CN111915166A

CN111915166A - Method and device for determining group activity

Info

Publication number: CN111915166A
Application number: CN202010684143.4A
Authority: CN
Inventors: 张兴; 夏恒毅; 张佳鑫
Original assignee: Beijing University of Posts and Telecommunications
Current assignee: Beijing University of Posts and Telecommunications
Priority date: 2020-07-16
Filing date: 2020-07-16
Publication date: 2020-11-10

Abstract

The embodiment of the invention provides a method and a device for determining group activity, relates to the field of data analysis, and can more accurately determine the group activity. The embodiment of the invention comprises the following steps: and constructing a criterion layer judgment matrix and an index layer judgment matrix aiming at each group of preset ratio sets. And then determining the weight of each item type according to the feature vector corresponding to the maximum feature root of the criterion layer judgment matrix, determining the weight of each index according to the feature vector corresponding to the maximum feature root of the index layer judgment matrix and the weight of each item type, and generating the weight vector corresponding to the group of preset ratio sets based on the weight of each index. And then calculating the sum of included angles of the weight vector and other weight vectors aiming at each weight vector included in the weight vector set, and determining the weight vector with the minimum sum of included angles as a target vector. And then, calculating the weighted sum of the attribute values of the indexes included in the designated group based on the weight of each index included in the target vector to obtain the activity of the designated group.

Description

Method and device for determining group activity

Technical Field

The invention relates to the technical field of data analysis, in particular to a method and a device for determining group liveness.

Background

The determination of group liveness is important in order to facilitate the management of group activities and to provide a good reference for the people who want to join the group. For example, volunteers in selecting the volunteer groups they wish to join will be very helpful in their decision making if the activity of the volunteer groups is available.

At present, there is no unified standard for determining the liveness of a group, and generally, a person determines the liveness of the group by experience according to the attribute parameters under each index of the group. For example, when a volunteer selects a volunteer group to join, it is determined empirically whether the volunteer group is active or not according to the number of projects each volunteer group participates in, the number of members included in the group, the accumulated duration of participating projects, and the like.

However, because the preference, mental state and level of each person are different, the method for determining the group activity is too subjective, and it is difficult to accurately determine the group activity.

Disclosure of Invention

The embodiment of the invention aims to provide a method and a device for determining group activity, so as to determine the group activity more accurately. The specific technical scheme is as follows:

in a first aspect, an embodiment of the present invention provides a method for determining group liveness, where each group includes multiple item categories, each item category includes multiple indicators, and each indicator is used to represent an attribute of the group, where the method includes:

aiming at each group of preset ratio sets, constructing a criterion layer judgment matrix and an index layer judgment matrix of the group; each set of preset ratios comprises: the method comprises the following steps that a first preset ratio of every two project categories to the importance of the group activity is determined, and a second preset ratio of every two indexes under the same project category to the importance of the group activity is determined; the criterion layer judgment matrix comprises a plurality of first preset ratios; the index layer judgment matrix comprises a plurality of second preset ratios;

determining the weight of each item category of the group according to the characteristic vector corresponding to the maximum characteristic root of the criterion layer judgment matrix of the group;

determining the weight of each index according to the feature vector corresponding to the maximum feature root of the index layer judgment matrix of the group and the weights of all item categories of the group, and generating the weight vector corresponding to the group of preset ratio sets on the basis of the determined weights of all indexes;

calculating the sum of included angles between each weight vector and other weight vectors included in the weight set aiming at each weight vector included in the weight vector set, and determining the weight vector with the smallest sum of included angles between the weight vector and other weight vectors included in the weight set as a target vector; the weight vector set comprises weight vectors corresponding to each group of preset ratio sets;

and calculating the weighted sum of the attribute values of the indexes included in the appointed group based on the weight of each index included in the target vector to obtain the activity of the appointed group.

Optionally, before determining the weight of each item category of the group according to the feature vector corresponding to the maximum feature root of the criterion layer determination matrix of the group, the method further includes:

calculating a first consistency index of the criterion layer judgment matrix of the group, and taking a quotient of the first consistency index and a first average random consistency index as a first consistency proportion; the first average random consistency index is a preset numerical value corresponding to the rank of the judgment matrix of the criterion layer of the group;

calculating a second consistency index of the index layer judgment matrix of the group, and taking a quotient of the second consistency index and a second average random consistency index as a second consistency proportion; the second average random consistency index is a preset numerical value corresponding to the rank of the judgment matrix of the index layer of the group;

and if the first consistency ratio and the second consistency ratio are both smaller than a preset ratio, executing the step of determining the weight of each item category of the group according to the feature vector corresponding to the maximum feature root of the criterion layer judgment matrix of the group.

