A multi-dimensional data analysis method and system for social group decision support

Abstract

The application discloses a kind of multidimensional data analysis method and system of social group decision support, it is related to social group activity planning field, main scheme is: through basic activity data, activity heat estimation value is calculated;Participation depth index is calculated by depth data and basic activity data;Conversion funnel efficiency value is calculated by conversion data and basic activity data;Retention network influence value is calculated by retention network data;Through the comprehensive analysis of activity heat estimation value, participation depth index, conversion funnel efficiency value and retention network influence value, decision support degree is obtained;Decision support degree is compared with the threshold in decision support threshold set, whether it should be supported according to the comparison result is judged.Solve the technical problem that the decision basis is not comprehensive enough in prior art, it is difficult to comprehensively and accurately evaluate the comprehensive effect of activity, leading to the technical problem that the decision basis is not comprehensive, and the evaluation accuracy is poor.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of social group activity planning, in particular to a multi-dimensional data analysis method for social group decision support. BACKGROUND

[0002] In today's society, social groups undertake a wide variety of functions, from cultural heritage, public welfare activities to professional field exchanges and cooperation, etc. The development of social group activities has far-reaching significance for consolidating group strength and promoting social progress. With the development of society and the growing demand for spiritual and cultural, professional interaction, etc., the scale and frequency of social group activities are also on the rise. Effective decision support analysis methods for social group activities can help social groups accurately position the direction of activities, optimize the content and form of activities, and improve the quality and appeal of activities, thereby better attracting members to participate, enhancing the influence and cohesion of the group, expanding the role of social groups in society, meeting the growing demand, and having a very broad application prospect.

[0003] The success or failure of social group activities is often closely related to the scientificity of decision-making. Proper decision-making can ensure the rational allocation of activity resources. At the same time, scientific decision-making helps social groups accurately grasp the needs and expectations of members and other stakeholders, improves the relevance and effectiveness of activities, and thus promotes the long-term stable development of social groups and enhances their competitiveness and irreplaceability in society, so decision support analysis methods are the key tool for social groups to grow continuously in a complex and changing social environment.

[0004] Many current decision support analysis methods for social group activities only focus on data analysis in one aspect, such as only focusing on the number of activity participants or activity costs, etc. single indicator, ignoring the comprehensive consideration of multi-dimensional factors such as activity popularity, participation depth, behavior conversion, and social network retention. This one-sided data analysis leads to an incomplete basis for decision-making, making it difficult to accurately assess the overall effectiveness of the activity, thereby affecting the scientificity and effectiveness of the decision. SUMMARY

[0005] (I) Technical problems solved

[0006] In view of the deficiencies of the prior art, the present application provides a multi-dimensional data analysis method for social group decision support, which solves the technical problems of incomplete basis for decision-making and poor assessment accuracy caused by the prior art only focusing on the number of activity participants or activity costs, etc. single indicator, ignoring multi-dimensional factors such as participation depth, behavior conversion, and social network retention.

[0007] (II) Technical solutions

[0008] To achieve the above objectives, the present invention provides the following technical solution: a multi-dimensional data analysis method for decision support in social groups, comprising:

[0009] Step 1: Obtain basic activity data and analyze it to obtain an estimated activity popularity value. ;

[0010] Step 2: Obtain participation depth data, and obtain the participation depth index by analyzing the participation depth data and basic activity data. ;

[0011] Step 3: Obtain behavioral conversion data through the HR system, and analyze the behavioral conversion data and basic activity data to obtain the conversion funnel effectiveness value. ;

[0012] Step 4: Obtain retention network data through social network systems, analyze the retention network data, and calculate the retention network influence value. ;

[0013] Step 5: Estimating the activity's popularity Participation Depth Index Conversion funnel efficiency value and retention of online influence value A comprehensive analysis was conducted to obtain the decision support level. ;

[0014] Step Six: Set the decision support threshold set, and include decision support scores. The decision is compared with a threshold in the decision support threshold set, and the result of the comparison determines whether support should be given.

[0015] In the preferred embodiment of the multi-dimensional data analysis method for social group decision support described above, the basic activity data includes the actual number of participants. Interaction frequency value and weather influence coefficient The specific method for obtaining it is as follows:

[0016] The attendance system was used to count the actual number of participants in similar club activities. ;

[0017] By counting the number of online comments for community activities per unit of time The number of questions asked on-site was identified and counted using a speech recognition device. Based on the number of comments Number of questions asked on site Calculate the interaction frequency value The formula used is: ;

[0018] Meteorological data is acquired through a meteorological platform and input into a dynamic compensation model to obtain the weather impact coefficient. The calculation formula used is as follows: ;in, is the environmental basis factor for environment type b; b is the ordinal number of the environment type, and its value is a positive integer. This represents the temperature decay coefficient.

