User demand big data mining method and system applied to smart community
By identifying key messages and updating tags in smart community conversation logs, and combining global organization and feature mining technologies, the resource overhead and accuracy issues in user demand mining were resolved, and efficient mining of business processing demand information was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHANGSHU HAOYU ELECTRONICS INFORMATION TECH
- Filing Date
- 2026-04-14
- Publication Date
- 2026-07-21
AI Technical Summary
In smart communities, existing technologies often change the data format by default during the user needs mining process, which increases resource consumption and makes it difficult to guarantee mining accuracy.
By identifying key messages and updating tag information in community session logs, and using intelligent processing threads to identify and organize key message sets in the session state, business processing requirements are determined. A global organization strategy and feature mining technology are adopted to reduce resource overhead and improve mining efficiency and accuracy.
Without changing the log recording format, the accuracy and efficiency of user demand mining were improved, the computational load and resource consumption were reduced, and efficient business processing demand information mining was achieved.
Smart Images
Figure CN122432228A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the fields of smart community and user demand mining technology, and in particular to a method and system for big data mining of user demand applied to smart communities. Background Technology
[0002] Smart communities are a type of urban area (community) that balances social, business, and environmental needs while optimizing available resources. They aim to provide various processes, systems, and products to promote urban (community) development and sustainability, benefiting residents, the economy, and the broader ecological environment upon which the urban (community) depends. This is achieved by applying information technology (IT) to plan, design, build, and operate urban (community) infrastructure, thereby improving quality of life and economic well-being.
[0003] As smart communities continue to develop, people are pursuing higher quality of life. To meet this demand, it is necessary to understand user needs and improve smart communities accordingly. However, relevant user need mining technologies often default to changing the data format of the logs to be mined. This not only increases the resource overhead for mining and analysis but also makes it difficult to ensure the accuracy of user need mining. Summary of the Invention
[0004] To address the technical problems existing in related technologies, this application provides a method and system for big data mining of user needs applied to smart communities.
[0005] On the one hand, this application provides a method for big data mining of user needs in smart communities, applied to a smart community service system. The method includes at least: identifying key messages in community session logs that meet user need mining conditions to obtain a first key message set under multiple session states; updating the tag information of the first key message set to obtain a second key message set corresponding to each session state; wherein the tag information of the second key message sets corresponding to the first key message sets under different session states is the same; sequentially updating the tag information of the second key message sets under each session state to obtain a third key message set corresponding to each session state, wherein the statistical results of the time-series description of the third key message set under each session state match the set indicators; and using the third key message set to determine the business processing demand information in the community session logs that meet the user need mining conditions. Thus, while ensuring mining accuracy, it reduces the resource overhead of mining and analyzing community session logs to a certain extent, and improves the mining efficiency for business processing demand information.
[0006] In one independently implementable embodiment, the step of identifying key messages in community session logs that meet user demand mining conditions to obtain a first key message set under multiple session states includes: identifying key messages in community session logs that meet user demand mining conditions through first intelligent processing threads under multiple session states to obtain a first key message set exported by the first intelligent processing thread under each session state; the step of updating the tag information of the first key message set to obtain a second key message set corresponding to the first key message set under each session state includes: determining the thread tag information of the second intelligent processing thread corresponding to the first intelligent processing thread under that session state based on the determined updated tag information and the tag information of the first key message set exported by the first intelligent processing thread under each session state; and performing feature mining operations on the first key message set exported by the first intelligent processing thread corresponding to the second intelligent processing thread under that session state, based on the second intelligent processing thread under each session state that includes the determined thread tag information, to obtain the second key message set exported by the second intelligent processing thread under that session state. This reduces the computational load and improves the efficiency of mining information for business processing needs.
[0007] In one independently implementable embodiment, the step of updating the tag information of the first key message set to obtain the second key message set corresponding to the first key message set in each session state includes: Determine the first key message set with the fewest community interaction constraints among the tag information corresponding to the first key message set in each session state, and update the other first key message sets except the first key message set with the fewest community interaction constraints to key message sets with the same tag information as the first key message set with the fewest community interaction constraints. The first key message set with the fewest community interaction constraints and the updated key message set with the same tag information as the first key message set with the fewest community interaction constraints are used as the second key message set. Alternatively, the first key message set in each session state can be updated to a key message set under a specified tag, and this key message set under the specified tag can be used as the second key message set. In this way, updating the first key message set in each session state to a minimum set of community interaction constraints can reduce the computational resource overhead and improve the efficiency of demand mining when mining business processing demand information contained in community session logs that meet user demand mining conditions.