Optionally, the determining the weight of each index according to the feature vector corresponding to the maximum feature root of the set of index layer determination matrices and the weights of each item category of the set includes:

normalizing the feature vector corresponding to the maximum feature root of the index layer judgment matrix of the group to obtain an initial weight corresponding to each index;

and for each index, taking the product of the initial weight corresponding to the index and the weight of the item category to which the index belongs in the group as the weight of the index.

Optionally, the number of the specified communities is at least two, and the weight vector set further includes: an entropy weight vector; before calculating, for each weight vector included in the weight vector set, a sum of included angles between the weight vector and other weight vectors included in the weight vector set, and determining a weight vector having a smallest sum of included angles with other weight vectors included in the weight vector set as a target vector, the method further includes:

respectively calculating the ratio of the attribute value number of the index to the total number of the designated groups of each designated group aiming at each index;

calculating the entropy value of the index according to the ratio of the attribute value number of the index of each designated group to the total number of designated groups;

and calculating the weight of each index according to the entropy value of each index, and forming the weight of each index into the entropy weight vector.

Optionally, the calculating the weight of each index according to the entropy of each index includes:

the weight of each index is calculated by the following formula:

wherein, w_jIs the weight of the index j, H_jIs the entropy value of the index j and n is the total number of indexes included by the given community.

Optionally, the item categories of the community include the community and the item; the indicators of the community include: at least two of group establishment time, group click rate, group participation activity accumulated time and group formal member number; the indicators of the items include: the number of participating items, the click rate of the items, the accumulated time of the items, the per-capita time of the items and the number of participating persons of the items.

In a second aspect, an embodiment of the present invention provides an apparatus for determining group liveness, each group including a plurality of item categories, each item category including a plurality of indicators, each indicator being used for representing an attribute of the group, the apparatus including:

the building module is used for building a criterion layer judgment matrix and an index layer judgment matrix of each group aiming at each group of preset ratio sets; each set of preset ratios comprises: the method comprises the following steps that a first preset ratio of every two project categories to the importance of the group activity is determined, and a second preset ratio of every two indexes under the same project category to the importance of the group activity is determined; the criterion layer judgment matrix comprises a plurality of first preset ratios; the index layer judgment matrix comprises a plurality of second preset ratios;

the determining module is used for determining the weight of each item category of the group according to the characteristic vector corresponding to the maximum characteristic root of the group of criterion layer judgment matrixes constructed by the constructing module;

the determining module is further configured to determine a weight of each index according to the feature vector corresponding to the maximum feature root of the set of index layer judgment matrices constructed by the constructing module and the weights of the item categories of the set, and generate a weight vector corresponding to the set of preset ratio sets based on the determined weights of the indexes;

the calculation module is used for calculating the sum of included angles between each weight vector included in the weight vector set and other weight vectors included in the weight set, and determining the weight vector with the smallest sum of included angles between the weight vector and other weight vectors included in the weight set as a target vector; the weight vector set comprises weight vectors corresponding to each group of preset ratio sets;

the calculation module is further configured to calculate, based on the weights of the indicators included in the target vector, a weighted sum of attribute values of the indicators included in the designated group, so as to obtain the liveness of the designated group.

Optionally, the apparatus further comprises: an execution module;

the calculation module is further configured to calculate a first consistency index of the criterion layer determination matrix of the group before determining a weight of each item category of the group according to the feature vector corresponding to the maximum feature root of the criterion layer determination matrix of the group, and take a quotient of the first consistency index and the first average random consistency index as a first consistency ratio; the first average random consistency index is a preset numerical value corresponding to the rank of the judgment matrix of the criterion layer of the group;

the calculation module is further configured to calculate a second consistency index of the set of index layer determination matrices, and use a quotient of the second consistency index and a second average random consistency index as a second consistency ratio; the second average random consistency index is a preset numerical value corresponding to the rank of the judgment matrix of the index layer of the group;

and the execution module is used for executing the step of determining the weight of each item category of the group according to the feature vector corresponding to the maximum feature root of the criterion layer judgment matrix of the group when the first consistency ratio and the second consistency ratio are both smaller than a preset ratio.