[0019] In the preferred embodiment of the multi-dimensional data analysis method for social group decision support described above, the activity popularity estimate is calculated using basic activity data. The formula used is as follows:

[0020] ;

[0021] in, The actual number of participants in the i-th club activity; Let i be the number of months since the i-th club activity; Let i be the interaction frequency value of the i-th club activity; Let be the weather impact coefficient for the i-th club activity; i represents the sequence number of the historical club activity, which takes a positive integer value, and n is the maximum number of club activities.

[0022] In the preferred embodiment of the multi-dimensional data analysis method for social group decision support described above, the participation depth data includes the number of people at different interaction levels k. The specific method for obtaining it is as follows:

[0023] The number of people at different interaction levels (k) is retrieved through the database management system. ;

[0024] By analyzing the number of people at different interaction levels k and actual number of participants Analyze and calculate the participation depth index. The formula used is as follows:

[0025] ;

[0026] in, This indicates that the i-th club activity has reached the interaction level k, where k is the sequence number of the interaction level. This represents the hierarchical weight coefficient.

[0027] In the preferred embodiment of the multi-dimensional data analysis method for social group decision support described above, the transformed data includes the number of people joining the community within a unit time period. and the number of people promoted ;

[0028] The number of people who join the community in a unit time period after the community activity is over is counted through the personnel system The number of people who join the community in a unit time period is counted The actual number of participants is counted The short-term conversion rate is calculated The formula is as follows: ;

[0029] The number of people who obtain a promotion in a unit time period after the community activity is over is counted through the personnel system The number of people who obtain a promotion in a unit time period is counted The number of people who join the community in a unit time period is counted The long-term conversion rate is calculated The formula is as follows: .

[0030] In the preferred solution of the multi-dimensional data analysis method for social group decision support, the conversion funnel efficiency value is calculated according to the conversion data and the basic activity data The formula is as follows:

[0031] ;

[0032] Wherein, represents the conversion parameter of the current conversion stage of the i-th community activity; represents the conversion parameter of the previous conversion stage of the i-th community activity, and j represents the serial number of different conversion stages;

[0033] When j is 1, is , the conversion parameter of the previous conversion stage is the total number of participants , is , the conversion parameter of the current conversion stage is the short-term conversion rate ; when j is 2, is , the conversion parameter of the previous conversion stage is the short-term conversion rate ; is , the conversion parameter of the current conversion stage is the long-term conversion rate .

[0034] In the preferred solution of the multi-dimensional data analysis method for social group decision support, the social network degree , the activity center degree , the community contribution degree , the average distance of direct interaction , and the active time Calculating retention network influence value The formula is as follows:

[0035] ;

[0036] Wherein, is the social network degree of the Pth community member, is the activity centrality of the Pth community member; is the community contribution degree of the Pth community member, is the average distance of direct interaction of the Pth community member, is the active time of the Pth community member.

[0037] In the preferred scheme of the above-mentioned multi-dimensional data analysis method for social group decision support, the activity heat estimation value , the participation depth index , the conversion funnel efficiency value and the retention network influence value are comprehensively analyzed to obtain the decision support degree , and the calculation formula is as follows:

[0038] ,

[0039] Wherein, is the weight coefficient of , and is the average activity cost of the community activity.

[0040] In the preferred scheme of the above-mentioned multi-dimensional data analysis method for social group decision support, the decision support threshold set includes a decision support threshold and a decision optimization threshold; the decision support degree is compared with the decision support threshold and the decision optimization threshold respectively, when the decision support degree is greater than or equal to the decision support threshold, the corresponding community activity suggestion is supported by policy; when the decision support threshold is greater than the decision support degree is greater than or equal to the decision optimization threshold, it is suggested that the scheme of the community activity is adjusted. When the decision support degree is less than the decision optimization threshold, the corresponding community activity is not suggested to be supported by policy.

[0041] (Three) beneficial effects

[0042] The application provides a multi-dimensional data analysis method for social group decision support, which has the following beneficial effects:

[0043] (1) By collecting basic activity data and using scientific data analysis algorithms for processing, it can estimate the possible level of activity heat in advance. This allows social groups to have a clear assessment of the potential impact of the activity at the planning stage, as one of the factors to assess the feasibility of the activity.

[0044] (2) The participation depth index obtained by cross analysis of these data and basic activity data can accurately reflect the real input of members to the activity. This helps social groups to identify which activity links can attract members to participate deeply, and then to optimize the content and form of the activity, improve the participation experience and satisfaction of members, enhance the cohesion and member stickiness of the group, and promote the long-term stable development of social groups.

[0045] (3) The conversion funnel efficiency value is obtained by using the personnel system to obtain behavior conversion data and analysis. This provides a direct quantitative index for the conversion of activity effect for social groups. Through the comparison and analysis of the conversion data at each link with the basic activity data, the conversion funnel efficiency value is formed. This allows social groups to clearly understand the conversion efficiency of each link of the activity, and to find the key links that may lead to potential member loss. Accordingly, the social group can optimize the activity process accurately, reduce the loss of potential members, improve the overall conversion efficiency of the activity, and ensure that the social group activity can effectively attract and retain members.

[0046] (4) By obtaining retention network data from social network systems and calculating retention network influence value, it broadens the evaluation perspective of the subsequent influence of social groups. In the social media era, the influence of social group activities is not limited to the activity itself, but also has a lasting radiation effect through members' sharing and discussion on social networks. By analyzing the retention network data and calculating the retention network influence value, social groups can fully understand the long-term influence of the activity in the social network, provide reference for subsequent activity planning of the group, such as developing more communicable activity programs based on the characteristics of activity types and content with high retention network influence, further expanding the social influence of the group, attracting more potential members' attention and joining, improving the group's reputation and popularity in the society, and enhancing its social value and competitiveness.

[0047] (5) Through the steps of multi-dimensional data analysis method, it provides comprehensive, in-depth and accurate quantitative evaluation basis for decision support of social groups, effectively solves the problems existing in the current decision support analysis method of social group activities, has significant innovation and practicality, and can effectively promote the efficient operation and sustainable development of social groups. BRIEF DESCRIPTION OF DRAWINGS

[0048] Figure 1 A schematic diagram of steps of a multi-dimensional data analysis method for social group decision support of the present application. DETAILED DESCRIPTION

[0049] The technical solutions in the embodiments of the present application will be clearly and completely described with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0050] Embodiment 1

[0051] Please refer to Figure 1 , the present application provides a multi-dimensional data analysis method for social group decision support, comprising:

[0052] Step 1: Obtain basic activity data, and obtain activity heat estimation value by analyzing the basic activity data .

[0053] It should be noted that in the process of calculating the activity heat estimation value , all the parameters involved in each calculation step need to be normalized and pre-processed to eliminate the dimensions of different parameters, so as to facilitate subsequent formula calculation.

[0054] The basic activity data includes the actual number of participants , the interaction frequency value and the weather influence coefficient .

[0055] Step 101: Count the actual number of participants in the same type of community activity through the check-in system ;

[0056] Step 102: Count the number of comments online in unit time of community activity ; for example, the number of comments of activity exclusive topic API can be counted, such as WeChat public number, microblog, etc. The number of on-site questions is identified and counted by setting a language recognition device on site, such as Keda Xunfei MICP-411 device, etc. ; the interaction frequency value is calculated according to the number of comments and the number of on-site questions , and the formula is: ;

[0057] Step 103: Obtain meteorological data through a meteorological platform, and input the meteorological data into a dynamic compensation model to obtain the weather influence coefficient , and the calculation formula is: ; wherein, is an environmental base factor of the environmental type b; can be set according to industry standards; b is the serial number of the environmental type, and is a positive integer; for example, is an environmental base factor representing a sunny state, and can be 1, is an environmental base factor representing a cloudy state, and can be 0.85, is an environmental base factor representing a light rain state, and can be 0.65, is an environmental base factor representing a heavy rain state, and can be 0.35; is a temperature attenuation coefficient, which can be set according to industry standards, and can be 0.015, is a temperature deviation, and the calculation formula is: , is an actual temperature, is an ideal temperature, which can be defined according to international standards and national standards, for example, 22℃; e is the base number of natural logarithm, and R is the precipitation intensity value.

[0058] The development of social group activities is often significantly affected by weather conditions. For example, outdoor activities may face problems such as insufficient number of participants, poor activity experience, and even safety risks in bad weather. However, traditional decision support methods lack quantitative evaluation of meteorological factors, making it difficult to accurately determine the actual impact of weather on activities, resulting in potential blindness in decision-making.