[0008] In one independently implementable embodiment, the step of identifying key messages in community session logs that meet user demand mining conditions to obtain a first key message set under multiple session states includes: identifying key messages in community session logs that meet user demand mining conditions through first intelligent processing threads under multiple session states to obtain a first key message set exported by the first intelligent processing thread under each session state; the step of sequentially updating the tag information of the second key message set under each session state to obtain a third key message set corresponding to the second key message set under each session state includes: statistical results based on the time-series description between the first intelligent processing threads under different session states, and the second key message set corresponding to the first intelligent processing thread under each session state. The temporal description of the first intelligent processing thread in each session state is used to determine the temporal description of the third key message set corresponding to each session state. Based on the determined temporal description of the third key message set corresponding to the first intelligent processing thread in each session state, and the temporal description of the second key message set corresponding to the first intelligent processing thread in each session state, the thread tag information of the third intelligent processing thread corresponding to the first intelligent processing thread in that session state is determined. According to the third intelligent processing thread in each session state that includes the determined thread tag information, feature mining is performed on the second key message set corresponding to the third intelligent processing thread in that session state to obtain the third key message set exported by the third intelligent processing thread in that session state. In this way, business processing requirement information included in community session logs that meet user requirement mining conditions can be mined more accurately, improving the accuracy of the mining.
[0009] In one independently implementable embodiment, determining the business processing requirement information in the community session logs that meet the user requirement mining conditions using the third key message set includes: performing a global reorganization operation on the third key message set corresponding to the second key message set in each session state to obtain a globally reorganized fourth key message set; and using the fourth key message set to determine the business processing requirement information in the community session logs that meet the user requirement mining conditions. This improves the accuracy of user requirement mining.
[0010] In one independently implementable embodiment, a global reorganization operation is performed on the third key message set corresponding to the second key message set in each session state to obtain a globally reorganized fourth key message set. This includes: performing a global reorganization operation on each of the third key message sets corresponding to the second key message set in each session state according to a pre-set global reorganization strategy to obtain a transitional key message set after each round of global reorganization; and using the transitional key message set after each round of global reorganization to obtain the fourth key message set. Thus, the global reorganization operation is performed on each of the third key message sets to ensure the accuracy of the transitional key message sets. Based on this, the fourth key message set is accurately obtained using the transitional key message set after each round of global reorganization.
[0011] In one independently implementable embodiment, the third key message set corresponding to the second key message set in each session state is used as the third key message set from the first session state to the Xth session state, where the temporal description of the third key message set in the Xth session state is greater than that of the third key message set in the (X-1)th session state, and X is a positive integer greater than 1. Then, according to a pre-set global reorganization strategy, the third key message sets corresponding to the second key message sets in each session state are globally reorganized one by one to obtain the transition key message set after each round of global reorganization, including: based on the third key message set from the first session state to the Xth session state... The global organization strategy for the third key message set involves sequentially organizing the third key message set in each session state to obtain the key message set after each round of global organization. The third key message set in the first session state and the key message set after each round of global organization are considered as the obtained transitional key message set; or, according to the global organization strategy from the third key message set in the Xth session state to the third key message set in the first session state, the global organization strategy for the third key message set in each session state is sequentially organized to obtain the key message set after each round of global organization. The third key message set in the Xth session state and the key message set after each round of global organization are considered as... The obtained transition key message set; or, according to the global organization strategy from the third key message set in the first session state to the third key message set in the Xth session state, the third key message set in each session state is globally organized, and key message sets after each round of global organization are obtained sequentially when the third key message set in the first session state to the third key message set in the Xth session state are globally organized. Feature mining operations are then performed on the third key message set in the first session state and the key message sets after each round of global organization to obtain the globally organized key message set from the first session state to the Xth session state. The label information of the globally organized key message set in the conversation state is the same as the label information of the key message set before the feature mining operation; according to the global organization strategy from the globally organized key message set in the Xth conversation state to the globally organized key message set in the first conversation state, the globally organized key message set in each conversation state is globally organized in turn, and the key message set after each round of global organization is obtained when the globally organized key message set from the Xth conversation state to the globally organized key message set in the first conversation state is globally organized. The key message set after each round of global organization and the globally organized key message set in the Xth conversation state are regarded as the obtained transition key message set.Alternatively, based on the global organization strategy from the third key message set in the first session state to the third key message set in the Xth session state, the third key message set in each session state is globally organized to obtain the key message set after each round of global organization. The third key message set in the first session state and the key message set after each round of global organization during the global organization operation from the third key message set in the first session state to the third key message set in the Xth session state are regarded as the obtained first transitional key message set. Based on the global organization strategy from the third key message set in the Xth session state to the third key message set in the first session state, the third key message set in each session state is globally organized to obtain the key message set after each round of global organization. The third key message set in the Xth session state and the key message set after each round of global organization during the global organization operation from the third key message set in the Xth session state to the third key message set in the first session state are regarded as the obtained second transitional key message set. The first transitional key message set and the second transitional key message set are regarded as the obtained transitional key message set. Thus, by setting various global organization strategies (fusion order), the third key message set in each session state is globally organized one by one. This enriches the methods for global organization of key message sets and expands the application environment for global organization of key message sets.
[0012] In one independently implementable embodiment, obtaining the fourth key message set by utilizing the transition key message set after each round of global reorganization includes: performing feature mining operations on the transition key message set after each round of global reorganization to obtain the fifth key message set corresponding to that transition key message set; wherein the temporal-level description of the fifth key message set corresponding to each transition key message set is the same; and concatenating the fifth key message sets corresponding to each transition key message set to obtain the fourth key message set. This improves the accuracy of mining business processing demand information included in community session logs that meet user demand mining conditions based on the fourth key message set.
[0013] In one independently implementable embodiment, the method further includes: pushing business services based on the business processing demand information.
[0014] On the other hand, this application also provides a smart community service system, including a processor and a memory; the processor and the memory are communicatively connected, and the processor is used to read a computer program from the memory and execute it to implement the above method.
[0015] The technical solutions provided by the embodiments of this application may include the following beneficial effects.
[0016] By updating the tag information of the first key message set, the second key message set corresponding to the first key message set in each session state is obtained. The temporal description of the second key message set in each session state is then updated, resulting in a statistical relationship between the temporal descriptions of the third key message set corresponding to the second key message set in each session state. This allows for the identification of business processing requirement information in community session logs that meet user requirement mining conditions, based on different temporal levels of the third key message set (which express different ways of business processing requirements through different temporal levels, thus obtaining key information about business processing requirements under different methods). This achieves the identification of business processing requirement information in community session logs that meet user requirement mining conditions based on the initial log recording format. Since the log recording format of community session logs that meet user requirement mining conditions does not need to be adjusted, the mining accuracy is maintained while reducing the resource overhead of community session log mining and analysis to a certain extent, thus improving the efficiency of mining business processing requirement information. Attached Figure Description
[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the specification, serve to explain the principles of this application.
[0018] Figure 1 This is a schematic diagram of the hardware structure of a smart community service system provided in an embodiment of this application.
[0019] Figure 2 This is a flowchart illustrating a user demand big data mining method for smart communities provided in an embodiment of this application.
[0020] Figure 3 This is a block diagram of a user demand big data mining device for smart communities provided in an embodiment of this application. Detailed Implementation
[0021] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.
[0022] It should be noted that the terms "first," "second," etc., in the specification, claims, and drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.
[0023] The methods and embodiments provided in this application can be executed in a smart community service system, computer equipment, or similar computing device. Taking the operation on a smart community service system as an example, Figure 1 This is a hardware structure block diagram of a smart community service system implemented in this application, which utilizes a user demand big data mining method for smart communities. (See also...) Figure 1 As shown, the smart community service system 10 may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. Optionally, the aforementioned smart community service system 10 may further include a transmission device 106 for communication functions. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the smart community service system 10 described above. For example, the smart community service system 10 may also include more than Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.
[0024] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the user demand big data mining method for smart communities in this embodiment of the application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the smart community service system 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.
[0025] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the smart community service system 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.
[0026] To facilitate understanding of the embodiments of this application, a method for mining user demand big data in smart communities, disclosed in the embodiments of this application, will first be described in detail below: Please see Figure 2 , Figure 2 This is a flowchart illustrating a user demand big data mining method for smart communities provided in this application embodiment. The method is applied to a smart community service system, and can be specifically described through the content recorded in steps 21-24 below.
[0027] Step 21: Identify key messages in community session logs that meet the user's needs mining criteria to obtain the first set of key messages in multiple session states.
[0028] In practice, community conversation logs that meet the user demand mining criteria can be understood as any community conversation log containing business processing demand information, such as community mutual assistance information, community medical information, and community housekeeping information. The user demand mining criteria can be understood as mining conditions set according to actual circumstances. The first key message set can be understood as a set of feature information from multiple conversation states obtained by identifying key messages in community conversation logs that meet the user demand mining criteria. For example, the key message set can be an expression or a feature map. Furthermore, the first key message set in the first conversation state is obtained by identifying key messages in community conversation logs that meet the user demand mining criteria. The first key message set in the second key message set of two related conversation states is obtained by identifying key messages in the first key message set of ...
[0029] In some embodiments, step 21, which records key message identification of community session logs that meet user demand mining conditions to obtain a first key message set under multiple session states, may specifically include the following description: Key message identification of community session logs that meet user demand mining conditions is performed by a first intelligent processing thread under multiple session states to obtain a first key message set exported by the first intelligent processing thread under each session state. In this embodiment, the intelligent processing thread can be understood as based on an intelligent convolutional neural network.
[0030] Step 22: Update the tag information of the first key message set to obtain the second key message set corresponding to the first key message set in each session state.
[0031] In practice, the tag information for the second key message set corresponding to the first key message set is the same across different session states. This tag information can be understood as identification information or parameter information.
[0032] In some embodiments, step 22, which records updating the tag information of the first key message set to obtain the second key message set corresponding to the first key message set in each session state, may specifically include the following: determining the first key message set with the fewest community interaction constraints (which can be understood as having the fewest restrictive conditions on community interaction matters) among the tag information corresponding to the first key message set in each session state, and updating other first key message sets except the first key message set with the fewest community interaction constraints to key message sets with the same tag information as the first key message set with the fewest community interaction constraints, and using the first key message set with the fewest community interaction constraints and the updated key message set with the same tag information as the first key message set with the fewest community interaction constraints as the second key message set; or, updating the first key message set in each session state to a key message set under specified tag information (in this embodiment, specified tag information can be understood as setting tag information), and using the key message set under the specified tag information as the second key message set.
[0033] By implementing the above, the first key message set in each session state will be updated to a minimum set of community interaction constraints. When mining business processing requirement information contained in community session logs that meet the user requirement mining conditions, the computational resource overhead of requirement mining can be reduced and the efficiency of requirement mining can be improved.
[0034] In some embodiments, step 22, which records updating the label information of the first key message set to obtain the second key message set corresponding to the first key message set in each session state, may further include the following: determining the thread label information of the second intelligent processing thread corresponding to the first intelligent processing thread in the session state based on the determined updated label information and the label information of the first key message set exported by the first intelligent processing thread in each session state; performing feature mining operation (which can be understood as convolution operation) on the first key message set exported by the first intelligent processing thread corresponding to the second intelligent processing thread in the session state, based on the second intelligent processing thread in each session state that includes the determined thread label information (network parameter information), to obtain the second key message set exported by the second intelligent processing thread in the session state.
[0035] It is understandable that by executing the above, by determining the thread tag information of the second intelligent processing thread in each session state, and by performing feature mining operations on the corresponding first key message set based on the second intelligent processing thread in each session state that contains the determined thread tag information, the community interaction item constraints in the tag information of the first key message set exported by the first intelligent processing thread in each session state are updated to a relatively smaller number of community interaction item constraints at a certain level. This reduces the computational load and improves the mining efficiency for business processing requirement information when mining community session logs that meet the user's needs.
[0036] Step 23: Update the tag information of the second key message set in each session state in sequence to obtain the third key message set corresponding to the second key message set in each session state.
[0037] In practice, the statistical results of the time-series description of the third key message set in each session state match the set indicators.
[0038] In some embodiments, the step 23 of sequentially updating the tag information of the second key message set in each session state to obtain the third key message set corresponding to the second key message set in each session state may specifically include the content recorded in steps 231-233 below.
[0039] Step 231: Based on the statistical results (proportions) of the temporal level description (which can be understood as the time dimension) between the first intelligent processing threads in different session states, and the temporal level description of the second key message set corresponding to the first intelligent processing thread in each session state, determine the temporal level description of the third key message set corresponding to the first intelligent processing thread in each session state.
[0040] Step 232: Determine the thread tag information of the third intelligent processing thread corresponding to the first intelligent processing thread in each session state by using the temporal description of the third key message set corresponding to the first intelligent processing thread in each session state, and the temporal description of the second key message set corresponding to the first intelligent processing thread in each session state.
[0041] Step 233: Based on the third intelligent processing thread in each session state that includes definite thread tag information, perform feature mining operation on the second key message set corresponding to the third intelligent processing thread in the session state to obtain the third key message set exported by the third intelligent processing thread in the session state.
[0042] When executing the content recorded in steps 231-233 above, by adjusting the temporal description of the second key message set corresponding to the first intelligent processing thread in each session state, the temporal description of the third key message set derived by the third intelligent processing thread in each session state is made to match the set statistical results (which is equivalent to adjusting the pattern of business processing requirement information included in the community session log that meets the user requirement mining conditions). Based on the third key message set after updating the temporal description, the business processing requirement information included in the community session log that meets the user requirement mining conditions can be mined more accurately, thus improving the accuracy of the mining.
[0043] Step 24: Using the third key message set, determine the business processing requirement information in the community session log that meets the user requirement mining conditions.
[0044] In practice, the business processing request information in community conversation logs that meet the user demand mining criteria can be understood as the actual business information that users need to process in the community conversation logs that meet the user demand mining criteria. Examples include: water, electricity, and gas repairs; access control and identity verification; community group buying; and optimization of intelligent anti-theft systems.
[0045] In some embodiments, the process of determining the business processing demand information in the community session log that meets the user demand mining conditions by utilizing the third key message set in step 24 may specifically include the content recorded in steps 241 and 242.
[0046] Step 241: Perform a global sorting operation on the third key message set corresponding to the second key message set in each session state (the global sorting operation can be understood as fusion processing, merging processing, etc.) to obtain the globally sorted fourth key message set.
[0047] Step 242: Using the fourth key message set, determine the business processing requirement information in the community session log that meets the user requirement mining conditions.
[0048] When implementing the content recorded in steps 241 and 242 above, after obtaining the third key message set corresponding to the second key message set in each session state, a global organization operation can be performed on the third key message set in each session state to obtain a globally organized fourth key message set. Based on the fourth key message set, business processing requirement information in the community session log that meets the user requirement mining conditions can be determined. In this way, by performing a global organization operation on the third key message set corresponding to the second key message set in each session state, the obtained fourth key message set can include key messages describing different third key message sets at the time sequence level. Therefore, when determining the business processing requirement information in the community session log that meets the user requirement mining conditions based on the fourth key message set, the accuracy of user requirement mining can be improved.
[0049] In some embodiments, the global sorting operation recorded in step 241 of the third key message set corresponding to the second key message set in each session state to obtain the globally sorted fourth key message set may specifically include the contents recorded in steps 2411 and 2412 below.
[0050] Step 2411: According to the pre-set global sorting strategy, the third key message set corresponding to the second key message set in each session state is sorted globally one by one to obtain the transition key message set after each round of global sorting.
[0051] Step 2412: Use the transition key message set after each round of global sorting to obtain the fourth key message set.
[0052] When executing steps 2411 and 2412, the pre-set global organization strategy can be understood as a preset fusion order. The transition key message set can be understood as an intermediate feature information set. It can be understood that a global organization strategy for the third key message set can be set, and the third key message sets corresponding to the second key message sets in each session state can be globally organized one by one according to the pre-set global organization strategy to obtain the transition key message set after each round of global organization. In this way, the third key message set is globally organized one by one to ensure the accuracy of the transition key message set. Based on this, the fourth key message set is accurately obtained using the transition key message set after each round of global organization.
[0053] In some embodiments, the third key message set corresponding to the second key message set in each session state is used as the third key message set in the first session state to the third key message set in the Xth session state, wherein the temporal description of the third key message set in the Xth session state is greater than the temporal description of the third key message set in the (X-1)th session state, and X is a positive integer greater than 1.
[0054] Based on the above, step 2411 records the global organization operation of the third key message set corresponding to the second key message set in each session state according to the pre-set global organization strategy, so as to obtain the transition key message set after each round of global organization. Specifically, it can include the following four embodiments.
[0055] First embodiment: Based on the global sorting strategy from the third key message set in the first session state to the third key message set in the Xth session state, the third key message set in each session state is sorted globally in sequence to obtain the key message set after each round of global sorting. The third key message set in the first session state and the key message set after each round of global sorting are regarded as the obtained transition key message set.
[0056] The second embodiment: Based on the global sorting strategy from the third key message set in the Xth session state to the third key message set in the first session state, the third key message set in each session state is sorted globally in sequence to obtain the key message set after each round of global sorting. The third key message set in the Xth session state and the key message set after each round of global sorting are regarded as the obtained transition key message set.
[0057] The third embodiment: Based on the global organization strategy from the third key message set in the first session state to the third key message set in the Xth session state, the third key message set in each session state is globally organized, sequentially obtaining the key message set after each round of global organization when organizing the third key message set from the first session state to the third key message set in the Xth session state. Feature mining is then performed sequentially on the third key message set in the first session state and the key message set after each round of global organization to obtain the globally organized key message set from the first session state to the Xth session state. In each session state... The label information of the globally organized key message set is the same as the label information of the key message set before the feature mining operation. According to the global organization strategy from the globally organized key message set in the Xth session state to the globally organized key message set in the first session state, the globally organized key message set in each session state is globally organized in sequence, and the key message set after each round of global organization is obtained when the globally organized key message set from the Xth session state to the globally organized key message set in the first session state is globally organized. The key message set after each round of global organization and the globally organized key message set in the Xth session state are regarded as the obtained transition key message set.
[0058] The fourth embodiment: Based on the global organization strategy from the third key message set in the first session state to the third key message set in the Xth session state, the third key message set in each session state is globally organized to obtain the key message set after each round of global organization. The third key message set in the first session state and the key message set after each round of global organization during the global organization operation from the third key message set in the first session state to the third key message set in the Xth session state are regarded as the obtained first transition key message set, and based on the third key message set in the Xth session state... The global sorting strategy for the message set to the third key message set in the first session state involves performing a global sorting operation on the third key message set in each session state to obtain the key message set after each round of global sorting. The third key message set in the Xth session state and the key message set after each round of global sorting operation when performing a global sorting operation from the third key message set in the Xth session state to the third key message set in the first session state are regarded as the obtained second transition key message set. The first transition key message set and the second transition key message set are regarded as the obtained transition key message set.
[0059] When implementing the above four embodiments, by setting multiple different global organization strategies (fusion order), the third key message set in each session state is globally organized one by one. This can enrich the global organization method of key message set and expand the application environment of global organization of key message set.
[0060] In some embodiments, the step 2412 of obtaining the fourth key message set by utilizing the transition key message set after each round of global reorganization may specifically include the following: performing feature mining operations on the transition key message set after each round of global reorganization to obtain the fifth key message set corresponding to the transition key message set; wherein, the temporal level description of the fifth key message set corresponding to each transition key message set is the same; and concatenating the fifth key message sets corresponding to each transition key message set to obtain the fourth key message set.
[0061] It is understandable that by performing feature mining operations on the transition key message set after each round of global sorting, and then splicing the fifth key message set obtained after feature mining operations, a fourth key message set is obtained. This fourth key message set includes both key messages with relatively concentrated demand information and key messages with relatively more detailed features. Furthermore, the fourth key message set also includes key messages described at different time-series levels. This improves the accuracy of mining when mining business processing demand information in community session logs that meet user demand mining conditions based on the fourth key message set.
[0062] Based on the above, services can also be pushed based on the business processing demand information, thereby improving the efficiency of business processing in smart communities.
[0063] In some independently implementable technical solutions, the above-mentioned service push based on the business processing demand information may include the following: obtaining a first predetermined number of annotated service demand information and a second predetermined number of unannotated service demand information; importing the first predetermined number of annotated service demand information and the second predetermined number of unannotated service demand information into a service demand information analysis network; binding service item identifier annotations of the included service items to the first predetermined number of annotated service demand information respectively; the service items bound to the first predetermined number of annotated service demand information and the service items bound to the second predetermined number of unannotated service demand information belong to the same service item category; determining the first identifier analysis result of the service items included in each annotated service demand information in the service demand information analysis network, and obtaining the associated service demand information of each annotated service demand information from the service demand information set; the service demand information set includes the first predetermined number of annotated service demand information. The service requirement information and the second set number of unannotated service requirement information; the associated service requirement information of each annotated service requirement information is not bound to the service item identifier annotation bound to the annotated service requirement information; a correlation difference evaluation is determined based on the correlation degree between each annotated service requirement information and its associated service requirement information, and an analysis evaluation is determined based on the first identifier analysis result and the bound service item identifier annotation of each annotated service requirement information; the network variables of the service requirement information analysis network are optimized based on the correlation difference evaluation and the analysis evaluation to obtain the target service requirement information analysis network; the service items belonging to the service item category in the business processing requirement information are identified based on the target service requirement information analysis network to obtain the pending service item identifier corresponding to the business processing requirement information, and a service push strategy is determined based on the pending service item identifier, and the business service is pushed according to the pending service push strategy.
[0064] Based on the above, the service demand information analysis network can be trained and optimized based on correlation difference evaluation and analysis evaluation, thereby ensuring the accuracy of identification of service items belonging to the service item category in the business processing demand information. This allows for accurate determination of the service push strategy based on the obtained service item identifiers to be processed, thereby improving the quality of business service push.
[0065] In some independently implementable technical solutions, obtaining the associated service demand information for each annotated service demand information from the service demand information set includes: generating service demand information expressions for each annotated service demand information and for each unannotated service demand information in the service demand information analysis network; generating a correlation degree distribution based on the service demand information expressions for each annotated service demand information and for each unannotated service demand information; obtaining the service demand information correlation degree between each annotated service demand information and the service demand information in the service demand information set from the correlation degree distribution; and determining the associated service demand information for each annotated service demand information from the service demand information set based on the service demand information correlation degree between each annotated service demand information and the service demand information in the service demand information set. In this way, the associated service demand information can be completely determined.
[0066] In summary, by updating the tag information of the first key message set, the second key message set corresponding to the first key message set in each session state is obtained. Furthermore, the temporal description of the second key message set in each session state is updated, resulting in a statistical relationship between the temporal descriptions of the third key message set corresponding to the second key message set in each session state. Therefore, based on the different temporal levels of the third key message set (which express different ways of business processing needs through different temporal levels, thus obtaining key information on business processing needs under different methods), business processing need information in community session logs that meet the user need mining conditions can be identified. This achieves the identification of business processing need information in community session logs that meet the user need mining conditions based on the initial log recording format. Since the log recording format of community session logs that meet the user need mining conditions does not need to be adjusted, the mining accuracy is maintained while reducing the resource overhead of community session log mining and analysis to a certain extent, thus improving the mining efficiency for business processing need information.
[0067] Based on the above, please combine Figure 3 This application also provides a block diagram of a user demand big data mining device 30 applied to smart communities, the device comprising the following functional modules: The message recognition module 31 is used to identify key messages in community session logs that meet the user's needs for mining, so as to obtain a first key message set in multiple session states; and to obtain a second key message set corresponding to the first key message set in each session state by updating the tag information of the first key message set; wherein the tag information of the second key message set corresponding to the first key message set in different session states is the same. The information determination module 32 is used to sequentially update the tag information of the second key message set in each session state to obtain the third key message set corresponding to the second key message set in each session state, wherein the statistical results of the time-series description of the third key message set in each session state match the set indicators; using the third key message set, the business processing demand information in the community session log that meets the user demand mining conditions is determined.
[0068] Furthermore, a readable storage medium is provided on which a program is stored, which, when executed by a processor, implements the above-described method.
[0069] It should be understood that this application is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.
Claims
1. A method for mining user demand big data applied to smart communities, characterized in that, Applied to a smart community service system, the method includes at least: Key messages are identified in community session logs that meet the user needs mining criteria to obtain a first key message set in multiple session states; the first key message set is updated with tag information to obtain a second key message set corresponding to the first key message set in each session state; wherein the tag information of the second key message set corresponding to the first key message set in different session states is the same. The tag information of the second key message set in each session state is updated sequentially to obtain the third key message set corresponding to the second key message set in each session state. The statistical results of the time-series description of the third key message set in each session state match the set indicators. Using the third key message set, the business processing demand information in the community session log that meets the user demand mining conditions is determined.
2. The method as described in claim 1, characterized in that, The process of identifying key messages in community session logs that meet user needs and mining criteria to obtain a first set of key messages across multiple session states includes: The first intelligent processing thread in multiple session states identifies key messages in community session logs that meet the user's mining criteria, so as to obtain the first key message set exported by the first intelligent processing thread in each session state. The step of updating the tag information of the first key message set to obtain the second key message set corresponding to the first key message set in each session state includes: Based on the determined updated tag information and the tag information of the first key message set exported by the first intelligent processing thread in each session state, the thread tag information of the second intelligent processing thread corresponding to the first intelligent processing thread in this session state is determined. Based on the second intelligent processing thread in each session state, which includes defined thread label information, feature mining is performed on the first key message set exported by the first intelligent processing thread corresponding to the second intelligent processing thread in that session state to obtain the second key message set exported by the second intelligent processing thread in that session state.
3. The method as described in claim 1, characterized in that, The step of updating the tag information of the first key message set to obtain the second key message set corresponding to the first key message set in each session state includes: Determine the first key message set with the fewest community interaction constraints among the tag information corresponding to the first key message set in each session state, and update the other first key message sets except the first key message set with the fewest community interaction constraints to key message sets with the same tag information as the first key message set with the fewest community interaction constraints. The first key message set with the fewest community interaction constraints and the updated key message set with the same tag information as the first key message set with the fewest community interaction constraints are used as the second key message set. Alternatively, the first key message set in each session state can be updated to the key message set under the specified tag information, and the key message set under the specified tag information can be used as the second key message set.
4. The method according to any one of claims 1 to 3, characterized in that, The process of identifying key messages in community session logs that meet user needs and mining criteria to obtain a first set of key messages across multiple session states includes: The first intelligent processing thread in multiple session states identifies key messages in community session logs that meet the user's mining criteria, so as to obtain the first key message set exported by the first intelligent processing thread in each session state. The step of sequentially updating the tag information of the second key message set in each session state to obtain the third key message set corresponding to the second key message set in each session state includes: Based on the statistical results of the temporal description of the first intelligent processing thread under different session states, and the temporal description of the second key message set corresponding to the first intelligent processing thread under each session state, the temporal description of the third key message set corresponding to the first intelligent processing thread under each session state is determined. By using the temporal description of the third key message set corresponding to the first intelligent processing thread in each session state, and the temporal description of the second key message set corresponding to the first intelligent processing thread in each session state, the thread tag information of the third intelligent processing thread corresponding to the first intelligent processing thread in that session state is determined. Based on the third intelligent processing thread in each session state, which includes defined thread label information, feature mining is performed on the second key message set corresponding to the third intelligent processing thread in that session state to obtain the third key message set exported by the third intelligent processing thread in that session state.
5. The method as described in claim 1, characterized in that, The step of using the third key message set to determine the business processing demand information in the community session logs that meet the user demand mining conditions includes: Perform a global sorting operation on the third key message set corresponding to the second key message set in each session state to obtain a globally sorted fourth key message set. Using the fourth key message set, the business processing requirement information in the community session logs that meet the user requirement mining conditions is determined.
6. The method as described in claim 5, characterized in that, The third key message set corresponding to the second key message set in each session state is globally processed to obtain a globally processed fourth key message set, including: According to the pre-set global sorting strategy, the third key message set corresponding to the second key message set in each session state is sorted out one by one to obtain the transition key message set after each round of global sorting. The fourth key message set is obtained by utilizing the transition key message set after each round of global sorting.
7. The method as described in claim 6, characterized in that, The third key message set corresponding to the second key message set in each session state is used as the third key message set from the first session state to the Xth session state, where the temporal description of the third key message set in the Xth session state is greater than that of the third key message set in the (X-1)th session state, and X is a positive integer greater than 1. Then, according to the pre-set global sorting strategy, the third key message set corresponding to the second key message set in each session state is globally sorted one by one to obtain the transition key message set after each round of global sorting, including: Based on the global organization strategy from the third key message set in the first session state to the third key message set in the Xth session state, the third key message set in each session state is globally organized in sequence to obtain the key message set after each round of global organization. The third key message set in the first session state and the key message set after each round of global organization are regarded as the obtained transition key message set. Alternatively, based on the global organization strategy from the third key message set in the Xth session state to the third key message set in the first session state, the third key message set in each session state is globally organized sequentially to obtain the key message set after each round of global organization. The third key message set in the Xth session state and the key message set after each round of global organization are regarded as the obtained transition key message set. Alternatively, based on the global organization strategy from the third key message set in the first session state to the third key message set in the Xth session state, the third key message set in each session state is globally organized to obtain the key message set after each round of global organization during the global organization operation from the third key message set in the first session state to the third key message set after each round of global organization. Feature mining operation is then performed on the third key message set in the first session state and the key message set after each round of global organization to obtain the globally organized key message set from the first session state to the globally organized key message set in the Xth session state. The label information of the globally organized key message set in each session state is the same as the label information of the key message set before the feature mining operation. Based on the global organization strategy from the global organization key message set in the Xth session state to the global organization key message set in the first session state, the global organization key message set in each session state is sequentially organized, and the key message set after each round of global organization is obtained sequentially when the global organization key message set from the Xth session state to the first session state is organized. The key message set after each round of global organization and the global organization key message set in the Xth session state are regarded as the obtained transition key message set. Alternatively, based on the global organization strategy from the third key message set in the first session state to the third key message set in the Xth session state, the third key message set in each session state is globally organized to obtain the key message set after each round of global organization. The third key message set in the first session state and the key message set after each round of global organization during the global organization operation from the third key message set in the first session state to the third key message set in the Xth session state are regarded as the obtained first transition key message set. Based on the global organization strategy from the third key message set in the Xth session state to the third key message set in the first session state, the third key message set in each session state is globally organized to obtain the key message set after each round of global organization. The third key message set in the Xth session state and the key message set after each round of global organization during the global organization operation from the third key message set in the Xth session state to the third key message set in the first session state are regarded as the obtained second transition key message set. The first transition key message set and the second transition key message set are regarded as the obtained transition key message set.
8. The method as described in claim 6 or 7, characterized in that, The process of obtaining the fourth key message set by utilizing the transition key message set after each round of global reorganization includes: Feature mining is performed on each round of global sorting of the transition key message set to obtain the fifth key message set corresponding to that transition key message set; wherein, the temporal level description of the fifth key message set corresponding to each transition key message set is the same; The fifth key message set corresponding to each transition key message set is concatenated to obtain the fourth key message set.
9. The method as described in claim 1, characterized in that, The method also includes: pushing business services based on the business processing demand information.
10. A smart community service system, characterized in that, Including processor and memory; The processor and the memory are communicatively connected. The processor is used to read a computer program from the memory and execute it to implement the method described in any one of claims 1-8.