CN117151914A - Crowd sensing user selection method and device based on comprehensive influence evaluation - Google Patents

Crowd sensing user selection method and device based on comprehensive influence evaluation Download PDF

Info

Publication number
CN117151914A
CN117151914A CN202311436096.1A CN202311436096A CN117151914A CN 117151914 A CN117151914 A CN 117151914A CN 202311436096 A CN202311436096 A CN 202311436096A CN 117151914 A CN117151914 A CN 117151914A
Authority
CN
China
Prior art keywords
user
score
users
seed
data
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202311436096.1A
Other languages
Chinese (zh)
Other versions
CN117151914B (en
Inventor
陈彬
周子涛
徐矿姗
朱正秋
赵勇
季雅秦
郭润康
杨芳
代云凯
颜梦宇
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
National University of Defense Technology
Original Assignee
National University of Defense Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by National University of Defense Technology filed Critical National University of Defense Technology
Priority to CN202311436096.1A priority Critical patent/CN117151914B/en
Publication of CN117151914A publication Critical patent/CN117151914A/en
Application granted granted Critical
Publication of CN117151914B publication Critical patent/CN117151914B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/01—Social networking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Strategic Management (AREA)
  • Data Mining & Analysis (AREA)
  • Economics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Physics (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Pure & Applied Mathematics (AREA)
  • Development Economics (AREA)
  • Educational Administration (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Tourism & Hospitality (AREA)
  • Operations Research (AREA)
  • Computational Mathematics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Computing Systems (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Quality & Reliability (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Game Theory and Decision Science (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Algebra (AREA)
  • Primary Health Care (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The application relates to a crowd sensing user selection method and device based on comprehensive influence evaluation. The method comprises the following steps: the method comprises the steps of obtaining a social network user data set in a set time-out area, constructing a social network model, calculating according to track data of users and social information data among the users in the social network model, obtaining comprehensive influence scores of all the users, adding the users with the highest comprehensive influence scores into a preselection seed user set, after the preselection seed user set reaches a quantity limit, carrying out simulation propagation of the users, calculating space-time coverage rate of the propagated preselection seed user set, and selecting the preselection seed user with the largest space-time coverage rate in the preselection seed user set as a final seed user under the budget limit. By adopting the method, seed user selection can be performed according to the comprehensive influence score and the space-time coverage rate, and the efficient propagation of group control perception tasks can be realized on the basis of acquiring a smaller number of seed users.

Description

Crowd sensing user selection method and device based on comprehensive influence evaluation
Technical Field
The application relates to the technical field of urban environment object crowd sensing, in particular to a crowd sensing user selection method and device based on comprehensive influence evaluation.
Background
Currently, data awareness for urban environments mainly includes two methods. The first, more traditional, method relies on manual unified execution. Each city requires a certain knowledge of the number, distribution and use of the various infrastructures in the city, and therefore specialized staff are usually assigned to conduct on-site surveys of the infrastructures. This method requires a lot of manpower and material resources, and the cost in each aspect is high. Another method is based on internet of things, deploying a fixed awareness network for a specific area. However, the arrangement of the internet of things is often high in cost, difficult in system maintenance, inflexible in service and low in cost performance. The problems bring great difficulty to the large-scale and large-scale practical application of the Internet of things, and the development and popularization of the Internet of things technology are hindered.
In recent years, due to demands of internet of things applications, rapid popularization of mobile intelligent terminals and the emergence of crowd-sourced computing modes, a mobile internet of things sensing mode based on sensing capability of mobile equipment (smart phones, wearable equipment, vehicle-mounted terminals and the like) is developed, which is called crowd sensing. The method realizes sensing task distribution and sensing data collection and utilization by combining conscious participation and unconscious cooperation by using a large number of common users as basic sensing units by using mobile sensing equipment, and finally completes large-scale and complex social sensing tasks.
Compared with the traditional sensing mode of the Internet of things of the fixed sensing network, the group participation and group intelligence endow the mode with three advantages: (1) The crowd sensing utilizes the existing sensor and communication infrastructure, so the deployment cost is much lower than that of a wireless sensor network; (2) The inherent mobility of mobile users provides an unprecedented space-time coverage effect on tasks; (3) In the crowd sensing practical application process, human task execution and data preprocessing are integrated with human understanding and intelligence of the task.
The current crowd sensing has been developed to some extent, and has been applied to life. In the civil aspect, the method has the fields of environment monitoring, public safety, smart city, travel, social service, health monitoring, indoor positioning and the like.
In current crowd-sourced research, it is mostly assumed that there are enough users for the platform to select. However, under current technical and social conditions, insufficient users may be recruited. First, while the system may provide rewards to encourage users to complete tasks, some users are reluctant to participate due to lack of interest or privacy awareness. Second, there is often a bias in the distribution of users. Spatially, users tend to concentrate on business and living areas of a city, while some development and suburban areas tend to have few or no users; the number of users within a sub-area varies over time with changes in daily activity. For example, there are more daytime users and fewer evening users may be. The biased distribution of the user results in some sub-regions lacking sensory data, which in turn results in a reduction in overall perceived accuracy.
Disclosure of Invention
Based on the above, it is necessary to provide a method and a device for selecting a crowd sensing user based on comprehensive influence evaluation, which can select seed users by sensing the propagation of tasks on a social network without a large amount of user basis for the above technical problem of insufficient initial user number of the crowd sensing platform, so as to realize efficient completion of the sensing tasks.
A crowd-sourced user selection method based on comprehensive impact assessment, the method comprising:
acquiring a social network user data set in a set time-out area, and constructing a social network model in an independent cascading model form according to the social network user data set; the social network user data set comprises social information data and historical track data of a user;
calculating according to historical track data of a user in the social network model, obtaining track scores of the user, calculating according to social information data of different users in the social network model, obtaining activation probability among the users, and calculating according to the track scores and the activation probability, obtaining comprehensive influence scores of the users;
selecting the user with the highest comprehensive influence score to add into the preselected seed user set, updating the comprehensive influence score of the user according to the historical track data of the preselected seed user set to obtain an updated comprehensive influence score, and selecting the user with the highest updated comprehensive influence score to add into the preselected seed user set until the number limit of the preselected seed user set is reached;
and carrying out crowd sensing task simulation propagation on the preselection seed user set, calculating the space-time coverage rate of the propagated preselection seed user set in a set space-time area, selecting the preselection seed user with the largest space-time coverage rate in the preselection seed user set as a final seed user under the condition of setting budget limit, and constructing to obtain the final seed user set.
In one embodiment, the calculating according to the historical track data of the user in the social network model to obtain the track score of the user includes:
calculating according to historical track data of the user in the social network model, obtaining track scores of the user, and representing the track scores as
；
Wherein,representing user +.>Track score of>Representing user +.>Time-space coverage generated in a set time-space region of historical track data of (a),>the historical track data of the user is represented,tthe sign-on time is indicated as such,sindicating that the sign-on space is available,prepresenting the sign-on probability.
In one embodiment, the calculating according to social information data of different users in the social network model, to obtain activation probability between users, includes:
calculating according to social information data of different users in the social network model to obtain social similarity among the different users, wherein the social similarity is expressed as
；
Wherein,representing user +.>Is->Social similarity between->And->Separate tableShow user +.>Is->Is>Representing the number of users;
calculating according to the social similarity, and acquiring the activation probability between users, wherein the activation probability is expressed as
；
Wherein,representing user +.>Activating user->Probability of->Representing a random probability +.>Is an adjustable parameter.
In one embodiment, the calculating according to the track score and the activation probability, to obtain the comprehensive influence score of the user, includes:
acquiring a userIs +.>Track score according to->Trajectory score and user->Activation ofThe probability of (2) is calculated to obtain the user +.>Is expressed as
；
Wherein,representing adjacent user +.>Track score of>Representing user +.>Is a set of adjacent users;