Optionally, the determining module is specifically configured to:

Optionally, the number of the specified communities is at least two, and the weight vector set further includes: an entropy weight vector; the computing module is further configured to:

before calculating the sum of included angles between each weight vector and other weight vectors included in the weight set and determining the smallest sum of included angles between each weight vector and other weight vectors included in the weight set as a target vector, respectively calculating the ratio of the number of the attribute values of each index of each designated group to the total number of the designated groups for each index;

Optionally, the calculation module is specifically configured to:

the weight of each index is calculated by the following formula:

In a third aspect, an embodiment of the present invention provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, where the processor and the communication interface complete communication between the memory and the processor through the communication bus;

a memory for storing a computer program;

and the processor is used for realizing the steps of the method for determining the activity of any group when the processor executes the program stored in the memory.

In a fourth aspect, the embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored, and the computer program, when executed by a processor, implements the steps of any one of the above methods for determining community liveness.

In a fifth aspect, embodiments of the present invention further provide a computer program product containing instructions, which when run on a computer, cause the computer to perform any of the above methods for determining community liveness.

The technical scheme of the embodiment of the invention can at least bring the following beneficial effects: the sum of included angles between the target vector and other vectors in the weight vector set is minimum, namely the target vector is closer to the average value of other vectors, so that the weight of each index included in the target vector is more objective and more universal, and the accuracy of determining the group liveness is higher by using the weight of each index included in the target vector.

Of course, not all of the advantages described above need to be achieved at the same time in the practice of any one product or method of the invention.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other embodiments can be obtained by using the drawings without creative efforts.

Fig. 1 is a flowchart of a method for determining group liveness according to an embodiment of the present invention;

FIG. 2 is a flowchart of another method for determining group liveness according to an embodiment of the present invention;

FIG. 3 is a schematic structural diagram of a group activity determination model according to an embodiment of the present invention;

fig. 4 is a schematic structural diagram of a device for determining group liveness according to an embodiment of the present invention;

fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

In order to determine the group activity more accurately, an embodiment of the present invention provides a method for determining the group activity, which is applied to an electronic device, where the electronic device may be a device with a data processing function, such as a mobile phone, a computer, and a tablet computer. Referring to fig. 1, the method comprises the steps of:

step 101, aiming at each group of preset ratio sets, a criterion layer judgment matrix and an index layer judgment matrix of the group are constructed.

Wherein, every group of preset ratio sets includes: the method comprises the following steps that a first preset ratio of every two project categories to the importance of the group activity is determined, and a second preset ratio of every two indexes under the same project category to the importance of the group activity is determined; the criterion layer judgment matrix comprises a plurality of first preset ratios; the index layer judgment matrix comprises a plurality of second preset ratios.

In an embodiment of the invention, each community comprises a plurality of item categories, each item category comprising a plurality of indicators, each indicator being for representing an attribute of the community. It can be understood that the index is a basic factor for evaluating the group activity, and can be understood as each dimension for evaluating the group activity.

In embodiments of the present invention, the project categories of the community may include communities and projects. The indicators of the community may include: at least two of group establishment time, group click rate, group participation activity accumulated time and group formal member number. The indicators of the items may include: the number of participating items, the click rate of the items, the accumulated time of the items, the per-capita time of the items and the number of participating persons of the items. Optionally, the item category and the index may also include other factors, which may be determined according to an actual application scenario, and this is not specifically limited in the embodiment of the present invention.

The group establishment time refers to a time length of the group establishment on the current day. The volume of a community click refers to the number of times a community is clicked by a user. The item click volume refers to the number of times an item is clicked by a user. The project aggregate duration refers to the aggregate duration of time that the community engages in each project activity. The term "average duration" refers to the ratio of the accumulated duration of a project to the number of participants in the project.

Optionally, each set of preset ratios may be manually set empirically in advance.

In the embodiment of the present invention, the criterion layer determination matrix a ═ a (a)_uv)_p×pWherein a is_uvA first ratio representing the importance of the u-th item category and the v-th item category for determining the group liveness, u, v being 1, 2, 3, … …, p, p being the total number of item categories.

An item type index layer determination matrix B ═ B_uv)_q×qWherein b is_uvAnd a second ratio of the u-th index and the v-th index representing the type of the item to the importance of determining the group activity, wherein u, v is 1, 2, 3, … …, q, and q is the total number of indexes of the item category.

a_uvAnd b_uvThe numerical values and numerical values of (b) are shown in Table 1.

TABLE 1

Numerical value	Means of
		1	Indicating that factor u and factor v are of equal importance for determining group liveness
3	Indicating that factor u is slightly more important than factor v for determining group liveness
		5	Indicating that factor u is significantly more important than factor v for determining group liveness
7	Indicating that factor u is more important than factor v for determining group liveness
		9	Indicating that factor u is extremely important over factor v for determining group liveness
2，4，6，8	Median value of the above two adjacent judgments

Where each even number in the values of table 1 is the median of its adjacent odd numbers, the median of the significance represented by its adjacent odd numbers. For example, scale 4, indicates that factor u is more slightly important and less significantly important than factor v. It can be seen that the larger the number, the larger the gap in importance between factor u and factor v.

Optionally, a_vu＝1/a_uv，b_vu＝1/b_uv。

And 102, determining the weight of each item category of the group according to the feature vector corresponding to the maximum feature root of the criterion layer judgment matrix of the group.

It can be understood that the feature vector corresponding to the maximum feature root of the criterion layer determination matrix is a one-dimensional vector, and each element value of the feature vector corresponds to the weight of one item category. Optionally, the element values of the feature vector may be normalized, and then each normalized element value may be used as a weight of one item category.

The method for obtaining the maximum feature root for the matrix belongs to the prior art, and is not described herein again.

Step 103, determining the weight of each index according to the feature vector corresponding to the maximum feature root of the set of index layer judgment matrices and the weights of the item categories of the set, and generating the weight vector corresponding to the set of preset ratio sets based on the determined weights of the indexes.

In one embodiment, the feature vector corresponding to the maximum feature root of the set of indicator layer determination matrices may be normalized to obtain an initial weight corresponding to each indicator. Then, for each index, taking the product of the initial weight corresponding to the index and the weight of the item category to which the index belongs in the group as the weight of the index.

For example, a community includes two categories of items, community and item, respectively, with the community having a weight of 0.6 and the item having a weight of 0.4. The group comprises two indexes which are the group click rate and the group formal member number respectively. The project comprises two indexes, namely the number of participating projects and the accumulated duration of the project. The initial weight of the group click rate is 0.7, the initial weight of the formal group member number is 0.3, the initial weight of the participation item number is 0.4, and the initial weight of the accumulated item duration is 0.6. The weight of the group click volume is 0.7 × 0.6 to 0.42, the weight of the group official member number is 0.3 × 0.6 to 0.18, the weight of the participation item number is 0.4 × 0.4 to 0.16, and the weight of the item accumulated time length is 0.6 × 0.4 to 0.24.

It can be understood that, the weight of the index is set as the product of the initial weight and the weight of the item category to which the index belongs, so that the total weight of each index included in the group is 1, the weight normalization of each index is realized, and the importance of each index for determining the group activity is more favorably measured and compared.

And 104, calculating the sum of included angles between the weight vector and other weight vectors included in the weight set aiming at each weight vector included in the weight vector set, and determining the weight vector with the smallest sum of included angles with other weight vectors included in the weight set as a target vector.

The weight vector set comprises weight vectors corresponding to each group of preset ratio sets.

In one embodiment, the target vector is determined by equation (1).

Where MAX denotes the maximum value, x_*＝(x_*1,x_*2…x_*n)^T∈Eⁿ，x_*The average value of each weight vector included in the weight vector set, n is the total number of indexes of the specified group, x_i＝(x_i1,x_i2…x_in)^T∈EⁿM is the number of weight vectors comprised by the weight vector, p_maxIs x^Tx maximum root of features, x being all x_iA matrix of components.

And 105, calculating the weighted sum of the attribute values of the indexes included in the appointed group based on the weight of each index included in the target vector to obtain the activity of the appointed group.

For example, the target vector is [0.5,0.2,0.3], which represents the weight of the group official member number, the number of participating items, and the item click amount, respectively. The formal number of members of a group is 50, the number of participating items is 5, the click rate of the items is 20, and the activity of the group is as follows: 0.5 × 50+0.2 × 5+0.3 × 20 ═ 32.

In the embodiment of the present invention, before determining the weight of the item category in step 102, a consistency check may be performed on the determination matrix, and a specific checking manner includes the following steps.

Step one, calculating a first consistency index of the criterion layer judgment matrix of the group, and taking a quotient of the first consistency index and the first average random consistency index as a first consistency proportion.

In one embodiment, the first consistency index of the criterion layer judgment matrix may be calculated using formula (2).

Wherein CI is a first consistency index, λ_max(A) And judging the maximum characteristic root of the matrix for the criterion layer, wherein n is the order of the criterion layer judgment matrix.

It can be understood that the average random consistency index is obtained by repeatedly performing the calculation of the characteristic value of the random judgment matrix for a plurality of times (more than 500 times) and then taking the arithmetic mean. In an embodiment of the present invention, the first average random consistency index (RI) is a preset value corresponding to the rank of the criterion layer determination matrix of the group. The first consistency ratio CR is CI/RI.

The corresponding relationship between RI and the order of the judgment matrix is shown in table 2.

TABLE 2

Order of matrix	1	2	3	4	5	6	7	8	9	10
											RI	0	0	0.58	0.90	1.12	1.24	1.32	1.41	1.45	1.49

And step two, calculating a second consistency index of the index layer judgment matrix of the group, and taking the quotient of the second consistency index and the second average random consistency index as a second consistency proportion.

And the second average random consistency index is a preset numerical value corresponding to the rank of the judgment matrix of the index layer of the group.

The method for calculating the second consistency ratio is the same as the method for calculating the first consistency ratio, and reference may be made to the related description in step one, which is not repeated herein.

And step three, if the first consistency proportion and the second consistency proportion are both smaller than the preset proportion, executing a step of determining the weight of each item category of the group according to the feature vector corresponding to the maximum feature root of the criterion layer judgment matrix of the group.

For example, the preset ratio may be 0.1, and when the first consistency ratio and the second consistency ratio are both less than 0.1, the step 102 is performed.

The technical scheme of the embodiment of the invention can also bring the following beneficial effects: due to the complexity of the objective world and the diversity of the human recognition problems, and the fact that the two elements are compared with each other and have no fixed reference, people can make judgment of common sense violation when comparing. For example, the importance of factor a > factor B, the importance of factor B > factor C, and the importance of factor C > factor a in the judgment matrix, but it should be the importance of factor a > factor C according to the common sense, which indicates that the judgment matrix has the judgment of the occurrence of the order inconsistency.

Such judgment against common sense appears, so that the judgment matrixes are not completely consistent, and therefore, consistency check needs to be performed on the judgment matrixes, and the judgment matrixes passing the consistency check are kept, so as to obtain the weight of each index. The obtained weight of each index is more accurate.

In this embodiment of the present invention, the number of designated communities is at least two, and the set of weight vectors may further include: an entropy weight vector. The entropy weight vector is obtained by the following steps.

Step 1, calculating the ratio of the attribute value number of the index to the total number of the designated groups of each designated group aiming at each index.

In one embodiment, it is assumed that there are m designated groups, each designated group has n indexes, and a matrix formed by attribute values of the indexes of each designated group is B ═ B (B)_ij)_m×n，b_ijAn attribute value of index j of a specified group m. Normalizing B to obtain the ratio of the attribute value number of the index of each designated group to the total number of designated groups

From d_ijThe composition matrix D ═ D_ij)_m×n。

For example: suppose there are 5 groups, each with 7 indicesFor one of the indexes 1, the attribute values of each community are: 1,1,2,2,2. It can be seen that there are 2 for attribute value 1 and 3 for attribute value 2, then d₁₁＝2/5，d₂₁＝2/5，d₃₁＝3/5，d₄₁＝3/5，d₅₁＝3/5。

And 2, calculating the entropy value of the index according to the ratio of the attribute value number of the index of each group to be specified to the total number of the specified groups.

In one embodiment, the entropy value of the indicator is calculated using equation (3).

Wherein H_jIs the entropy of index j, n is the total number of indexes included in the specified group, m is the number of specified groups, d_ijThe ratio of this attribute value of index j to the total number of designated communities for designated community i.

And 3, calculating the weight of each index according to the entropy value of each index, and forming the weight of each index into the entropy weight vector.

In one embodiment, the weight of each index is calculated by equation (4).

It will be appreciated that entropy is a measure of the degree of disorder of the system. If the entropy of an index is smaller, the amount of information provided by the index is larger, the more it plays a role in determining the group liveness, so the weight of the index is higher.

The technical scheme of the embodiment of the invention can also bring the following beneficial effects: since the decision level of a person depends not only on his professional level, experience, knowledge and comprehensive abilities, but also is closely related to his mental state, mood and preference at the time of decision. Therefore, the maximum value of the decision reliability of the manually preset ratio set for the importance of the factor is 1, or the uncertainty of the manual decision is not 0. The multiple groups of preset ratio sets can reduce errors caused by personal factors such as the level, experience, knowledge and mental state preference of a certain person. Therefore, in the embodiment of the present invention, an Analytic Hierarchy Process (AHP) is used to determine the weight vector for each Group of preset ratio sets, a non-subjective entropy weight method is used to obtain the entropy weight vector, and a Group eigen root (GEM) method is used to determine the target vector from a plurality of groups of weight vectors including the entropy weight vector.

Referring to fig. 2, a method for determining community liveness provided by an embodiment of the present invention is described below by way of an example.

Step 201, the volunteer service data of each volunteer service group is sorted and screened to obtain the attribute value of each index.

The embodiment of the invention adopts Chinese volunteer service data to evaluate volunteer service groups, and screens the data meeting preset conditions from the Chinese volunteer service data to obtain 94512 active groups which respectively come from 415 cities across the country, 1025241 volunteer service items and 44339850 participators, and the accumulated time reaches 797786241.4 hours. For example, the preset condition may be that the attribute value of each index is not null.

The volunteer service data presence field of each volunteer service community includes: the establishment time of the volunteer group, the number of volunteers in streets of province, city, district and county, the total volunteer duration, the contact way and the like; volunteers' names, ages, streets of province, city, district and county, volunteer duration, volunteer items, contact ways and the like; volunteer project name, time, number of participating people, location, total duration, etc. The attribute values of the indexes required by the embodiment of the invention are screened out.

Step 202, obtaining a weight vector corresponding to each group of preset ratio sets by using an AHP analytic hierarchy process.

As shown in FIG. 3, the model for determining community liveness provided by the embodiment of the invention comprises a target layer, a criterion layer and an index layer. Wherein the target layer is used for determining the activity of the volunteer service group, and the criterion layer is used for determining the weight of two project categories (group (O) and project (P)). The index layer is used for determining the weight of each index. Wherein, the group includes the indexes: group establishment time (Y), group click rate (H), group participation activity accumulated time (T), and group official member number (M). The items include the following indexes: the number (c) of participating items, the click rate (h) of the items, the accumulated time (t) of the items, the average time (a) of the item persons and the number (m) of the item participants.

According to each group of preset ratio sets, constructing a criterion layer judgment matrix of the group:

	O	P
			O	1	1/3
P	3	1

and constructing two index layer judgment matrixes of each group according to each group of preset ratio sets. One is the index layer judgment matrix of the community, and the other is the index layer judgment matrix of the project.

The index layer of the community judges the matrix:

	Y	H	T	M
					Y	1	1/3	1/7	1/7
H	3	1	1/5	1/5
					T	7	5	1	1
M	7	5	1	1

an index layer judgment matrix of the item:

	c	h	t	a	m
						c	1	1	1/5	1/3	1/3
h	1	1	1/5	1/3	1/3
						t	5	5	1	3	3
a	3	3	1/3	1	1
						m	3	3	1/3	1	1

it can be understood that the above judgment matrix is shown in a table form, and the ratio of the importance of the two factors for determining the group activity can be more intuitively shown.

It can be understood that the AHP analytic hierarchy process carries out formal expression and processing on the artificial subjective judgment, and gradually eliminates the subjectivity, thereby converting into objective description as much as possible. And judging whether the preset importance ratio value is reasonable or not, and carrying out consistency check on the judgment matrix to obtain a consistency ratio CR shown in a table 3.

TABLE 3

And after the test, the CR of each judgment matrix is less than 0.1, namely the consistency test is passed, the weight of each index is further obtained, and a weight vector is generated according to a set of determined weights. As shown in table 4.

TABLE 4

Index (I)	Y	H	T	M	c	h	t	a	m
										Weight of	0.012	0.024	0.098	0.098	0.056	0.056	0.358	0.149	0.149

For a specific method for determining the weight of the item category and the weight of the index, reference may be made to the above description, and details are not repeated here.

And step 203, obtaining an entropy weight vector based on an entropy weight method.

The weights of the indices were obtained by the entropy weight method, as shown in table 5.

TABLE 5

For a specific method for determining the entropy weight vector, reference may be made to the above description, and details are not repeated here.

Step 204, determining a target vector from the weight vector and the entropy weight vector corresponding to each group of preset ratio sets.

It can be understood that the error of each ratio set can be balanced by the multiple groups of preset ratio sets, the cognition of each person on the volunteer service is different, the evaluation emphasis is different, and the states of spirit, mood and the like during evaluation are also different. If a group of preset ratios is used, the fluctuation of the determined group activity is large, so that a plurality of groups of preset ratios are adopted, and the entropy weight method is synthesized for comprehensive evaluation, so that the determined target vector can more objectively reflect the importance of each factor.

For example, the target vector includes the weights for each live broadcast as shown in table 6.

TABLE 6

Index (I)	Y	H	T	M	c	h	t	a	m
										Weight of	0.035	0.066	0.193	0.15	0.108	0.055	0.197	0.111	0.084

For a specific method for determining the target vector, reference may be made to the above description, and details are not repeated here.

Step 205, calculating the weighted sum of the attribute values of the indexes included in the volunteer service group based on the weights of the indexes included in the target vector to obtain the activity of the volunteer service group.

For example, in connection with Table 7, each row of Table 7 represents data for one volunteer service group.

TABLE 7

Y	H	T	M	c	h	t	a	m	Group liveness
										6	1925	51982	1115	10	2218	39400.5	29.7362	1325	89.6873516
7	24605	586496	17297	544	53219	290641	16.5137	17600	98.72286106
										7	756	708	500	2	77	378	7.41176	51	78.70145742
6	419	1490	30	5	0	1166	11.66	100	78.10643389
										6	668	1118	80	2	770	268	7.44444	36	77.87601262

It can be understood that, for each volunteer service group, the sum of the product of the attribute value of each index and the weight thereof is calculated to obtain the activity of the volunteer service group.

Based on the same inventive concept, corresponding to the above method embodiment, the embodiment of the present invention further provides an apparatus for determining group liveness, where each group includes a plurality of item categories, each item category includes a plurality of indexes, and each index is used for representing an attribute of the group. As shown in fig. 4, the apparatus includes: a construction module 401, a determination module 402 and a calculation module 403;

a building module 401, configured to build, for each group of preset ratio sets, a criterion layer judgment matrix and an index layer judgment matrix of the group; each set of preset ratios comprises: the method comprises the following steps that a first preset ratio of every two project categories to the importance of the group activity is determined, and a second preset ratio of every two indexes under the same project category to the importance of the group activity is determined; the criterion layer judgment matrix comprises a plurality of first preset ratios; the index layer judgment matrix comprises a plurality of second preset ratios;

a determining module 402, configured to determine a weight of each item category of the group according to a feature vector corresponding to a maximum feature root of the group of criterion layer determination matrices constructed by the constructing module 401;

the determining module 402 is further configured to determine a weight of each index according to the feature vector corresponding to the maximum feature root of the set of index layer determination matrices constructed by the constructing module 401 and the weights of the item categories of the set, and generate a weight vector corresponding to the set of preset ratio sets based on the determined weights of the indexes;

a calculating module 403, configured to calculate, for each weight vector included in the weight vector set, a sum of included angles between the weight vector and other weight vectors included in the weight set, and determine, as a target vector, a weight vector having a smallest sum of included angles between the weight vector and other weight vectors included in the weight set; the weight vector set comprises weight vectors corresponding to each group of preset ratio sets;

the calculating module 403 is further configured to calculate a weighted sum of attribute values of the indexes included in the designated group based on the weights of the indexes included in the target vector, so as to obtain the activity of the designated group.

Optionally, the apparatus further comprises: an execution module;

the calculating module 403 is further configured to calculate a first consistency index of the criterion layer determination matrix of the group before determining a weight of each item category of the group according to the feature vector corresponding to the maximum feature root of the criterion layer determination matrix of the group, and take a quotient of the first consistency index and the first average random consistency index as a first consistency ratio; the first average random consistency index is a preset numerical value corresponding to the rank of the judgment matrix of the criterion layer of the group;

the calculating module 403 is further configured to calculate a second consistency index of the index layer judgment matrix of the group, and use a quotient of the second consistency index and the second average random consistency index as a second consistency ratio; the second average random consistency index is a preset numerical value corresponding to the order of the judgment matrix of the index layer of the group;

and the execution module is used for executing the step of determining the weight of each item category of the group according to the feature vector corresponding to the maximum feature root of the criterion layer judgment matrix of the group when the first consistency ratio and the second consistency ratio are both smaller than the preset ratio.

Optionally, the determining module 402 is specifically configured to:

Optionally, the number of specified communities is at least two, and the weight vector set further includes: an entropy weight vector; a calculation module 403, further configured to:

before calculating the sum of included angles between the weight vector and other weight vectors included in the weight set aiming at each weight vector included in the weight vector set and determining the smallest sum of included angles between the weight vector and other weight vectors included in the weight set as a target vector, respectively calculating the ratio of the number of the attribute values of the index to the total number of the designated groups of each designated group aiming at each index;

and calculating the weight of each index according to the entropy value of each index, and forming the weight of each index into an entropy weight vector.

Optionally, the calculating module 403 is specifically configured to:

the weight of each index is calculated by the following formula:

An embodiment of the present invention further provides an electronic device, as shown in fig. 5, which includes a processor 501, a communication interface 502, a memory 503 and a communication bus 504, where the processor 501, the communication interface 502 and the memory 503 complete mutual communication through the communication bus 504,

a memory 503 for storing a computer program;

the processor 501 is configured to implement the method steps in the above-described embodiment when executing the program stored in the memory 503.

The communication bus mentioned in the electronic device may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.

The communication interface is used for communication between the electronic equipment and other equipment.

The Memory may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.

The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components.

In yet another embodiment of the present invention, a computer-readable storage medium is further provided, in which a computer program is stored, and the computer program, when executed by a processor, implements the steps of any of the above-mentioned methods for determining community liveness.

In yet another embodiment, a computer program product containing instructions is provided, which when run on a computer causes the computer to perform the method for determining community liveness of any of the above embodiments.

In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.

It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, as for the apparatus embodiment, since it is substantially similar to the method embodiment, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

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

1. A method for determining group liveness, wherein each group comprises a plurality of item categories, each item category comprising a plurality of indicators, each indicator being indicative of an attribute of the group, the method comprising:

2. The method of claim 1, wherein before determining the weight of each item category of the group according to the feature vector corresponding to the largest feature root of the criterion layer judgment matrix of the group, the method further comprises:

3. The method according to claim 1, wherein the determining the weight of each index according to the eigenvector corresponding to the largest feature root of the index layer judgment matrix of the group and the weights of the item categories of the group comprises:

4. The method of claim 1, wherein the number of the specified communities is at least two, and wherein the set of weight vectors further comprises: an entropy weight vector; before calculating, for each weight vector included in the weight vector set, a sum of included angles between the weight vector and other weight vectors included in the weight vector set, and determining a weight vector having a smallest sum of included angles with other weight vectors included in the weight vector set as a target vector, the method further includes:

5. The method of claim 4, wherein calculating the weight for each index based on the entropy of each index comprises:

the weight of each index is calculated by the following formula:

6. The method of any of claims 1-5, wherein the item categories of communities include communities and items; the indicators of the community include: at least two of group establishment time, group click rate, group participation activity accumulated time and group formal member number; the indicators of the items include: the number of participating items, the click rate of the items, the accumulated time of the items, the per-capita time of the items and the number of participating persons of the items.

7. An apparatus for determining group liveness, wherein each group includes a plurality of item categories, each item category including a plurality of indicators, each indicator indicating an attribute of the group, the apparatus comprising:

8. The apparatus of claim 7, further comprising: an execution module;

9. The apparatus of claim 7, wherein the determining module is specifically configured to:

10. The apparatus of claim 7, wherein the number of the designated communities is at least two, and wherein the set of weight vectors further comprises: an entropy weight vector; the computing module is further configured to: