individual users in a group, the method comprising:
acquiring group data in observation time, and acquiring message quantity interval time probability distribution or/and message quantity length probability distribution of the group in the observation time according to the group data; the message volume interval time is the interval time of two continuous message volume records in the group, and the message volume length is the character length of each record of sending the message volume in the group;
establishing a group response model based on the rule that the adjustment change rate of the group response after being disturbed by the individual user is in direct proportion to the difference between the current state of the group response and the balanced state of the group response;
obtaining a multi-step recursion calculation mode of group response according to the group response model; the multi-step recursion calculation mode is that the calculation result of the previous time period is used as the initial condition of the next time period, and the group response values after a plurality of time periods are obtained by recursion time periods;
and counting to obtain the activity of each individual user in the group in the observation time according to the multi-step recursion calculation mode of the group response and the message quantity interval time probability distribution or/and the message quantity length probability distribution of the group in the observation time.
2. The method of claim 1,
and the obtained probability distribution of the interval time of the group message volumes and the probability distribution of the length of the group message volumes are subjected to power law distribution in the main part of the probability distribution under the log-log coordinate.
3. The method of claim 2, wherein after obtaining the message volume interval probability distribution or/and the message volume length probability distribution of the group in the observation time according to the group data, the method further comprises:
and determining the liveness of the group according to the message quantity interval time probability distribution or/and the power exponent of the message quantity length probability distribution of the group in the observation time.
4. The method of claim 1,
the group response model is:
where y is the current state of the group response,yeFor the equilibrium state of the group response, β is the coefficient, dy/dt is the rate of change of the adjustment of the group response, and t is time.
5. The method of claim 4,
the multi-step recursion calculation mode of the group response is as follows:
where Vt represents an interval time length obtained by dividing the observation time T into n equal time periods, yeiBalance value, y, for group response at i time period0The initial value of the group response, β, is a coefficient.
6. The method of claim 1, wherein the step of statistically deriving the activity level of each individual user in the group during the observation time according to the multi-step recursive computation pattern of the group response and the message volume interval probability distribution or/and the message volume length probability distribution of the group during the observation time comprises:
inputting the message quantity interval time probability distribution and/or the message quantity length probability distribution of the group at the initial moment in the observation time into the multi-step recursion calculation mode as initial values to obtain a group response value of each time period;
according to the message volume interval time probability distribution or/and the message volume length probability distribution of the group in the observation time, counting to obtain active individual users and the active frequency of the active individual users in each time period;
and counting the activity of each individual user in the group within the observation time according to the group response value of each time period, the active individual users and the active frequency thereof.
7. The method according to claim 6, wherein the inputting the message volume interval time probability distribution and/or the message volume length probability distribution of the group at an initial time within the observation time as an initial value into the multi-step recursion calculation mode comprises any one of the following situations:
inputting any one of message quantity interval time probability distribution and message quantity length probability distribution of the group at the initial time in the observation time into the multi-step recursion calculation mode as an initial value; or,
inputting the sum of the message quantity interval time probability distribution and the message quantity length probability distribution of the group at the initial moment in the observation time as an initial value into the multi-step recursion calculation mode; or,
and inputting the product of the message volume interval time probability distribution and the message volume length probability distribution of the group at the initial moment in the observation time as an initial value into the multi-step recursion calculation mode.
8. The method of claim 6, wherein the counting the activity of each individual user in the group during the observation time according to the group response value of each time period, the active individual users and the active frequency thereof comprises:
obtaining the active data of each individual user in each time period according to the group response value of each time period, the active individual users and the active frequency thereof, wherein the active data comprises: the individual user active frequency of each time period and the corresponding group response value;
and respectively carrying out statistical analysis on the activity data of each individual user in the group according to a set mode to obtain the activity of each individual user in the group in the observation time.
9. The method according to any one of claims 1-8, further comprising:
acquiring group data in another observation time, and acquiring message quantity interval time probability distribution or/and message quantity length probability distribution of the group in the another observation time according to the group data;
and counting the activity of each individual user in the group in the other observation time according to the multi-step recursion calculation mode of the group response and the message quantity interval time probability distribution or/and the message quantity length probability distribution of the group in the other observation time.
10. The method of claim 9, further comprising:
and integrating the activity of each individual user in the group in the observation time with the activity of each individual user in at least one other observation time to obtain the integrated activity of each individual user in the group.
11. An apparatus for obtaining activity of individual users in a group, the apparatus comprising:
a probability distribution obtaining unit, configured to obtain group data within observation time, and obtain message quantity interval time probability distribution or/and message quantity length probability distribution of the group within the observation time according to the group data; the message volume interval time is the interval time of two continuous message volume records in the group, and the message volume length is the character length of each record of sending the message volume in the group;
the model establishing unit is used for establishing a group response model based on the rule that the adjustment change rate of the group response after being disturbed by the individual user is in direct proportion to the difference between the current state of the group response and the balanced state of the group response;
the recursive calculation unit is used for obtaining a multi-step recursive calculation mode of the group response according to the group response model; the multi-step recursion calculation mode is that the calculation result of the previous time period is used as the initial condition of the next time period, and the group response values after a plurality of time periods are obtained by recursion time periods;
and the individual activity counting unit is used for counting to obtain the activity of each individual user in the group in the observation time according to the multi-step recursion calculation mode of the group response and the message quantity interval time probability distribution or/and the message quantity length probability distribution of the group in the observation time.
12. The apparatus according to claim 11, wherein the probability distribution of the group message volume interval time and the probability distribution of the group message volume length obtained by the probability distribution obtaining unit both obey a power law distribution in a main part of the probability distribution in a log-log coordinate.
13. The apparatus of claim 12, further comprising:
and the group activity determining unit is suitable for determining the activity of the group according to the power exponent of the message quantity interval time probability distribution or/and the message quantity length probability distribution of the group in the observation time.
14. The apparatus of claim 11,
the group response model established by the model establishing unit is as follows:
where y is the current state of the group response, yeFor the equilibrium state of the group response, β is the coefficient, dy/dt is the rate of change of the adjustment of the group response, and t is time.
15. The apparatus of claim 14,
the multi-step recursive computation mode of the group response obtained by the recursive computation unit is as follows:
wherein Vt represents a division of the observation time T into n pieces of eachInterval time length of equal time period, yeiBalance value, y, for group response at i time period0The initial value of the group response, β, is a coefficient.
16. The apparatus of claim 11, wherein the individual liveness statistics unit comprises:
the group response value calculation module is used for inputting the message quantity interval time probability distribution and/or the message quantity length probability distribution of the group at the initial time in the observation time into the multi-step recursion calculation mode as initial values to obtain a group response value of each time period;
the individual user counting module is used for counting and obtaining active individual users and active frequency thereof in each time period according to message quantity interval time probability distribution or/and message quantity length probability distribution of the group in the observation time;
and the individual activity counting module is used for counting the activity of each individual user in the group within the observation time according to the group response value of each time period, the active individual users and the active frequency thereof.
17. The apparatus of claim 16, wherein the individual user statistics module comprises any one of the following sub-modules:
a first statistical submodule configured to input any one of a message size interval time probability distribution and a message size length probability distribution of the group at an initial time within the observation time as an initial value into the multi-step recursive computation mode;
the second statistic submodule is used for inputting the sum of the message quantity interval time probability distribution and the message quantity length probability distribution of the group at the initial time in the observation time into the multi-step recursion calculation mode as an initial value;
and the third statistical submodule is used for inputting the product of the message quantity interval time probability distribution and the message quantity length probability distribution of the group at the initial moment in the observation time into the multi-step recursion calculation mode as an initial value.
18. The apparatus of claim 16, wherein the individual liveness statistics module comprises:
the active data obtaining sub-module is configured to obtain active data of each individual user in each time period according to the group response value of each time period, the active individual users and the active frequency thereof, where the active data includes: the individual user active frequency of each time period and the corresponding group response value;
and the individual activity degree counting submodule is used for respectively carrying out statistical analysis on the activity data of each individual user in the group according to a set mode to obtain the activity degree of each individual user in the group in the observation time.
19. The apparatus of any one of claims 11-18, further comprising:
a second probability distribution obtaining unit, configured to obtain group data in another observation time, and obtain, according to the group data, message quantity interval time probability distribution or/and message quantity length probability distribution of the group in the another observation time;
and the second individual activity counting unit is used for counting the activity of each individual user in the group in the other observation time according to the multi-step recursion calculation mode of the group response and the message volume interval time probability distribution or/and the message volume length probability distribution of the group in the other observation time.
20. The apparatus of claim 19, further comprising:
and the comprehensive unit is used for synthesizing the activity of each individual user in the group in the observation time and the activity of each individual user in at least one other observation time to obtain the comprehensive activity of each individual user in the group.
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 CN105409162A (en) | 2016-03-16 |
CN105409162B true 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) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
CN101478450A (en) * | 2009-02-12 | 2009-07-08 | 阿里巴巴集团控股有限公司 | Monitoring method and server for network user active degree |
US7940702B1 (en) * | 2005-09-23 | 2011-05-10 | Avaya Inc. | Method and apparatus for allowing communication within a group |
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 |
-
2014
- 2014-06-30 CN CN201480012240.1A patent/CN105409162B/en active Active
- 2014-06-30 WO PCT/CN2014/081248 patent/WO2016000164A1/en active Application Filing
Patent Citations (4)
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 |
CN101478450A (en) * | 2009-02-12 | 2009-07-08 | 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 |
Also Published As
Publication number | Publication date |
---|---|
WO2016000164A1 (en) | 2016-01-07 |
CN105409162A (en) | 2016-03-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Keller | Predictors of the performance of project groups in R & D organizations | |
Gainous et al. | Tweeting to power: The social media revolution in American politics | |
CN109684446B (en) | Text semantic similarity calculation method and device | |
Anwaruddin | Teaching language, promoting social justice | |
Sutan et al. | Using social media as tools of social movement and social protest in omnibus law of job creation bill policy-making process in Indonesia | |
Poesen-Vandeputte et al. | Rich schools, poor schools. Hidden resource inequalities between primary schools | |
CN106570763A (en) | User influence evaluation method and system | |
CN105409162B (en) | A kind of method and apparatus obtaining individual consumer's liveness in group | |
Olebara | Mobile devices as tools for implementing digital classroom in Nigeria | |
Ebbers et al. | The co-evolution of social networks and selection system orientations as core constituents of institutional logics of future entrepreneurs at school | |
Crawford et al. | E-learning in context: An assessment of student inequalities in a university outreach program | |
Edwards | Consumer power and market control: Exploring consumer behaviour in affluent contexts (1946–1980) | |
Lal et al. | Analyzing the effect of social media on recruitment | |
Ulrich | Links between experiential learning and simulation & gaming | |
Charness | Technology as multiplier effect for an aging work force | |
Riivits-Arkonsuo et al. | Consumer online word-of-mouth–analysis through an experience pyramid model | |
Chigbuson et al. | Information and Communication Technology Skills Needed by Office Technology and Management Students for Self Sustenance and National Development | |
Gunduz | Relationship between Social Networks Adoption and Social Intelligence. | |
CN110020375B (en) | Evaluation method for influence of social network users | |
Sutarno | The Impact of Social Media Usage On Students’ Academic Performance (SAP) | |
Kandpal et al. | A study on perception of students towards use of social media in education | |
Aruguete | Las hipotesis de la agenda setting en el nuevo entorno mediatico. | |
CN111047194A (en) | System and method for managing client creation platform | |
Gurin et al. | Discursive Structure of Media Content Discussion by Users of Online Social Networks | |
Lancaster et al. | Personal response systems as a learning aid in an epidemiology course for postgraduate statistics students |
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 |
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. |
|
CP02 | Change in the address of a patent holder |