by means of the userWeighted summation of the trajectory score and the propagation score of (2) to obtain the user +.>Is expressed as
；
Wherein,representing the set parameter between (0, 1), is shown in the specification>Representing user +.>Is a trajectory score of (a).
In one embodiment, updating the comprehensive influence score of the user according to the historical track data of the pre-selected seed user set to obtain an updated comprehensive influence score comprises:
and updating the track score of the user according to the historical track data of the preselected seed user set to obtain an updated track score, and obtaining an updated comprehensive influence score by carrying out weighted summation on the propagation score of the user and the updated track score.
In one embodiment, updating the track score of the user according to the historical track data of the pre-selected seed user set to obtain an updated track score comprises:
updating the track score of the user according to the historical track data of the preselected seed user set to obtain an updated track score expressed as
；
Wherein,representing user +.>Updated trajectory score,/>Representing a pre-selected seed subset->Joining user->Post spatiotemporal coverage, < >>Representing a preselected seed user set->The space-time coverage of the historical track data in the set space-time region.
In one embodiment, the cluster intelligence aware task simulation propagation is performed on a set of preselected seed users, traversing the set of preselected seed usersPropagated final activated user set +.>All users of->And calculating the space-time coverage rate of the pre-selected seed user set in the set space-time region after the propagation, comprising:
for a preselected seed user setPerforming crowd sensing task simulation propagation, and calculating the space-time coverage rate of the historical track data of the pre-selected seed user set after the propagation in a set space-time region, wherein the space-time coverage rate is expressed as
；
Wherein,representing user +.>Sign-on probability +.in each spatiotemporal region in the historical trajectory data of (a)>The sum of the two values,indicating the size of the set spatiotemporal region.
A crowd-aware user selection device based on comprehensive impact assessment, the device comprising:
the social network model construction module is used for acquiring a social network user data set in a set time-out area and constructing a social network model in an independent cascade model form according to the social network user data set; the social network user data set comprises social information data and historical track data of a user;
the comprehensive influence scoring module is used for calculating according to historical track data of the user in the social network model, obtaining track scores of the user, calculating according to social information data of different users in the social network model, obtaining activation probability among the users, and calculating according to the track scores and the activation probability, obtaining comprehensive influence scores of the users;
the seed user preselection module is used for selecting the user with the highest comprehensive influence score to add into the preselection seed user set, updating the comprehensive influence score of the user according to the historical track data of the preselection seed user set to obtain the updated comprehensive influence score, and selecting the user with the highest updated comprehensive influence score to add into the preselection seed user set until the number limit of the preselection seed user set is reached;
and the final seed user set construction module is used for carrying out cluster intelligent perception task simulation propagation on the preselected seed user set, calculating the space-time coverage rate of the propagated preselected seed user set in a set space-time area, selecting the preselected seed user with the largest space-time coverage rate in the preselected seed user set as the final seed user under the condition of setting budget limit, and constructing to obtain the final seed user set.
According to the crowd sensing user selection method and device based on comprehensive influence evaluation, the social network user data set in the set time-out area is obtained, the social network model is built, calculation is carried out according to the track data of the users in the social network model and social information data among the users, comprehensive influence scores of all the users are obtained, the users with the highest comprehensive influence scores are selected to join in a preselected seed user set, after the preselected seed user set reaches the quantity limit, simulation propagation of the users is carried out, the space-time coverage rate of the preselected seed user set after the propagation is calculated, the preselected seed user with the largest space-time coverage rate in the preselected seed user set is selected as the final seed user under the budget limit, and execution of the crowd sensing task of the final seed user set is built. According to the scheme, comprehensive influence evaluation is performed based on social relations of users in a social network, and users with high influence scores and wide space-time coverage rate are selected as seed users, so that efficient propagation of group control perception tasks can be realized on the basis of acquiring a small number of seed users.
Drawings
FIG. 1 is a flow diagram of a crowd-aware user selection method based on comprehensive impact assessment in one embodiment;
FIG. 2 is a flow chart of simulation experiment implementation steps of a crowd sensing user selection method based on comprehensive impact assessment in one embodiment;
FIG. 3 is a schematic diagram of a structural framework of a crowd sensing platform according to one embodiment;
FIG. 4 is a block diagram of a crowd-aware user-selection device based on comprehensive impact assessment in one embodiment.
Detailed Description
The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.
In one embodiment, as shown in fig. 1, a crowd-aware user selection method based on comprehensive impact assessment is provided, comprising the steps of:
step S1, acquiring a social network user data set in a set time-out area, and constructing a social network model in an independent cascade model form according to the social network user data set; wherein the social network user data set includes social information data and historical track data of the user.
It will be appreciated that the social network user data set originates from the social platform and that prior to acquisition of the data set, the task of crowd-sourced perception is considered to have a spatiotemporal limit, with extraneous user data outside the spatiotemporal region of the task to be deleted.
And S2, calculating according to historical track data of the users in the social network model, obtaining track scores of the users, calculating according to social information data of different users in the social network model, obtaining activation probabilities among the users, and calculating according to the track scores and the activation probabilities, so as to obtain comprehensive influence scores of the users.
It can be understood that the task completion group control perception task can be divided into two parts, namely that the user transmits task information to the social partner to complete the task and the user personally complete the task, so that the influence of the user to be selected needs to be considered, the influence of the user is comprehensively evaluated by considering the user's own task completion capability and task transmission capability, wherein the user's own task completion capability can be quantitatively expressed by the increment of the user's own task completion effect (namely, the track score of the user), and the task transmission capability of the user can be quantitatively expressed by the sum of the task completion capabilities of adjacent users of the user (namely, the transmission score of the user).
And S3, selecting the user with the highest comprehensive influence score to add into the preselected seed user set, updating the comprehensive influence score of the user according to the historical track data of the preselected seed user set to obtain the updated comprehensive influence score, and selecting the user with the highest updated comprehensive influence score to add into the preselected seed user set until the quantity limit of the preselected seed user set is reached.
It can be understood that all users are ranked according to the magnitude of the comprehensive influence score, and the user with the highest comprehensive influence score is selected to add into the preselected seed user set, and it is pointed out that the magnitude of the influence of the user is dynamically changed, and the comprehensive influence scores of other users are different according to the different preselected seed user sets, namely, the historical track data of the preselected seed user set updates the comprehensive influence score of the user. Also, since the task costs of crowd-sensing are limited, there are limits to the number of preselected seed users and the number of end seed users based on cost.
And S4, performing cluster intelligent perception task simulation propagation on the preselected seed user set, calculating the space-time coverage rate of the propagated preselected seed user set in a set space-time area, and selecting the preselected seed user with the largest space-time coverage rate in the preselected seed user set as the final seed user under the condition of setting budget limit, so as to construct and obtain the final seed user set.
It will be appreciated that in simulated propagation, multiple simulations are employed, taking the form of an average. Because of the randomness of the propagation, the final user activation is the number of times the user is activated divided by the number of simulations, i.e., a number between 0,1, by simulating the propagation results multiple times. In the simulation propagation, the user has only two possibilities of activation and deactivation in a non-0 or 1 form.
In one embodiment, the calculating according to the historical track data of the user in the social network model to obtain the track score of the user includes:
calculating according to historical track data of the user in the social network model, obtaining track scores of the user, and representing the track scores as
；
Wherein,representing user +.>Track score of>Representing user +.>Time-space coverage generated in a set time-space region of historical track data of (a),>the historical track data of the user is represented,tthe sign-on time is indicated as such,sindicating that the sign-on space is available,prepresenting the sign-on probability.
In one embodiment, the calculating according to social information data of different users in the social network model, to obtain activation probability between users, includes:
calculating according to social information data of different users in the social network model to obtain social similarity among the different users, wherein the social similarity is expressed as
；
Wherein,representing user +.>Is->Social similarity between->And->Respectively represent user +>Is->Is>Representing the number of users;
calculating according to the social similarity, and acquiring the activation probability between users, wherein the activation probability is expressed as
；
Wherein,representing a user/>Activating user->Probability of->The random probability is represented, other conditions except the affinity are represented, and the occurrence of the activation probability 0 between two users is avoided; />The actual state of the activation probability is embodied as an adjustable parameter, and can be set according to the data set and the actual situation.
It will be appreciated that it takes a certain amount of time and effort to complete a crowd sensing task, so users will not easily accept task information from unrelated persons, but rather will tend to accept invitations of friends more closely related to themselves, and thus after social network data of users are obtained, the relationship affinity between users is evaluated through social similarity, so that the probability of users being activated can be calculated.
In one embodiment, the calculating according to the track score and the activation probability, to obtain the comprehensive influence score of the user, includes:
acquiring a userIs +.>Track score according to->Trajectory score and user->Activation->The probability of (2) is calculated to obtain the user +.>Is expressed as
；
Wherein,representing adjacent user +.>Track score of>Representing user +.>Is a set of adjacent users;
by means of the userWeighted summation of the trajectory score and the propagation score of (2) to obtain the user +.>Is expressed as
；
Wherein,representing the set parameter between (0, 1), is shown in the specification>Representing user +.>Is a trajectory score of (a).
In one embodiment, updating the comprehensive influence score of the user according to the historical track data of the pre-selected seed user set to obtain an updated comprehensive influence score comprises:
and updating the track score of the user according to the historical track data of the preselected seed user set to obtain an updated track score, and obtaining an updated comprehensive influence score by carrying out weighted summation on the propagation score of the user and the updated track score.
In one embodiment, updating the track score of the user according to the historical track data of the pre-selected seed user set to obtain an updated track score comprises:
updating the track score of the user according to the historical track data of the preselected seed user set to obtain an updated track score expressed as
；
Wherein,representing user +.>Updated trajectory score,/>Representing a pre-selected seed subset->Joining user->Post spatiotemporal coverage, < >>Representing a preselected seed user set->The space-time coverage of the historical track data in the set space-time region.
It will be appreciated that after a collection of preselected seed users has been selected, it is desirable to consider that multiple perceptions in the same spatio-temporal region are nonsensical, and that re-selection of users with similar trajectories is also considered nonsensical, and therefore it is desirable to update the trajectory scores of the users based on historical trajectory data for the set of preselected seed users.
In one embodiment, performing a cluster intelligence perception task simulation propagation on the pre-selected seed user set, and calculating a space-time coverage rate of the pre-selected seed user set within a set space-time region after the propagation, including:
for a preselected seed user setPerforming crowd sensing task simulation propagation, traversing a preselected seed user set +.>Propagated final activated user set +.>All users of->And calculating the space-time coverage rate of the historical track data of the pre-selected seed user set after propagation in the set space-time region, expressed as
；
Wherein,representing user +.>Sign-on probability +.in each spatiotemporal region in the historical trajectory data of (a)>The sum of the two values,indicating the size of the set spatiotemporal region.
In an embodiment taking a noise monitoring task as an example, the application further carries out a simulation test on a crowd sensing user selection method based on comprehensive influence evaluation, and the specific implementation steps are as shown in fig. 2, including:
1) Task generation, wherein a task publisher (in various forms such as enterprises, individuals and the like) publishes tasks related to noise monitoring in a set space-time area on a crowd sensing platform;
2) The crowd sensing platform receives the noise monitoring task, selects proper seed users, sends task information to the seed users, and describes an excitation mechanism;
3) After receiving the task issued by the crowd sensing platform, the seed user transmits task information to the social partner;
4) Seed users and social partners activated by the seed users are jointly involved in the task of noise monitoring;
5) All participants use their mobile phone microphones or other intelligent wearable devices to sense noise data in the area of their own action tracks, and upload the data to a crowd sensing platform;
6) The crowd sensing platform collects task data uploaded by all participants, processes and analyzes the data and feeds the data back to the task publisher.
In this embodiment, the specific steps of step 1) are described as follows:
1.1 Gathering social network data for selectable users within a city.
1.2 A space-time area of the noise monitoring task is determined, and effective data in the set space-time area is screened out.
1.3 Determining user social information data and user history track data in a set time-out area, and releasing a noise monitoring task to a crowd sensing platform shown in fig. 3, wherein the crowd sensing platform comprises an application layer, a transmission layer and a sensing layer, a task publisher releases the task through a cloud service center in the application layer and transmits the task to a user through the transmission layer, and the user monitors noise in the sensing layer through a smart phone or intelligent wearable device.
In this embodiment, the specific steps of step 2) are described as follows:
the crowd sensing platform receives the noise monitoring task, adopts the crowd sensing user selection method based on comprehensive influence evaluation to select seed users, sends task information to the seed users, and describes an excitation mechanism.
It should be noted that budget and perceived task sets are two constraints that are used when a seed user selects. In the aspect of the propagation set, the set contains all users to be activated, and when any user is activated, the user is moved out of the set, so that when the propagation set is not empty, the program can normally run; when the propagation set is empty, the propagation is ended.
In addition, under the current technical conditions, the crowd sensing platform can theoretically directly acquire the geographic position of the participant, but the problem of participant information leakage possibly exists. The application firstly carries out a fuzzy process when the position information of the participator is obtained, and then carries out encryption process on the geographical position information of the participator, thus, although the precision of part of the position information of the participator can be sacrificed, the information leakage in the transmission process can be avoided.
In this embodiment, the specific steps of step 3) are described as follows:
3.1 After the seed user receives the noise monitoring task, task information is displayed on the social media, and all social partners of the seed user can see the information;
3.2 The other users decide whether to accept the information according to the self situation, and the accepted probability is the activation probability among the users;
3.3 The activated user in turn continues to show the task information in social media, propagating to his social partner, each user may be tried to activate numerous times before being activated;
3.4 When the budget limit is reached or there are no users being attempted to be activated, the task propagation process ends.
It should be understood that, although the steps in the flowcharts of fig. 1 and 2 are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 1 and 2 may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor does the order in which the sub-steps or stages are performed necessarily occur in sequence, but may be performed alternately or alternately with at least a portion of the other steps or sub-steps of other steps.
In one embodiment, as shown in FIG. 4, there is provided a crowd-aware user-selection device based on comprehensive impact assessment, comprising:
the social network model building module 401 is configured to obtain a social network user data set in a set time-out area, and build a social network model in an independent cascading model form according to the social network user data set; the social network user data set comprises social information data and historical track data of a user;
the comprehensive influence scoring module 402 is configured to calculate according to historical track data of a user in the social network model, obtain track scores of the user, calculate according to social information data of different users in the social network model, obtain activation probabilities among the users, calculate according to the track scores and the activation probabilities, and obtain comprehensive influence scores of the users;
the seed user preselection module 403 is configured to select a user with the highest comprehensive influence score to add into the preselection seed user set, update the comprehensive influence score of the user according to the historical track data of the preselection seed user set to obtain an updated comprehensive influence score, and select the user with the highest updated comprehensive influence score to add into the preselection seed user set until the number limit of the preselection seed user set is reached;
and the final seed user set construction module 404 is used for carrying out cluster intelligence perception task simulation propagation on the pre-selected seed user set, calculating the space-time coverage rate of the pre-selected seed user set in a set space-time area after the propagation, and selecting the pre-selected seed user with the largest space-time coverage rate in the pre-selected seed user set as the final seed user under the condition of setting budget limit, so as to construct the final seed user set.
For specific limitations on the crowd-aware user selection means based on the comprehensive impact assessment, reference may be made to the above limitations on the crowd-aware user selection method based on the comprehensive impact assessment, and will not be described in detail herein. The above-described individual modules in the crowd-sourced user selection apparatus based on integrated impact assessment may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in hardware or may be independent of a processor in the computer device, or may be stored in software in a memory in the computer device, so that the processor may call and execute operations corresponding to the above modules.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The above examples merely represent a few embodiments of the present application, which are described in more detail and are not to be construed as limiting the scope of the present application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of the application should be assessed as that of the appended claims.

Claims (8)

1. A crowd sensing user selection method based on comprehensive impact assessment, the method comprising:
acquiring a social network user data set in a set time-out area, and constructing a social network model in an independent cascading model form according to the social network user data set; wherein the social network user data set comprises social information data and historical track data of a user;
calculating according to historical track data of users in the social network model, obtaining track scores of the users, calculating according to social information data of different users in the social network model, obtaining activation probabilities among the users, and calculating according to the track scores and the activation probabilities, obtaining comprehensive influence scores of the users;
selecting the user with the highest comprehensive influence score to add into a preselected seed user set, updating the comprehensive influence score of the user according to the historical track data of the preselected seed user set to obtain an updated comprehensive influence score, and selecting the user with the highest updated comprehensive influence score to add into the preselected seed user set until the number limit of the preselected seed user set is reached;
and carrying out crowd sensing task simulation propagation on the preselection seed user set, calculating the space-time coverage rate of the preselection seed user set in a set space-time area after propagation, and selecting the preselection seed user with the largest space-time coverage rate in the preselection seed user set as a final seed user under the condition of setting budget limit, so as to construct and obtain the final seed user set.
2. The method of claim 1, wherein calculating from historical track data of the user itself in the social network model to obtain the track score of the user comprises:
calculating according to historical track data of the user in the social network model, obtaining track scores of the user, and representing the track scores as
；
Wherein,representing user +.>Track score of>Representing user +.>Time-space coverage generated in a set time-space region of historical track data of (a),>the historical track data of the user is represented,tthe sign-on time is indicated as such,sindicating that the sign-on space is available,prepresenting the sign-on probability.
3. The method of claim 2, wherein calculating based on social information data of different users in the social network model to obtain activation probabilities between users comprises:
calculating according to social information data of different users in the social network model to obtain social similarity among different users, wherein the social similarity is expressed as
；
Wherein,representing user +.>Is->Social similarity between->And->Respectively represent user +>Is->Is>Representing the number of users;
calculating according to the social similarity, and acquiring activation probability among users, wherein the activation probability is expressed as
；
Wherein,representing user +.>Activating user->Probability of->Representing a random probability +.>Is an adjustable parameter.
4. A method according to claim 3, wherein calculating based on the trajectory score and the activation probability to obtain a comprehensive impact score for the user comprises:
acquiring a userIs +.>Trajectory score of (2)According to->Trajectory score and user->Activation->The probability of (2) is calculated to obtain the user +.>Is expressed as
；
Wherein,representing adjacent user +.>Track score of>Representing user +.>Is a set of adjacent users;
by means of the userWeighted summation of the trajectory score and the propagation score of (2) to obtain the user +.>Is expressed as
；
Wherein,representing the set parameter between (0, 1), is shown in the specification>Representing user +.>Is a trajectory score of (a).
5. The method of claim 4, wherein updating the user's integrated influence score based on the historical track data of the preselected seed user set to obtain an updated integrated influence score comprises:
and updating the track score of the user according to the historical track data of the preselected seed user set to obtain an updated track score, and obtaining an updated comprehensive influence score by carrying out weighted summation on the propagation score of the user and the updated track score.
6. The method of claim 5, wherein updating the user's trajectory score based on historical trajectory data of a preselected set of seed users, resulting in an updated trajectory score, comprises:
updating the track score of the user according to the historical track data of the preselected seed user set to obtain an updated track score expressed as
；
Wherein,representing user +.>Updated trackTrace score, ->Representing a pre-selected seed subset->Joining user->Post spatiotemporal coverage, < >>Representing a preselected seed user set->The space-time coverage of the historical track data in the set space-time region.
7. The method of claim 6, wherein performing a crowd-sourced task simulation propagation on the set of preselected seed users and calculating a space-time coverage of the set of preselected seed users within a set space-time region after propagation, comprises:
for the preselected seed user setPerforming crowd sensing task simulation propagation, traversing a preselected seed user set +.>Propagated final activated user set +.>All users of->And calculating the space-time coverage rate of the historical track data of the pre-selected seed user set in the set space-time region after propagation, which is expressed as
；
Wherein,representing user +.>Sign-on probability +.in each spatiotemporal region in the historical trajectory data of (a)>The sum of the two values,indicating the size of the set spatiotemporal region.
8. A crowd-aware user selection device based on comprehensive impact assessment, the device comprising:
the social network model construction module is used for acquiring a social network user data set in a set time-out area and constructing a social network model in an independent cascade model form according to the social network user data set; wherein the social network user data set comprises social information data and historical track data of a user;
the comprehensive influence scoring module is used for calculating according to historical track data of the user in the social network model, obtaining track scores of the user, calculating according to social information data of different users in the social network model, obtaining activation probability among the users, calculating according to the track scores and the activation probability, and obtaining comprehensive influence scores of the users;
the seed user preselection module is used for selecting the user with the highest comprehensive influence score to add into a preselection seed user set, updating the comprehensive influence score of the user according to the historical track data of the preselection seed user set to obtain an updated comprehensive influence score, and selecting the user with the highest updated comprehensive influence score to add into the preselection seed user set until the number limit of the preselection seed user set is reached;
and the final seed user set construction module is used for carrying out cluster intelligence perception task simulation propagation on the pre-selected seed user set, calculating the space-time coverage rate of the pre-selected seed user set in a set space-time area after the propagation, and selecting the pre-selected seed user with the largest space-time coverage rate in the pre-selected seed user set as the final seed user under the condition of setting budget limit, so as to construct and obtain the final seed user set.
CN202311436096.1A 2023-11-01 2023-11-01 Crowd sensing user selection method and device based on comprehensive influence evaluation Active CN117151914B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311436096.1A CN117151914B (en) 2023-11-01 2023-11-01 Crowd sensing user selection method and device based on comprehensive influence evaluation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311436096.1A CN117151914B (en) 2023-11-01 2023-11-01 Crowd sensing user selection method and device based on comprehensive influence evaluation

Publications (2)

Publication Number Publication Date
CN117151914A true CN117151914A (en) 2023-12-01
CN117151914B CN117151914B (en) 2024-01-30

Family

ID=88903207

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311436096.1A Active CN117151914B (en) 2023-11-01 2023-11-01 Crowd sensing user selection method and device based on comprehensive influence evaluation

Country Status (1)

Country Link
CN (1) CN117151914B (en)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016197982A1 (en) * 2015-12-25 2016-12-15 中兴通讯股份有限公司 Crowd sensing method and apparatus
US20170323313A1 (en) * 2015-02-04 2017-11-09 Alibaba Group Holding Limited Information propagation method and apparatus
US20210014654A1 (en) * 2019-07-11 2021-01-14 Arizona Board Of Regents On Behalf Of Arizona State University Systems and methods for crowdsourcing real-time mobile crowd sensing applications
CN113191023A (en) * 2021-05-28 2021-07-30 中国人民解放军国防科技大学 Crowd-sourcing-aware task allocation and user recruitment model cross-validation method and system
CN113298668A (en) * 2021-06-07 2021-08-24 福州大学 Mobile crowd-sourcing aware user large-scale rapid recruitment method considering social network
WO2021213293A1 (en) * 2020-04-24 2021-10-28 西北工业大学 Ubiquitous operating system oriented toward group intelligence perception
CN114943340A (en) * 2022-06-03 2022-08-26 哈尔滨理工大学 Social relationship reasoning and task allocation method for crowd sensing system
CN115511650A (en) * 2022-09-22 2022-12-23 国网智联电商有限公司 Method and device for determining user propagation in crowd sensing task
CN116070852A (en) * 2023-01-10 2023-05-05 深圳大学 Task allocation method, device, equipment and storage medium based on crowd sensing

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170323313A1 (en) * 2015-02-04 2017-11-09 Alibaba Group Holding Limited Information propagation method and apparatus
WO2016197982A1 (en) * 2015-12-25 2016-12-15 中兴通讯股份有限公司 Crowd sensing method and apparatus
US20210014654A1 (en) * 2019-07-11 2021-01-14 Arizona Board Of Regents On Behalf Of Arizona State University Systems and methods for crowdsourcing real-time mobile crowd sensing applications
WO2021213293A1 (en) * 2020-04-24 2021-10-28 西北工业大学 Ubiquitous operating system oriented toward group intelligence perception
CN113191023A (en) * 2021-05-28 2021-07-30 中国人民解放军国防科技大学 Crowd-sourcing-aware task allocation and user recruitment model cross-validation method and system
CN113298668A (en) * 2021-06-07 2021-08-24 福州大学 Mobile crowd-sourcing aware user large-scale rapid recruitment method considering social network
CN114943340A (en) * 2022-06-03 2022-08-26 哈尔滨理工大学 Social relationship reasoning and task allocation method for crowd sensing system
CN115511650A (en) * 2022-09-22 2022-12-23 国网智联电商有限公司 Method and device for determining user propagation in crowd sensing task
CN116070852A (en) * 2023-01-10 2023-05-05 深圳大学 Task allocation method, device, equipment and storage medium based on crowd sensing

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
JIANGTAO WANG，ETC: "Social-Network-Assisted Worker Recruitment in Mobile Crowd Sensing", 《JOURNAL OF LATEX CLASS FILES》, pages 1 - 13 *
方文凤;周朝荣;孙三山;: "移动群智感知中任务分配的研究", 计算机应用研究, no. 11, pages 3206 - 3212 *

Also Published As

Publication number Publication date
CN117151914B (en) 2024-01-30

Similar Documents

Publication Publication Date Title
AU2019226201B2 (en) Complex situation analysis system
US20170316324A1 (en) Computerized Event-Forecasting System and User Interface
Wu et al. On computing mobile agent routes for data fusion in distributed sensor networks
Perez et al. An agent-based approach for modeling dynamics of contagious disease spread
Li et al. An agent-based learning-embedded model (ABM-learning) for urban land use planning: A case study of residential land growth simulation in Shenzhen, China
Bian A conceptual framework for an individual-based spatially explicit epidemiological model
Chessa et al. Empowering mobile crowdsensing through social and ad hoc networking
Wang et al. What can be learned from analyzing university student travel demand?
Silveira et al. MobHet: Predicting human mobility using heterogeneous data sources
CN113191023B (en) Crowd-sourcing-aware task allocation and user recruitment model cross-validation method and system
Ji et al. MOEA/D-based participant selection method for crowdsensing with social awareness
JP2014106585A (en) Information processing device, terminal device, information processing method and program
Yu et al. Supporting serendipitous social interaction using human mobility prediction
CN103501232A (en) Group cooperation system and method based on virtual terminal in mobile environment
Ye et al. Integrating geographic activity space and social network space to promote healthy lifestyles
Ding et al. Dynamic delayed-decision task assignment under spatial-temporal constraints in mobile crowdsensing
Wu et al. A utility-based subcontract method for sensing task in mobile crowd sensing
CN117151914B (en) Crowd sensing user selection method and device based on comprehensive influence evaluation
Frei et al. Modeling spatial embedded social networks
Mare et al. Spatial inequality, neighborhood mobility, and residential segregation
CN110072298B (en) Robust mobile crowd sensing algorithm based on edge calculation
Shen et al. Human mobility in virtual and real worlds: Characterization, modeling, and implications
Montori et al. Performance evaluation of hybrid crowdsensing systems with stateful CrowdSenSim 2.0 simulator
Gao et al. An improved optimization model for crowd evacuation considering individual exit choice preference
Cárcamo et al. Generative models for synthetic populations.

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant