Crowd sensing task allocation method based on multivariate participant preferences

By constructing a multi-dimensional preference model and a Logit model, integrating participant historical task data, the problem of failure to accurately consider users' subjective wishes and multi-dimensional preferences in the existing technology is solved, and a more efficient group intelligence perceived task allocation is achieved.

CN119027955AActive Publication Date: 2024-11-26NORTHWESTERN POLYTECHNICAL UNIV

Patent Information

Application Number
CN202411046345.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-01
Publication Date
2024-11-26
Estimated Expiration
2044-08-01

AI Technical Summary

Technical Problem

The existing group intelligence-perceptual task allocation method based on participants' preferences fails to fully consider the user's subjective wishes and multi-dimensional preferences, resulting in inaccurate task allocation.

Method used

By obtaining participant historical task data, a content, time and place preference model is constructed, and the Logit model is used to integrate preferences in different dimensions, predict the matching degree between participants and tasks, and achieve efficient assignment of tasks.

Benefits of technology

It improves the accuracy and efficiency of task allocation, ensures that tasks are assigned to the most matching participants, thereby improving the effectiveness of group intelligence perception tasks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119027955A_ABST
    Figure CN119027955A_ABST
Patent Text Reader

Abstract

The embodiment of the invention relates to the technical field of crowd sensing, and discloses a crowd sensing task allocation method based on multivariate participant preferences, and the method comprises the steps: obtaining historical task data of a plurality of participants, obtaining a task preference model of each participant based on the historical task data and a new task, and converting the task preference model of each participant into a matching probability of each participant and the new task, and allocating the new task based on the matching probability. According to the multi-participant preference-based crowd sensing task allocation method provided by the embodiment of the invention, the accuracy of deep feature extraction of user historical task data is improved by guiding and fusing the features of two complementary contents of the image and the text; and meanwhile, content information and spatio-temporal information in the historical task data of the user are integrated to represent preferences of the user to the tasks, so that the task allocation accuracy is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present application relate to the field of crowd sensing technology, and in particular to a crowd sensing task allocation method based on the preferences of multiple participants. Background Art

[0002] With the rapid development of mobile devices and sensor technology, we are entering a data-driven era. Mobile Crowd Sensing (MCS), as an emerging perception mode, is gradually becoming one of the key technologies to solve the problem of large-scale data collection and processing. Through the participation of widely distributed mobile devices and users, MCS realizes real-time perception of the environment and dynamic collection of data, providing strong data support for smart cities, environmental monitoring, traffic management and other fields. However, the successful implementation of MCS faces many challenges, among which efficient task allocation is one of the key issues. Among the numerous mobile devices and users, how to reasonably allocate tasks to maximize the perception effect has become an urgent problem to be solved.

[0003] Many algorithms have been proposed in the existing task allocation research, which can be roughly divided into two categories: platform-centric task allocation and user-centric task allocation. In platform-centric task allocation, the methods used include optimization-based methods, auction-based methods, prediction-based methods, and social network-based methods. In user-centric task allocation, the main method used is task allocation based on participant preferences. However, there are still some defects in the above research: 1. In platform-centric task allocation, existing studies generally assume that users will definitely accept the tasks assigned by the platform, but the fact is that the user's subjective willingness has a great influence on the acceptance of the task. At the same time, these studies generally simplify the tasks when dealing with tasks, that is, use several attributes (such as time, place, reward, etc.) as the definition of a task. As the complexity of crowd intelligence perception tasks increases, some complex tasks are difficult to describe with a few attributes. 2. In user-centric task allocation, existing studies generally use text data when utilizing historical task data. In practice, with the development of technology, users often generate data in multiple modalities, such as text, images, and sounds, when completing crowd intelligence perception tasks. In many cases, data in various modalities complement each other. Only by integrating all data can the user's intention be accurately captured. At the same time, when analyzing user preferences, existing studies mostly only consider preferences in a single dimension, while user preferences may be multidimensional. Summary of the invention

[0004] The purpose of the embodiment of the present application is to provide a method for allocating tasks based on the preferences of multiple participants, aiming to solve the technical problem that the existing task allocation method based on the preferences of participants only considers the preferences of a single dimension of the user when analyzing the user preferences, and does not consider the user's subjective willingness to accept the task, resulting in inaccurate task allocation. The embodiment of the present application provides a method for allocating tasks based on the preferences of multiple participants, which makes full use of the large amount of multimodal data generated by the participants when completing the group intelligence perception tasks, models the participants' preferences for tasks from the three dimensions of content, time and place, and then regards the prediction of the matching degree between the task and the participant as a binary classification problem, and uses the Logit model to integrate the preferences of different dimensions together, so as to predict the matching degree of each pair of participants and tasks. Finally, the tasks to be assigned are assigned to several participants with the highest matching degree, and the accuracy and efficiency of task allocation are improved by accurately modeling the preferences of the participants.

[0005] To solve the above technical problems, an embodiment of the present application provides a method for allocating group intelligence perception tasks based on the preferences of multiple participants, comprising the following steps: obtaining historical task data of several participants, wherein the historical task data includes text data, image data, time data and location data of each task; constructing and acquiring a content preference model and the content preference of each participant based on the text data and image data, constructing and acquiring a time preference model and the time preference of each participant based on the time data, and constructing a location preference model based on the location data; receiving a new task, wherein the new task includes new text data, new image data, new time data and new location data; inputting the new text data and the new image data into the content preference model to obtain a new content preference, using the new content preference and the content preference of each participant to obtain the content matching degree of each participant, using the new time data and the time preference of each participant to obtain the time matching degree of each participant, and inputting the new location data into the location preference model to obtain the location matching degree of each participant; weightedly fusing the content matching degree, the time matching degree and the location matching degree to construct a task preference model for each participant; converting the task preference model of each participant into a matching probability between each participant and the new task; and allocating the new task based on the matching probability.

[0006] The embodiment of the present application also provides a group intelligence perception task allocation system based on the preferences of multiple participants, including: a data acquisition module, a one-dimensional preference acquisition module, a new task acceptance processing module, a matching probability acquisition module, and a new task allocation module; the data acquisition module is used to acquire the historical task data of several participants, wherein the historical task data includes the text data, image data, time data and location data of each task; the one-dimensional preference acquisition module is used to train and acquire a content preference model and the content preference of each participant based on text data and image data, train and acquire a time preference model and the time preference of each participant based on time data, and train and acquire a location preference model based on location data; the new task acceptance processing module is used to receive new tasks, wherein the new tasks It includes new text data, new image data, new time data and new location data; the new text data and the new image data are input into the content preference model to obtain the new content preference, the new content preference and the content preference of each participant are used to obtain the content matching degree of each participant, the new time data and the time preference of each participant are used to obtain the time matching degree of each participant, and the new location data is input into the location preference model to obtain the location matching degree of each participant; the matching probability acquisition module is used to weightedly fuse the content matching degree, the time matching degree and the location matching degree to construct a task preference model for each participant; the task preference model of each participant is converted into the matching probability of each participant and the new task; the new task allocation module is used to allocate new tasks based on the matching probability.

[0007] An embodiment of the present application also provides an electronic device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned method of group intelligence perception task allocation based on the preferences of multiple participants.

[0008] An embodiment of the present application further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned method for allocating group intelligence perception tasks based on the preferences of multiple participants.

[0009] The embodiment of the present application provides a method for group intelligence perception task allocation based on the preferences of multiple participants. First, the accuracy of deep feature extraction of user historical task data is improved by guiding and fusing the features of two complementary contents, image and text. Secondly, the embodiment of the present application also integrates the content information and spatiotemporal information in the user's historical task data to characterize the user's preference for tasks, thereby improving the accuracy of task allocation. Finally, the embodiment of the present application processes task allocation as a binary classification problem, and uses the Logit model to unify all factors into a framework, calculates the matching probability of tasks and participants, and improves the efficiency of task allocation.

[0010] In some optional embodiments, constructing and acquiring a content preference model and each participant's content preference based on text data and image data includes:

[0011] The content preference model is trained and obtained based on text data and image data, and the content preference of each participant is obtained by inputting the result of all historical tasks of each participant into the content preference model.

[0012] In some optional embodiments, the content preference model includes: a feature extraction module, a feature fusion module and a multi-label classification module; the input of the content preference model is the text data and image data of the Nth task, and the output is the content preference of the Nth task.

[0013] In some optional embodiments, the feature fusion module includes:

[0014] The text features and image features are used for semantic alignment and cross-guidance to form guided image features and guided text features respectively, which are then weightedly fused to construct the content features of each task.

[0015] In some optional embodiments, the content category of the Nth task is embedded as a label in the multi-label classification module.

[0016] In some optional embodiments, allocating the new task based on the matching probability includes: allocating the new task to several participants with the highest matching probability.

[0017] In some optional embodiments, after allocating the new task based on the matching probability, the method further includes: uploading the task allocation result to the platform, and the task publisher obtains the task allocation result through the platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] One or more embodiments are exemplarily described by the pictures in the corresponding drawings, and these exemplary descriptions do not constitute limitations on the embodiments.

[0019] Figure 1is a flowchart of a method for allocating group intelligence sensing tasks based on preferences of multiple participants provided by an embodiment of the present application;

[0020] Figure 2 It is a framework diagram of a method for allocating group intelligence sensing tasks based on preferences of multiple participants provided by an embodiment of the present application;

[0021] Figure 3 is a framework diagram of a content preference extraction step provided by an embodiment of the present application;

[0022] Figure 4 is a set of coordinates of a participant when participating in a task provided by an embodiment of the present application;

[0023] Figure 5 is a Gaussian mixture distribution diagram provided by an embodiment of the present application;

[0024] Figure 6 Another embodiment of the present application provides a schematic diagram of a group intelligence sensing task allocation system based on preferences of multiple participants;

[0025] Figure 7 It is a schematic diagram of the structure of an electronic device provided by another embodiment of the present application. DETAILED DESCRIPTION

[0026] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the embodiments of the present application will be described in detail below in conjunction with the accompanying drawings. However, it will be appreciated by those skilled in the art that in the present application, many technical details are proposed in order to enable the reader to better understand the present application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical scheme claimed in the present application can also be implemented. The division of the following embodiments is for the convenience of description, and the specific implementation of the present application should not be construed as any limitation, and the various embodiments can be combined and referenced with each other under the premise of no contradiction.

[0027] An embodiment of the present application relates to a method for allocating group intelligence perception tasks based on the preferences of multiple participants, which is applied to an electronic device, wherein the electronic device can be a terminal or a server. In this embodiment and the following embodiments, the electronic device is described by taking the server as an example. The implementation details of the method for allocating group intelligence perception tasks based on the preferences of multiple participants in this embodiment are described in detail below. The following content is only the implementation details provided for the convenience of understanding and is not necessary for the implementation of this solution.

[0028] The specific process of the method for allocating group intelligence tasks based on the preferences of multiple participants in this embodiment can be as follows: Figure 1 As shown, including:

[0029] Step 101, obtaining historical task data of several participants, wherein the historical task data includes text data, image data, time data and location data of each task.

[0030] It should be noted that the core of the embodiment of the present application is to assign tasks to several participants with the highest degree of adaptability. The task assignment method based on the preferences of multiple participants proposed in the embodiment of the present application first analyzes the historical task data generated when the participants participate in the task. The embodiment of the present application analyzes the preferences of the participants for the task from the three dimensions of content, time and location in the historical task data, integrates the preferences of different dimensions, and thus predicts the degree of matching between different participants and tasks, and then assigns the task to be assigned to the participant with the highest degree of matching with the task, completing the group intelligence perception task assignment based on the preferences of multiple participants.

[0031] See also Figure 2 , Figure 2 A framework diagram of a task allocation method based on the preferences of multiple participants provided in an embodiment of the present application is presented, and the framework includes: preference extraction and task allocation, wherein preference extraction includes: participant content preference extraction, time preference extraction and location preference extraction.

[0032] In this embodiment, the task publisher publishes the task content to the crowd-sensing platform, where the task content includes: the location of the task, the time of the task and the content of the task, and the task data is in the format of text or picture. When the crowd-sensing platform receives a new task, it analyzes the historical task data generated when several participants participate in the crowd-sensing task, calculates the task preference of each participant, and then obtains the matching degree between each participant and the new task based on the task preference.

[0033] In this embodiment, T = {t1, t2, ..., t n} is the set of tasks in the platform, where t i represents the i-th task, let T u Represents the historical task data of the u-th participant participating in the task, and the historical task data includes text data, image data, location data and time data.

[0034] Step 102, training and acquiring a content preference model and each participant's content preference based on text data and image data, training and acquiring a time preference model and each participant's time preference based on time data, and training and acquiring a location preference model based on location data.

[0035] This embodiment is TP u , L.P. u , CP u is the time preference, location preference and content preference of the u-th participant for the task.

[0036] This embodiment first constructs and obtains a content preference model and each participant's content preference based on text data and image data.

[0037] See also Figure 3 , Figure 3 The overall framework diagram of content preference extraction in this embodiment, that is, the overall architecture diagram of the content preference model, is presented, wherein the framework mainly includes three parts: feature extraction module, feature fusion module and multi-label classification module. The process of framework execution is: first, the features of text and image are extracted separately, then the text features and image features are fused using the cross-attention method, and then the Transformer decoder is used for multi-label classification to obtain the category label of the data, and the category labels of all historical task data of the participants are merged to obtain the participants' preferences for the task content. The specific steps are as follows:

[0038] First, the text data and image data in the historical task data are preprocessed to obtain text features and image features. The specific steps of preprocessing are: use the pre-trained BERT model to extract text features and obtain text features FT = BERT (t). Then, based on the Transformer model, the CNN model is integrated and the outputs of the two models are spliced ​​so that the model has both local and global feature extraction capabilities, thereby obtaining image features FI = Transformer (t) + CNN (t).

[0039] At this point, the preprocessing operations on the text data and image data in each participant's historical task data are completed, and the text features and image features are extracted.

[0040] It should be noted that in group intelligence tasks, data of various modalities are often very concise, and data of any modality are equally important for the task. Therefore, the embodiment of the present application considers using text features and image features for semantic alignment and cross-guidance to form text-guided attention features and image-guided attention features, respectively, to achieve complementarity and enhancement of the two modalities.

[0041] The embodiment of the present application uses a feature fusion method based on a cross-attention mechanism to achieve the complementarity and enhancement of the two modalities. Specifically, when text guides image features, the text is used to strengthen the features of the image features that the text also pays attention to; when the image guides the text features, the image is used to strengthen the features of the text features that the image also pays attention to.

[0042] The specific steps of the embodiment of the present application for fusing image features and text features based on the cross attention mechanism are as follows:

[0043] Let the text feature be FT = {α1, α2, ..., α n}, the image feature is FI = {η1, η2, ..., ηn}, when text guides image features, the text is regarded as the query (Q), and the image features are regarded as the key (K) and value (V), that is, The guided image features are:

[0044]

[0045] Where d is the dimension of the key vector.

[0046] When images guide text features, image features are considered as queries (Q) and text features are considered as keys (K) and values ​​(V), that is, The text features after guidance are:

[0047]

[0048] Where d is the dimension of the key vector.

[0049] The guided image feature FI′ and the guided text feature FT′ are fused to obtain the content feature F. The fusion calculation formula is:

[0050] F=λFI′+(1-λ)FT′

[0051] Wherein, λ is the weight of the image feature. It should be noted that, in this embodiment, λ is set to 0.39, and the implementer may adjust it according to the specific implementation scenario.

[0052] At this point, the image features and text features are fused to obtain the content features of each task in the historical task data of each participant.

[0053] It should be noted that after the above processing, the content features of each task of each participant's historical task data are obtained, and then the multi-label classification process is performed. The embodiment of the present application uses a Transformer decoder to implement multi-label classification. Compared with the traditional Transformer decoder, which needs to set labels first for neural network training, the embodiment of the present application converts all category labels of the data into label embeddings, which are used as queries for the decoder to check the existence of each label, and execute a multi-head attention mechanism to strengthen the features related to the category in the features.

[0054] The specific steps of implementing multi-label classification using a Transformer decoder in this application embodiment are as follows:

[0055] This embodiment takes the Nth task in the historical task data of the Hth participant as an example, takes the categories of all the contents in the Nth task as labels, and embeds the labels into Q through the self-attention layer to learn the correlation between labels. Then, it enters the multi-head cross attention layer with the feature vector extracted by the feature extraction layer, i.e., the content feature, to strengthen the category-related features, and finally enters the feedforward neural network layer to perform linear processing on the features. The input of each Transformer decoder layer i comes from the i-1th decoder layer, and the output of each layer in the decoder is as follows:

[0056]

[0057] O i =FFN(O′ i )

[0058] in:

[0059] MultiHead(Q,K,V)=Concat(h1,h2,…,h n )W p

[0060]

[0061] FFN(x)=max(0,xW1+b1)W2+b2

[0062] W is a different transformation matrix.

[0063] In order to reduce the impact of label imbalance on the results, the loss function used in this embodiment is focal loss, which enables the model to pay more attention to samples that are difficult to classify and improve the recognition ability of minority categories.

[0064]

[0065] The final output of the model is: j =[o1, o2, …, o h ], where h is the number of label categories, o i is the predicted probability of the i-th category in the feature.

[0066] At this point, the content preference model is completed.

[0067] The above processing is performed on the text data and image data of each task in all historical task data of the participants, and the output results are combined to obtain the content preference CP of the participants. u .

[0068] CP u =Concat(O j )

[0069] It should be noted that this embodiment obtains all outputs O of all historical task data of participants. j This was summed and divided by the total number of categories to obtain the content preference for that participant.

[0070] At this point, the content preference of each participant is obtained.

[0071] Furthermore, this embodiment constructs and obtains a time preference model and the time preference of each participant based on time data.

[0072] Using a 30-dimensional vector D g Indicates the task execution time in a month. If the task execution time is within a specific time period, the corresponding part of the vector is set to 1, and the rest is set to 0. In this embodiment, a month is set to 30 days. Therefore, a time preference model is constructed. The input is the execution time of a task, and the output is the time preference of the task. The time preference TP of the uth participant for the task is u For example, the specific calculation formula of the time preference model is:

[0073]

[0074] Among them, t i represents the i-th task, T u Represents the historical task data of the u-th participant.

[0075] At this point, the time preference model and the time preference of each participant are obtained.

[0076] Furthermore, this embodiment constructs a location preference model based on location data.

[0077] This embodiment uses a method based on Gaussian mixture distribution to model the location data and obtain the location preference of each participant. The input of the location preference model is the location data of the task, and the output is the location preference of the participant for the task. It should be noted that the location data is the coordinate set of the participant when participating in the task. Figure 4 , Figure 4 The coordinate sets of several participants participating in the task are presented. Let the coordinate set P of the task selected by the uth participant be u for:

[0078]

[0079] in, represents the i-th task t i The coordinates of T u Represents the historical task data of the u-th participant.

[0080] The Gaussian mixture distribution of the historical task location of the uth participant is:

[0081]

[0082] in:

[0083]

[0084] In the formula, D represents the dimension of the historical task data of the uth participant. It should be noted that this formula is a Gaussian mixture distribution formula, and the meaning of the letters in this formula is consistent with that in the Gaussian mixture distribution formula.

[0085] Furthermore, participants' location preferences for tasks u =p(x).

[0086] See also Figure 5 , Figure 5 The Gaussian mixture distribution obtained in this embodiment is shown. It should be noted that when a new task appears, the matching degree between the task and the location of the participant can be calculated based on the Gaussian mixture distribution. As can be seen from the figure, the darker the color, the more the scope of the task conforms to the activity pattern of the participant.

[0087] At this point, the location preference model is completed.

[0088] Step 103 , receiving a new task and obtaining a content matching degree, a time matching degree and a location matching degree based on the new task, the content preference model, the time preference model and the location preference model.

[0089] It should be noted that after obtaining each participant's preference for the three dimensions of the task, when a new task appears on the platform, the degree of match between each participant and each dimension of the task can be calculated.

[0090] Receive new tasks released by the platform, where the new tasks include new text data, new image data, new time data and new location data; input the new text data and new image data into the content preference model to obtain new content preferences, use the new content preferences and the content preferences of each participant to obtain the content matching degree of each participant, use the new time data and the time preferences of each participant to obtain the time matching degree of each participant, and input the new location data into the location preference model to obtain the location matching degree of each participant.

[0091] Assume that the new task is A, then the content matching degree of the u-th participant to the new task A is: The time matching and location matching are: Among them, cos(*) means to find the cosine similarity, O A is the output result obtained after inputting the new task A into the content preference model, D A is the time data of the new task A, i.e., the execution time, l AThe location data for the new task is the coordinates.

[0092] At this point, the content matching degree, time matching degree and location matching degree of each participant are obtained.

[0093] Step 104, weighted fusion of the content matching degree, time matching degree and location matching degree is performed to construct a task preference model for each participant; and the task preference model for each participant is converted into a matching probability between each participant and the new task.

[0094] This embodiment takes the u-th participant as an example to construct the preference model of the u-th participant. The specific formula is:

[0095]

[0096] Where W = (w1, w2, w3) represents the weight of different features, b represents the deviation, represents the content matching degree of the u-th participant to the new task A, represents the time matching degree of the u-th participant to the new task A, It represents the location matching degree of the u-th participant for the new task A.

[0097] Since the result of whether the participant accepts the task or not is binary, this paper converts the prediction of the degree of match between the participant and the task into a binary classification problem. This paper uses a logistic function to convert the preference model into a probability function, that is, to convert the preference model into a matching probability, as follows:

[0098]

[0099] Let the cost function be:

[0100]

[0101] At this point, the matching probability between each participant and the new task is obtained.

[0102] Step 105: Allocate new tasks based on the matching probability.

[0103] After the platform receives a new task, it calculates the matching probability between each participant and the task, and assigns the task to several participants with the highest matching probability based on task requirements.

[0104] After receiving the assigned task, participants can accept or reject the task at their own will. If they accept the task, they must complete it as required and upload the results to the platform.

[0105] The task publisher obtains the task assignment results through the platform.

[0106] The step division of the above methods is only for clear description. When implemented, they can be combined into one step or some steps can be split and decomposed into multiple steps. As long as they include the same logical relationship, they are all within the scope of protection of this application; adding insignificant modifications to the algorithm or process or introducing insignificant designs without changing the core design of the algorithm and process are all within the scope of protection of this application.

[0107] Another embodiment of the present application relates to a group intelligence perception task allocation system based on the preferences of multiple participants. The implementation details of the group intelligence perception task allocation system based on the preferences of multiple participants in this embodiment are specifically described below. The following content is only for the convenience of understanding the implementation details provided, and is not necessary for the implementation of this solution. The schematic diagram of the group intelligence perception task allocation system based on the preferences of multiple participants in this embodiment can be as follows: Figure 6 As shown, including:

[0108] Data acquisition module 201, single-dimensional preference acquisition module 202, new task acceptance processing module 203, matching probability acquisition module 204, new task allocation module 205;

[0109] The data acquisition module 201 is used to acquire the historical task data of several participants, wherein the historical task data includes text data, image data, time data and location data of each task;

[0110] A one-dimensional preference acquisition module 202, for training and acquiring a content preference model and the content preference of each participant based on text data and image data, training and acquiring a time preference model and the time preference of each participant based on time data, and training and acquiring a location preference model based on location data;

[0111] A new task acceptance processing module 203, for receiving a new task and obtaining a content matching degree, a time matching degree and a location matching degree based on the new task, a content preference model, a time preference model and a location preference model;

[0112] The matching probability acquisition module 204 is used to construct a task preference model for each participant by weighted integration of the content matching degree, the time matching degree and the location matching degree; and to convert the task preference model for each participant into a matching probability between each participant and the new task;

[0113] The new task allocation module 205 is used to allocate new tasks based on the matching probability.

[0114] It is not difficult to find that this embodiment is a system embodiment corresponding to the above method embodiment, and this embodiment can be implemented in conjunction with the above method embodiment. The relevant technical details and technical effects mentioned in the above embodiments are still valid in this embodiment, and in order to reduce repetition, they are not repeated here. Accordingly, the relevant technical details mentioned in this embodiment can also be applied in the above embodiments.

[0115] It is worth mentioning that all modules involved in this embodiment are logic modules. In practical applications, a logic unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. In addition, in order to highlight the innovative part of this application, this embodiment does not introduce units that are not closely related to solving the technical problems proposed by this application, but this does not mean that there are no other units in this embodiment.

[0116] Another embodiment of the present application relates to an electronic device, such as Figure 7 As shown, it includes: at least one processor 301; and a memory 302 that is communicatively connected to the at least one processor 301; wherein the memory 302 stores instructions that can be executed by the at least one processor 301, and the instructions are executed by the at least one processor 301 so that the at least one processor 301 can execute the group intelligence perception task allocation method based on the preferences of multiple participants in the above-mentioned embodiments.

[0117] Among them, the memory and the processor are connected in a bus manner, and the bus may include any number of interconnected buses and bridges, and the bus connects various circuits of one or more processors and memories together. The bus can also connect various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be one element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices on a transmission medium. The data processed by the processor is transmitted on a wireless medium via an antenna, and further, the antenna also receives data and transmits the data to the processor.

[0118] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory can be used to store data used by the processor when performing operations.

[0119] Another embodiment of the present application relates to a computer-readable storage medium storing a computer program, which implements the above method embodiment when executed by a processor.

[0120] That is, those skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium, including a number of instructions to enable a device (which can be a single-chip microcomputer, chip, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as: ROM), random access memory (Random Access Memory, referred to as: RAM), disk or optical disk and other media that can store program codes.

[0121] Those skilled in the art will appreciate that the above embodiments are specific embodiments for implementing the present application, and in actual applications, various changes may be made thereto in form and detail without departing from the spirit and scope of the present application.

Claims

1. Crowd-sensing task allocation method based on preferences of multiple participants, It is characterized in that include: Acquire historical task data of several participants, wherein the historical task data includes text data, image data, time data and location data; Build and obtain a content preference model and the content preference of each participant based on text data and image data, build and obtain a time preference model and the time preference of each participant based on time data, and build a location preference model based on location data; receiving a new task, wherein the new task includes new text data, new image data, new time data, and new location data; Input the new text data and the new image data into the content preference model to obtain the new content preference, use the new content preference and the content preference of each participant to obtain the content matching degree of each participant, use the new time data and the time preference of each participant to obtain the time matching degree of each participant, input the new location data into the location preference model to obtain the location matching degree of each participant; The content matching degree, time matching degree and location matching degree are weighted and integrated to construct the task preference model of each participant; Convert each participant's task preference model into the matching probability between each participant and the new task; New tasks are assigned based on the matching probability.

2. According to claim 1, the method for allocating group intelligence tasks based on the preferences of multiple participants, It is characterized in that The step of constructing and acquiring a content preference model and each participant's content preference based on text data and image data includes: A content preference model is constructed based on text data and image data, and the content preference of each participant is obtained by inputting the result of all historical tasks of each participant into the content preference model.

3. According to claim 2, the method for group intelligence sensing task allocation based on the preferences of multiple participants, It is characterized in that The content preference model includes: Feature extraction module, feature fusion module and multi-label classification module; The input of the content preference model is the text data and image data of the Nth task, and the output is the content preference of the Nth task.

4. The method for allocating group intelligence tasks based on preferences of multiple participants according to claim 3, It is characterized in that The feature fusion module comprises: The text features and image features are used for semantic alignment and cross-guidance to form guided image features and guided text features respectively, which are then weightedly fused to construct the content features of each task.

5. The method for allocating group intelligence tasks based on preferences of multiple participants according to claim 3, It is characterized in that The content category of the Nth task is embedded as a label into the multi-label classification module.

6. The method for allocating group intelligence tasks based on preferences of multiple participants according to claim 1, It is characterized in that The allocating the new task based on the matching probability includes: allocating the new task to a number of participants with the highest matching probability.

7. The method for allocating group intelligence tasks based on preferences of multiple participants according to claim 1, It is characterized in that After allocating the new task based on the matching probability, the method further includes: uploading the task allocation result to the platform, and the task publisher obtains the task allocation result through the platform.

8. Crowd-sensing task allocation system based on preferences of multiple participants, It is characterized in that The system includes a data acquisition module, a unidimensional preference acquisition module, a new task acceptance processing module, a matching probability acquisition module, and a new task allocation module; The data acquisition module is used to acquire historical task data of several participants, wherein the historical task data includes text data, image data, time data and location data of each task; The one-dimensional preference acquisition module is used to train and acquire a content preference model and the content preference of each participant based on text data and image data, train and acquire a time preference model and the time preference of each participant based on time data, and train and acquire a location preference model based on location data; The new task acceptance processing module is used to receive a new task, wherein the new task includes new text data, new image data, new time data and new location data; input the new text data and the new image data into the content preference model to obtain the new content preference, use the new content preference and the content preference of each participant to obtain the content matching degree of each participant, use the new time data and the time preference of each participant to obtain the time matching degree of each participant, and input the new location data into the location preference model to obtain the location matching degree of each participant; The matching probability acquisition module is used to construct a task preference model for each participant by weighted fusion of the content matching degree, time matching degree and location matching degree; and convert the task preference model for each participant into a matching probability between each participant and a new task; The new task allocation module is used to allocate new tasks based on the matching probability.

9. An electronic device, It is characterized in that include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the crowd-sensing task allocation method based on the preferences of multiple participants as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, It is characterized in that When the computer program is executed by a processor, the method for allocating group intelligence tasks based on the preferences of multiple participants as described in any one of claims 1 to 7 is implemented.

Citation Information

Patent Citations

  • Social relation reasoning and task allocation method for crowd sensing system

    CN114943340A

  • Mobile crowd sensing task recommendation method fusing multi-modal data features

    CN114969534A

  • Multi-person cooperation task allocation method for crowd sensing

    CN115577910A

  • Crowd sensing task allocation method fusing willingness of participants under multiple constraints and application

    CN116957235A

  • Software-defined satellite shared network-oriented on-satellite task order grabbing system

    CN117319505A

Cited By

  • Matching method considering preferences of tasks and participants

    CN119441907A