[0059] By calculating the temperature deviation , the actual temperature is compared with the ideal temperature (such as 22℃), and the degree of temperature deviation from the ideal state is quantified. In the formula, the term is introduced, where is a temperature attenuation coefficient, which can be 0.015, representing the attenuation effect of temperature deviation on activities. When the actual temperature is higher or lower than the ideal temperature, the value of increases or decreases accordingly, causing the value of the entire expression to change. If the actual temperature is too high or too low, the value of will decrease, thereby reducing the final value of , reflecting the impact of adverse temperature conditions on activities; conversely, if the temperature is close to the ideal value,

[0060] the value is relatively high. This quantitative treatment of temperature deviation enables the decision support method to more comprehensively evaluate the impact of meteorological conditions on activities, prompting decision-makers to fully consider temperature factors and reasonably arrange activity times or take temperature adjustment measures to improve the participation and experience of activities. , the formula is as follows:

[0061] ;

[0062] Wherein, is the actual number of participants in the i-th community activity; is the number of months from the i-th community activity; is the interaction frequency value of the i-th community activity; is the weather influence coefficient of the i-th community activity; i represents the serial number of historical community activities, which is a positive integer, and n is the maximum number of community activities.

[0063] It should be noted that the time decay factor is introduced to weight the actual number of participants in the historical activities. With the increase of the distance between the activity time and the current time, the weight of this item decays exponentially. This reflects the natural decline of activity popularity over time, making the model pay more attention to the participation of recent activities, so as to more accurately reflect the current trend of community activity popularity, help community managers focus on recent activities that members pay more attention to, and adjust the activity strategy in time to maintain the activity of the community. In the formula, the average value of interaction frequency is introduced in the form of logarithmic function to modify and improve the activity popularity. The use of logarithmic function not only ensures the positive promoting effect of interaction frequency on popularity, but also avoids the excessive inflation of popularity value caused by the too high interaction frequency of individual activities, so that the influence of interaction frequency on popularity is more smooth and reasonable. This processing method fully reflects the promoting effect of members' initiative participation and communication in activities on activity popularity, prompting community managers to pay attention to how to improve the interaction of activities and enhance the connection and cohesion among members, and further improve the attraction and influence of activities. The average value of weather influence coefficient is included in the calculation of activity popularity prediction value, which quantifies the comprehensive influence of weather on community activity popularity. In adverse weather conditions, the value is low, which will correspondingly lower the overall value, reminding community managers to carefully arrange activities or take measures; while in favorable weather, it will improve the value, encouraging to actively organize outdoor activities. This quantitative consideration of weather factors makes the activity popularity evaluation more close to the actual situation, helping the community to better plan the activity time and improve the success rate and member satisfaction of the activity.

[0064] Step two: obtain participation depth data, and obtain participation depth index by analyzing the participation depth data and basic activity data.

[0065] It should be noted that in the calculation of participation depth index In the process of calculation, all the parameters involved in each calculation step need to be normalized and pre-processed respectively to eliminate the dimensions of different parameters, so as to facilitate subsequent formula calculation.

[0066] The number of people participating in deep data at different interaction levels k .

[0067] Step 201: The number of people participating in deep data at different interaction levels k is read by the analysis tool of the log system (such as Splunk, ELK Stack) or the database management system (such as MySQL, MongoDB) through SQL query or API call ;

[0068] Step 202: Calculate the participation depth index by analyzing the number of people participating at different interaction levels k and the actual number of participants , according to the following formula:

[0069] ;

[0070] Where, represents the number of people participating in the i-th community activity at the interaction level k, k is the serial number of the interaction level, and takes the value of 1-4; is the level weight coefficient, which can take the value of 1.5.

[0071] It should be noted that the number of people at different interaction levels k can be set as the number of people at the watching level, the number of people at the questioning level, the number of people assisting in organization, and the number of people initiating sub-activities, etc.

[0072] It should be noted that the double summation operation integrates all activity and interaction level data, eliminates data silos, and constructs a comprehensive participation depth evaluation system. It systematically analyzes the participation of community historical activities, ensures that the data of each activity and each interaction level are considered, and makes the overall participation depth trend of the community accurately reflect, providing strong support for macro decision-making. In the formula, the number of people at each level is counted, the higher value of high-level interaction highlights the higher value of high-level interaction, and this level weighting processing amplifies the impact of deep interaction, so that the changes in member participation depth can be captured more sensitively. The community can accurately locate the activities that members like deeply, provide a basis for designing high-interactive activities, improve the quality and attractiveness of activities, and stimulate member enthusiasm. The sum of the number of people at each interaction level is divided by the actual number of participants ​This process calculates the percentage of deep engagement. It correlates interaction level data with overall engagement scale, visually presenting the proportion of deep participants and reflecting the activity's broad and deep appeal to members. A high percentage of deep participants signifies high interactivity and attractiveness, providing crucial information for community organizations to assess the success and development potential of activities, and helping them create more impactful branded events.

[0073] Step 3: Obtain behavioral conversion data through the HR system, and analyze the behavioral conversion data and basic activity data to obtain the conversion funnel effectiveness value. .

[0074] It should be noted that when calculating the efficiency value of the conversion funnel... During the process, all parameters involved in each calculation step need to be normalized and preprocessed to eliminate the dimensions of different parameters, so as to facilitate subsequent formula calculations.

[0075] Conversion data includes the number of people joining the community within a unit of time period. and the number of people promoted ;

[0076] Step 301: Use the personnel system to count the number of people who joined the club within a unit time period after the club activity ended. Based on the number of people joining the club within a unit time period and actual number of participants Calculate short-term conversion rate The formula used is: .

[0077] It should be noted that the statistics only include the number of people who have joined the club. The unit time period can be set to one week or ten days, and you can refer to the experience of similar club activities for setting it.

[0078] Step 302: Use the HR system to count the number of people who received promotions within a given time period after the club activities ended. Based on the number of people who receive a promotion within a given time period. The number of people joining the club within a unit of time period Calculate long-term conversion rate The formula used is .

[0079] It should be noted that the statistics include the number of people who were promoted. The unit time period can be set to 90 days, which can be referenced from the experience of similar club activities.

[0080] Step 303: Based on short-term conversion rate and long-term conversion rate Calculate the conversion funnel performance value , the formula is as follows:

[0081] ;

[0082] Wherein, represents the conversion parameter of the current conversion stage of the i-th community activity; represents the conversion parameter of the previous conversion stage of the i-th community activity, and j represents the serial number of different conversion stages, taking values of 1-2.

[0083] It should be noted that when j takes 1, is , the conversion parameter of the previous conversion stage is the total number of participants , is , the conversion parameter of the current conversion stage is the short-term conversion rate ; when j takes 2, is , the conversion parameter of the previous conversion stage is the short-term conversion rate ; is , the conversion parameter of the current conversion stage is the long-term conversion rate .

[0084] It should be noted that by comprehensively calculating the short-term and long-term conversion rates , the overall performance of the community activity in the entire conversion cycle is quantified. It clearly presents the loss and retention of members in different conversion stages, helping community managers to locate the key problems in the conversion process, such as whether the participation link in the early stage of the activity can effectively attract members, and whether the subsequent deep conversion link can retain members. Based on this, the community can optimize the activity planning and execution, improve the efficiency of each conversion stage, and thus improve the overall member conversion rate, enhance the attractiveness and vitality of the community. In the formula, the product term is calculated by sequentially calculating the ratio of adjacent conversion stages, and the product of 1-the ratio is calculated, which quantifies the degree of member loss in the conversion process. When the ratio of a certain conversion stage is low, the product result will be larger, which means that there is a problem in the conversion continuity of the activity, and members are prone to lose at this stage. This calculation logic sensitively captures the weak link in the conversion process, prompting the community to focus on the smoothness of the conversion process, optimize the connection between stages, reduce member loss, and improve overall conversion performance.

[0085] Step four: obtain retention network data through a social network system, analyze the retention network data, and calculate the retention network influence value .

[0086] It should be noted that in the process of calculating the retention network influence value , all the parameters involved in each calculation step need to be normalized and pre-processed respectively to eliminate the dimensions of different parameters, so as to facilitate subsequent formula calculation.

[0087] Step 401: Use the API interface provided by the social network platform (such as WeChat, Weibo, Facebook, etc.) to access the friend list or follow list of the user through user authorization, and count the sum of the number of friends or the number of followers and fans of each community user as the social network degree .

[0088] Step 402: Collect the behavior data of community members in the social network about community activities, including the number of posts, comments, likes, and shares. Calculate the activity centrality based on the behavior data, and the formula is as follows:

[0089] ;

[0090] Where, is the activity centrality of the Pth community member; is the number of posts of the Pth community member, is the number of comments of the Pth community member, is the number of likes of the Pth community member, is the number of shares of the Pth community member; P is the serial number of the community member, taking positive integer value, and Q is the number of community members; is the weight coefficient of the number of posts; is the weight coefficient of the number of comments, is the weight coefficient of the number of likes, is the weight coefficient of the number of comments, and , each can take the value of 0.25.

[0091] It should be noted that the numerator part of the formula gives different weight coefficients (a, b, c, d) to the post, comment, like and share behaviors of the community member , , , , and performs weighted summation. This weighted processing fully considers the different influence degrees of different behaviors on activity. In this way, the comprehensive activity of the community member can be reasonably quantified, and the evaluation result is more in line with the actual situation. The denominator part takes the maximum value of the weighted summation result of all community members, which is used for normalization of the numerator. The normalized activity centrality The interval is [0, 1], which eliminates the influence of different member behavior data dimensions, making the activity of different members comparable. This helps the community manager quickly identify the most active members and understand the core strength of the community, while also providing a basis for the community's internal incentive mechanism to encourage members to actively participate in social network interactions.

[0092] Step 403: Collect the activity records of community members in historical community activities, including the number of times participating in community activities and the amount of financial support provided, and calculate the community contribution degree through the number of times participating in community activities and the amount of financial support , The formula is as follows:

[0093] ,

[0094] Among them, is the number of times the Pth community member participates in community activities; is the financial support amount of the Pth community member; is the weight coefficient of the number of times participating in community activities, is the weight coefficient of the amount of financial support. And , Each item can be valued at 0.5.

[0095] It should be noted that the numerator part of the formula gives different weight coefficients ( and ) to the number of times participating and the amount of financial support of community members, and performs weighted summation. This weighted processing fully considers the different importance of the number of times participating and the amount of financial support to the community contribution. For example, the higher the number of times participating reflects the time and effort of the member, and the financial support directly reflects the member's contribution to the community's material resources. In this way, the comprehensive contribution degree of community members can be reasonably quantified, ensuring that the evaluation results are more in line with the actual situation, balancing the contributions of members in different aspects to the community. The denominator takes the maximum value of the weighted summation results of all community members, which is used for normalization processing of the numerator. The normalized community contribution degree will be in the interval [0, 1], eliminating the influence of different member data dimensions, making the contribution degree of different members comparable. This helps the community manager quickly identify the members who contribute most to the community and understand the core strength of the community, while also providing a basis for resource allocation and incentive measures within the community, encouraging members to actively participate in community activities and provide financial support.

[0096] Step 404: Use social network analysis tools (such as Gephi, UCINET, etc.) to analyze user interaction networks. Import the user interaction data of community members into the tool, build a network graph, and through the tool's function of calculating the average distance of the network, obtain the direct interaction average distance between community members within the community ;

[0097] Step 405: Collect the activity records of community members in the community, including the date of participating in activities, the login date on the community platform, etc., find the date of the first time the user participates in community activities and the date of the last time the user participates in community activities, calculate the number of days between the two dates as the active time of the community member in the community .

[0098] Step 405: Calculate the retention network influence value according to the social network degree , activity centrality , community contribution , direct interaction average distance and active time , the formula is as follows:

[0099] ;

[0100] Where, is the social network degree of the Pth community member, is the activity centrality of the Pth community member; is the community contribution of the Pth community member, is the direct interaction average distance of the Pth community member, is the active time of the Pth community member.

[0101] It should be noted that by collecting and integrating the social network degree, activity centrality, community contribution, direct interaction average distance and active time of community members, etc. Multi-dimensional data can comprehensively quantify the overall influence of community members in the network. This not only reflects the breadth and activity of members' social connections, but also reflects the members' contribution to the community and the efficiency and persistence of their participation. Based on this, the community can better understand the overall influence of each member in the network, identify key opinion leaders and core members, provide scientific basis for the community's network promotion, member management and development strategy, and enhance the network cohesion and influence of the community. The activity centrality and social network degree are combined, and the hyperbolic tangent function is used for nonlinear transformation of . Activity centrality reflects the activity level of members in the network, and social network degree reflects the breadth of members' social connections. The hyperbolic tangent function ​The use of the social network degree of influence has a certain saturation, that is, when the social connection of the member reaches a certain number, the growth effect of the influence will gradually weaken. This combination considers the activity and social connection breadth of the member, while avoiding the over-reliance on the number of social connections to cause the influence to be overestimated, and can more accurately reflect the actual influence of the member in the network. The community contribution degree The square root of the average distance of direct interaction The square root of the average distance of direct interaction. The community contribution degree reflects the comprehensive contribution of the member to the community, and the average distance of direct interaction reflects the close degree of interaction between the member and other members. The square root processing of the average distance of direct interaction makes the influence of the interaction distance gradually weaken, that is, the closer the interaction distance between members, the greater the role in promoting the influence. This combination considers the contribution of the member to the community and the interaction efficiency between members, and comprehensively evaluates the value and actual interaction effect of the member in the community, which helps to identify core members who have important contributions to the community and maintain close interaction with other members.

[0102] Step five: through the normalization preprocessing of the activity heat estimation value , the participation depth index , the conversion funnel efficiency value and the retention network influence value , then comprehensive analysis is carried out to obtain the decision support degree , the calculation formula is as follows:

[0103] ,

[0104] Among them, is the weight coefficient of , and is the average activity cost of the community activity.

[0105] It should be noted that through the comprehensive , , and four-dimensional indicators, the performance of the community activity can be evaluated comprehensively. reflects the expected heat of the activity, embodies the depth of member participation, measures the conversion efficiency of the activity, and shows the influence of the activity in the network. This comprehensive analysis method fully considers the multiple key aspects of the community activity, provides a more comprehensive and objective basis for decision-making, and helps the community to more accurately judge the value and potential of the activity, so as to make more reasonable decisions. embodies the depth of member participation, reflects the influence of the activity in the network, represents the average cost of community activities. The and are multiplied, and then divided by , taking into account the depth of participation, network influence and cost-effectiveness of the activities. Multiply the weight coefficient , further emphasizing the importance of this comprehensive index in decision support. This processing method can more comprehensively evaluate the comprehensive benefits of activities in terms of participation depth and network influence, while considering the cost factor, helping the community to choose more valuable activities under limited resources. measures the effectiveness of community activities in converting members at different stages, reflecting the conversion effect from potential members to actual participants to deeply involved members. form ensures that occupies a proper weight in the comprehensive analysis, enabling it to balance with part. This makes the decision support comprehensively reflect the conversion effectiveness of activities, helping the community to identify activities that perform well in the conversion process, thereby optimizing activity strategies and improving member conversion rates.

[0106] Step six: set a decision support threshold set, compare the decision support degree with the thresholds in the decision support threshold set, and judge whether it should be supported according to the comparison result.

[0107] Step 601: set a decision support threshold set, which includes a decision support threshold and a decision optimization threshold.

[0108] It should be noted that the median of VCC of Top20% activities is obtained through industry databases such as "Community Activity Effectiveness White Paper", which is set as Vbench, the decision support threshold can be set as 1.2*Vbench, and the decision optimization threshold can be set as 0.7*Vbench. The specific value can be adjusted according to the actual situation.

[0109] Step 602: compare the decision support degree with the decision support threshold and the decision optimization threshold respectively, when the decision support degree ≥ decision support threshold, the corresponding community activity suggestion is supported by policy; when the decision support threshold > decision support degree ≥ decision optimization threshold, then the scheme of community activities is recommended to be adjusted. When the decision support degree < decision optimization threshold, the corresponding community activity is not recommended to be supported by policy.

[0110] Example 2:

[0111] The application discloses a multi-dimensional data analysis system for social group decision support, which is used for realizing the multi-dimensional data analysis method for social group decision support.

[0112] When implemented by using software, the above-mentioned embodiments can be implemented in the form of a computer program product in whole or in part. Those skilled in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed in the present text can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are executed in hardware or software mode depends on the specific application and design constraints of the technical solution.

[0113] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, can be located in one place, or can be distributed on multiple network units. Part or all of the units can be selected to achieve the purpose of the embodiment scheme according to actual needs.

[0114] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed in the present application, which should be covered within the protection scope of the present application.

Claims

1. A multi-dimensional data analysis method for social group decision support, characterized by, Comprising: Step one: Obtain basic activity data, and obtain activity heat estimation value by analyzing the basic activity data ; Step two: Obtain the participation depth data, and obtain the participation depth index by analyzing the participation depth data and the basic activity data ; Step 3: Obtain behavioral conversion data through the HR system, and analyze the behavioral conversion data and basic activity data to obtain the conversion funnel effectiveness value. The conversion data includes the number of people joining the community within a given time period. and the number of people promoted ; The number of people who joined the community in a unit time period after the community activity is over is counted by the personnel system The short-term conversion rate is calculated according to the number of people who joined the community in a unit time period and the actual number of participants The formula is as follows: ; ; The maximum number of community activities The number of people who get promoted in a unit time period after the end of the club activities is counted by the personnel system The long-term conversion rate is calculated according to the number of people who get promoted in a unit time period and the number of people who join the club in a unit time period The long-term conversion rate is calculated according to the number of people who get promoted in a unit time period The formula is ; Calculating conversion funnel performance values from conversion data and base activity data According to the following formula: ； wherein, represents a conversion parameter of a current conversion stage of the represents a conversion parameter of a previous conversion stage of the represents a sequence number of different conversion stages;​​ When is 1, is the conversion parameter of the previous conversion stage is the total number of participants , is the conversion parameter of the current conversion stage is the short-term conversion rate ; when is 2, is the conversion parameter of the previous conversion stage is the short-term conversion rate ; is the conversion parameter of the current conversion stage is the long-term conversion rate ; Step four: obtaining the retention network data through the social network system, and analyzing the retention network data to calculate the retention network influence value ; Specifically: according to the social network degree , the activity center degree , the community contribution degree , the direct interaction average distance and the active time , the retention network influence value is calculated , and the formula is as follows: ； wherein, is the social network degree of the th community member, is the activity centrality of the th community member; is the community contribution of the th community member, is the average distance of direct interaction of the th community member, is the active time of the th community member; is the number of community members; Step 5: Estimating the activity's popularity Participation Depth Index Conversion funnel efficiency value and retention of online influence value A comprehensive analysis was conducted to obtain the decision support level. ; Step six: setting a decision support threshold set, comparing the decision support degree with the threshold in the decision support threshold set, and judging whether support should be obtained according to the comparison result. with the threshold in the decision support threshold set, and judging whether support should be obtained according to the comparison result. The basic activity data includes actual number of participants , interaction frequency value and weather influence coefficient ; the specific acquisition method is: Count the actual number of participants in the same type of community activities through the sign-in system ; By counting the number of online comments for community activities per unit of time The number of questions asked on-site was identified and counted using a speech recognition device. Based on the number of comments Number of questions asked on site Calculate the interaction frequency value The formula used is: The weather data is obtained through a weather platform, and the weather data is input into a dynamic compensation model to obtain a weather influence coefficient , and the calculation formula is: ; wherein, is an environmental base factor of an environmental type ; is a serial number of the environmental type, and is a positive integer; represents a temperature attenuation coefficient; is a temperature deviation; is a precipitation intensity value.

2. The multi-dimensional data analysis method for social group decision support according to claim 1, wherein, Calculating activity heat estimation value through basic activity data The formula is as follows: ； wherein, is the number of actual participants of the nth community activity; is the number of actual participants of the nth community activity; is the number of months from now to the nth community activity; is the number of months from now to the nth community activity; is the interaction frequency value of the nth community activity; is the interaction frequency value of the nth community activity; is the weather influence coefficient of the nth community activity; is the weather influence coefficient of the nth community activity; represents the serial number of the historical community activity, and takes a positive integer, is the maximum number of community activities.

3. The multi-dimensional data analysis method for social group decision support according to claim 2, wherein, Engagement depth data includes different levels of interaction number of people The specific acquisition method is: Reading different interaction levels by a database management system Number of people ; By analyzing the number of different interaction levels and the actual number of participants the engagement depth index is calculated, based on the following formula:​ ； in, Indicates the first The club activities reached an interactive level. The number of people; This refers to the sequence number of the interaction level; This represents the hierarchical weight coefficient.

4. The multi-dimensional data analysis method for social group decision support according to claim 3, wherein, The decision support threshold set includes decision support thresholds and decision optimization thresholds; it considers decision support levels... The decision support score is compared with the decision support threshold and the decision optimization threshold, respectively. When the decision support threshold is ≥, the corresponding community activity is recommended for policy support; when the decision support threshold is >, the decision support level is recommended. If the decision support level is ≥ the decision optimization threshold, it is recommended to adjust the club activity plan; when the decision support level is ≥ the decision optimization threshold, it is recommended to adjust the club activity plan. If the decision optimization threshold is reached, then policy support is not recommended for the corresponding community activities.

5. A multi-dimensional data analysis system for social group decision support, characterized by, A multidimensional data analysis method for social group decision support implementing any of the above claims 1-4.

Images