Method and device for generating management decision, equipment and storage medium

Abstract

The invention discloses a method and device for generating a management decision, equipment and a storage medium, and the method comprises the steps: collecting related data of a property management system, and obtaining a long text data set; cleaning and preprocessing the long text data set by using an RAG technology to obtain preprocessed property management data; constructing a property manager agent; performing reasoning analysis on the property management data through the property manager agent to generate a management decision; wherein the property manager agent construction comprises the following steps: setting a task; calling a large language model to process the task; executing an instruction output by the large language model to obtain a historical processing result; wherein the instruction comprises an analysis instruction, a selection execution tool and an execution tool sequence; and storing the historical processing result by utilizing a cache processing mechanism. The property manager intelligent agent not only can improve the efficiency and quality of property management, but also can improve the life experience of residents.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence and property information management technology, and in particular to a method, device, equipment and storage medium for generating management decisions. Background Art

[0002] In recent years, with the advancement of science and technology and the improvement of residents' living standards, property management is undergoing a transformation from traditional manual services to intelligent management. In this process, the internal management of property companies has also undergone earth-shaking changes. In the past, property companies relied on a large number of manual workers for daily management, which resulted in low work efficiency and uneven service quality. As a property manager, facing the growing amount of information, you need to process a large amount of redundant and complex data every day, which makes it difficult to make effective decisions quickly. Therefore, it is often difficult for property managers to quickly extract the key information of the community on the day from the complex information, resulting in unreasonable work arrangements, which in turn affects the quality of service. Summary of the invention

[0003] The embodiments of the present invention provide a method, apparatus, device and storage medium for generating management decisions, aiming to solve the problems of low information processing efficiency and uneven service quality in traditional property management.

[0004] In a first aspect, an embodiment of the present invention provides a method for generating management decisions based on a property manager agent, comprising: Collect relevant data from the property management system and obtain long text data sets; Using RAG technology to clean and preprocess the long text data set to obtain preprocessed property management data; Build a property manager agent; The property manager agent performs reasoning analysis on the property management data to generate management decisions; The steps of constructing the property manager agent include: Setting tasks, wherein the tasks include defining objectives, setting output constraints and resource lists; Calling a large language model to process the task; Executing the instructions output by the large language model to obtain historical processing results; wherein the instructions include parsing instructions, selecting execution tools and execution tool sequences; A cache processing mechanism is used to store the historical processing results.

[0005] In a second aspect, an embodiment of the present invention provides a device for generating management decisions based on a property manager agent, comprising: A collection unit is used to collect relevant data of the property management system and obtain a long text data set; A preprocessing unit, used for cleaning and preprocessing the long text data set using RAG technology to obtain preprocessed property management data; Construction unit, used to build property manager agent; A reasoning and analysis unit, used to perform reasoning and analysis on the property management data through the property manager agent to generate management decisions; The steps of constructing the property manager agent include: Setting tasks, wherein the tasks include defining objectives, setting output constraints and resource lists; Calling a large language model to process the task; Executing the instructions output by the large language model to obtain historical processing results; wherein the instructions include parsing instructions, selecting execution tools and execution tool sequences; A cache processing mechanism is used to store the historical processing results.

[0006] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for generating management decisions based on a property manager agent as described in the first aspect is implemented.

[0007] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to execute the method for generating management decisions based on a property manager agent as described in the first aspect.

[0008] The embodiment of the present invention provides a method, device, equipment and storage medium for generating management decisions. The present invention collects relevant data of a property management system to obtain a large amount of long text data sets, uses RAG (Retrieval-Augmented Generation) technology to clean and preprocess the long text data sets to obtain property management data, then constructs a property manager intelligent body, and uses the property manager intelligent body to perform reasoning and analysis on the property management data to generate management decisions; in terms of constructing the property manager intelligent body, the present invention sets specific tasks and calls a large language model to process these tasks. The large language model can understand and process complex instructions, including parsing instructions, selecting execution tools and execution tool sequences, etc., and obtains historical processing results by executing the output instructions of the large language model, and finally uses a cache processing mechanism to store historical processing results. The embodiment of the present invention significantly improves the efficiency and quality of property management, optimizes the accuracy and timeliness of decision-making, improves service efficiency, reduces labor costs, and reduces workload; at the same time, a more humane service model is realized, and the property management industry is promoted to develop in the direction of intelligence and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0010] Figure 1 A flow chart of a method for generating management decisions based on a property manager agent provided by an embodiment of the present invention; Figure 2 A schematic diagram of a sub-process of a method for generating management decisions based on a property manager agent provided by an embodiment of the present invention; Figure 3 A schematic diagram of another sub-process of a method for generating management decisions based on a property manager agent provided by an embodiment of the present invention; Figure 4 A schematic diagram of the RAG technical architecture provided by an embodiment of the present invention; Figure 5 A schematic diagram of another sub-process of a method for generating management decisions based on a property manager agent provided by an embodiment of the present invention; Figure 6 A schematic diagram of another sub-process of a method for generating management decisions based on a property manager agent provided by an embodiment of the present invention; Figure 7 A schematic diagram of another sub-process of a method for generating management decisions based on a property manager agent provided by an embodiment of the present invention; Figure 8 A schematic block diagram of the overall process of generating management decisions based on a property manager agent provided by an embodiment of the present invention; Fig. 9 A schematic diagram of a device for generating management decisions based on a property manager agent provided by an embodiment of the present invention; Fig.10 A schematic diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0011] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0012] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0013] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0014] It should be further understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0015] See also Figure 1 , Figure 1 A flow chart of a method for generating management decisions based on a property manager agent provided by an embodiment of the present invention, the method comprising steps S101 to S104: S101, collect relevant data of the property management system to obtain a long text data set; In this embodiment, in order to realize intelligent property management, we first conduct in-depth research based on property domain knowledge, industry regulations and system documents to build a comprehensive long text data set. This long text data set revolves around the business situation of the property project, including key indicators such as operating data, customer satisfaction, work order operations, and emergencies, aiming to provide a data basis for deep reasoning and analysis. On this basis, this embodiment summarizes the matters that property managers need to pay attention to and deal with on a daily basis into ten categories: building area and number of service households, weather conditions of the day, yesterday's operating data, yesterday's key arrears households, yesterday's emergency summary, yesterday's customer complaints, follow-up of long-term unclosed complaints, yesterday's work order completion status, customer satisfaction surveys, and today's work plan. In order to collect relevant data in the above categories, the internal management system resources of the property are integrated and collected to obtain long text data sets, realizing comprehensive data aggregation.

[0016] S102, using RAG technology to clean and preprocess the long text data set to obtain preprocessed property management data; In this embodiment, the data extracted from the long text data set obtained in step S101 often have problems such as duplication, inaccuracy, invalidity and errors. In view of this situation, this embodiment performs strict data cleaning on the long text data set, including removing useless, duplicate and erroneous data to ensure the quality and accuracy of the data. At the same time, considering the dependence of the large language model used in the following steps on property domain knowledge, this embodiment uses RAG (Retrieval-Augmented Generation) technology for data cleaning and preprocessing.

[0017] RAG technology can retrieve information related to property management from a large amount of data in long text data sets, ensuring that the pre-processed property management data is more focused and accurate. By combining professional knowledge in the property field, RAG technology can better understand the data content, provide industry-specific context for data preprocessing, and provide high-quality input for subsequent large language model processing.

[0018] In one embodiment, if Figure 2 As shown, the step S102 includes steps S201 to S204: S201. Convert the property-related queries and documents in the long text dataset into vector representations according to the following formula: , where Q and C represent property-related queries and documents, respectively. is an n-dimensional real vector space, represents the vector representation of each document in the document collection, and q represents the query vector; S202, calculating the vector search between the query vector and the document vector according to the following formula: ,in, Represents the query vector q and the document vector The vector dot product of Represents the Euclidean norm of a vector; S203, concatenate and merge the retrieved relevant documents and property-related queries according to the following formula to form context information: ,in, Indicates related documents; S204: Input the context information into the generation model to generate a text set according to the following formula: , where G is the generative model and m is the dimension of the context vector.

[0019] In this embodiment, in step S201, the architecture of the RAG technology consists of two core parts: a retrieval module (Retriever) and a generator module (Generator). In the retrieval module, the property-related query Q and document C need to be converted into vector representations. is an embedding function, where Q and C represent property-related queries and documents, respectively. is an n-dimensional real vector space, then for each document in the document collection , the vector representations of query Q and document C are and .

[0020] In step S202, vector search is performed by calculating the query vector q and the document vector The calculation of cosine similarity not only takes into account the directional relationship between vectors, but also has high computational efficiency.

[0021] Specifically, the dot product of the query vector and each document vector in the dataset is calculated, and then the Euclidean norm of the query vector and each document vector is calculated respectively. The dot product result is divided by the norm product of the query vector and the document vector to obtain the cosine similarity of each document vector to the query vector. The documents are sorted according to the size of the cosine similarity, and the document with the highest similarity is considered to be the most relevant to the query.

[0022] Through this vector search method, the present embodiment is able to quickly and accurately retrieve the most relevant information to the query from a large number of documents, which is essential for providing key information support for property manager agents when making management decisions.

[0023] In step S203, the generation module (generation model) has learned the grammar, semantics and contextual relationships of natural language through a large amount of text data during the training phase. This pre-training can be based on a general corpus or fine-tuned for a specific field (such as property management). The generation model needs to receive a context information as input, which is composed of two parts, namely the retrieved relevant documents and the property-related query.

[0024] Furthermore, the retrieved relevant documents and property-related queries are spliced ​​and merged to form a new and more comprehensive context information. Specifically, the extracted key information is spliced ​​with the property-related query text to form a continuous text sequence, ensuring that the spliced ​​text is coherent in format and easy for the generation model to understand and process.

[0025] In step S204, based on step S203, the generation model generates a coherent, accurate and information-rich text set T according to the input context information, and each text in the text set T is an answer to the property-related query from a different angle. Specifically, during the generation process, the generation model will continuously refer to the input context information to ensure that the text in the generated text set T is consistent with the query and related document content.

[0026] In summary, in steps S201 to S204, in the property data integration and preprocessing stage, that is, the preprocessing stage of the long text data set, not only efficient information retrieval is achieved, but also the quality of text generation is ensured, providing a solid data foundation for the property manager intelligent agent.

[0027] In one embodiment, if Figure 3 As shown, the step S102 also includes steps S301 to S304: S301, using the preset rules generated by the RAG technology to perform hash deduplication, correction and completion operations on the long text data set; S302: Calculate the accuracy of the data in the long text data set according to the following formula: ,in, represents a dataset, Representation dataset A single data item in Representation dataset The correct data item in S303, calculating the completeness rate of the data in the long text data set according to the following formula: , where n represents the data set The total number of data fields in represents the jth field, It is an indicator function that returns 0 or 1 according to the value of the input data field; S304: Calculate the consistency of the data in the long text data set according to the following formula: , where z represents the data set Data items that conform to the specifications.

[0028] In this embodiment, the preset rules generated by RAG technology are used , perform deduplication, correction and completion operations on the data extracted from the long text dataset. Specifically, the hash function is used , to achieve data deduplication, where H is a hash function and X is a data item, ensuring the uniqueness of each data item. After data cleaning, data quality assessment is performed to ensure the validity and reliability of the data. Accuracy, completeness and consistency are key indicators for evaluating data quality. Through the above process, this embodiment effectively solves the quality problems encountered in the data integration process and ensures the high accuracy and integrity of the data. The application of RAG technology not only improves the speed of data processing, but also reduces the need for manual intervention, providing efficient and reliable data support for the intelligent property manager system, thereby laying a solid foundation for the intelligent and refined property management.

[0029] In summary, if Figure 4 As shown in the figure, the RAG technology architecture is divided into the following stages: data source, data preprocessing, data storage, vectorization and retrieval, prompt enhancement and model response. First, the data source is divided into the following modules: business: revenue, profit, collection rate data; customer: satisfaction, complaint data; emergency: fire, extreme weather data; work order operation data; equipment operation data; project data, historical experience and system documents, industry regulations. The original data (long text data set) is collected from different channels of the property management system. The diverse data sources provide a comprehensive information basis for data collection. Then, the above data is preprocessed to form vector data, and then the preprocessed vector data is stored for subsequent query and matching. Then, the user inputs in a certain way, the user input content is vectorized, and then retrieved in the vector database. Then, the relevant information retrieved is used to enhance the prompt (Prompt) to improve the generated effect. Finally, according to the enhanced prompt, the model generates the final response or result. The whole process reflects how to enhance the generation ability of the model by combining business data and knowledge base, and using RAG technology to provide more accurate and useful responses.

[0030] S103, constructing a property manager intelligent agent; In one embodiment, if Figure 5 As shown, the step S103 includes steps S401 to S404: S401, setting a task, wherein the task includes defining a goal, setting output limits and a resource list; S402, calling a large language model to process the task; S403, executing the instructions output by the large language model to obtain historical processing results; wherein the instructions include parsing instructions, selecting execution tools and execution tool sequences; S404: Utilize a cache processing mechanism to store the historical processing results.

[0031] In this embodiment, four core links are involved in the construction of the property manager agent: task setting, large language model calling, command execution and cache processing.

[0032] In step S401, the user sets tasks for the intelligent property manager in the form of role description, which can be input in a configuration file or interactively. The user defines the target tasks of the intelligent property manager and optimizes and integrates the tasks, including setting output restrictions and resource lists, and forming a standardized large language model interface data format to ensure the parsability of the call results. The task setting requires the output to contain two parts of data: process and command. The process is used to describe the progress of the goal completion and the self-reflection of the intelligent agent, while the command contains the commands and their parameters required to execute the task. After the task setting is completed, the large language model M is called in step S402 to process the task set by the user.

[0033] Further, in step S403, when executing a task, the property manager agent first parses the instructions output by the large language model M. These instructions will tell the property manager agent what to execute, thereby obtaining historical processing results, and then selects the optimal tool sequence from the tool set based on the reinforcement learning algorithm to maximize the expected return. Specifically, this embodiment allows users to customize the command tool set , each tool The corresponding specific capabilities of the intelligent agent, such as text analysis, web search, and knowledge integration, etc. When performing a task, the intelligent agent first parses the instructions output by the large model, and then selects the optimal tool sequence O from the tool set based on the reinforcement learning algorithm to maximize the expected return.

[0034] To achieve this goal, the agent uses the evaluation function To make a decision, this function takes into account the current state Potential Actions And the parameter set A of the reinforcement learning model. The selection process can be formalized as: After the selected tool sequence O is executed, the result is It is the key output of the agent to complete the task. This execution process can be expressed by the following formula: .

[0035] Furthermore, in step S404, in order to ensure the continuity of the process and the optimization of the results, this embodiment adopts a cache processing mechanism to save the historical processing results. The present invention will feed this historical information h into the model together with the current input to ensure that the model can make more optimized decisions based on a more comprehensive set of information. First, through the embedding function, the current input x is fused with the historical information h extracted from the cache to generate a new input vector Then, this new input vector is fed into the generating function G to obtain the optimized output vector : .

[0036] in, It is a new input vector combined with historical information, which contains the data of the current step and relevant information of historical processing. It is an optimized output vector, which is based on a new input vector that incorporates historical information and aims to provide more accurate and task-oriented predictions or decisions.

[0037] To sum up, through the above four core links, this embodiment constructs an efficient and flexible property manager intelligent body, which can automatically execute commands according to the task objectives set by the user, and continuously optimize the processing flow to improve the intelligence level of property management.

[0038] In one embodiment, if Figure 6 As shown, the step S402 includes steps S501 to S503: S501, encode the input sequence according to the following formula to obtain an input code: ,in, Embedding represents embedding, Positional Encoding represents the positional encoding, and t represents the input sequence; S502: linearly transform the input code according to the following formula to obtain a query matrix, a key matrix and a value matrix: ,in , , is a learnable weight matrix, Q represents the query matrix, K represents the key matrix, and V represents the value matrix; S503, calculating the self-attention weight through the query matrix, key matrix and value matrix, and obtaining the output result of the self-attention layer according to the following formula: ,in, is the vector dimension of the key matrix K, is the transposed matrix of the key matrix K.

[0039] In this embodiment, in step S501, the large language model M first needs to encode the input sequence, which is usually achieved through an embedding layer, and then positional encoding is added to retain the position information of the words in the sequence.

[0040] In step S502, the large language model M is processed. That is, the input code is linearly transformed. The core of the large language model M is the self-attention mechanism and the feed-forward neural network. Specifically, first, the correlation between each element in the input sequence is calculated through the self-attention layer. Specifically, the embedding representation of the input sequence (Input encoding) is linearly transformed into three matrices: query, key, and value, represented as , and Then, in step S503, the self-attention weight is calculated by Function and Value Matrix To obtain the output of the self-attention layer.

[0041] S601, using a multi-head attention mechanism to send the input code to multiple self-attention layers respectively, then splicing the output results of the self-attention layers, and performing a linear transformation through a weight matrix, and obtaining the linear transformation result of the multi-head attention mechanism according to the following formula: ,in, represents the weight matrix, Represents the i-th output result of the self-attention layer; S602, further performing nonlinear transformation on the linear transformation result of the multi-head attention mechanism through a feedforward neural network, and the nonlinear transformation result is expressed as follows: ,in, and is the weight matrix in the feedforward neural network, and is the bias term in the feedforward neural network; S603: Integrate the features of different layers of the large language model according to the following formula: ; S604, generate probability distribution through linear layer and softmax function, and obtain output code according to the following formula: ,in, is the weight matrix of the output embedding layer, is the predicted next word.

[0042] In this embodiment, in step S601, in order to enhance the expressiveness of the large language model M, a multi-head attention mechanism is used to feed the input sequence into multiple self-attention layers respectively, and then concatenate the outputs of these layers and pass an additional weight matrix In step S602, after the self-attention layer, the output of each position is further nonlinearly transformed through a feedforward neural network. is the output of step S601 (i.e., the linear transformation result of the multi-head attention mechanism). In step S603, in order to improve the stability of model training, layer normalization is applied after the self-attention layer and the feedforward neural network, and residual connections are added to ensure that the large language model M can effectively integrate features from different layers.

[0043] Further, in step S604, by passing the output of the linear layer to Function, the large language model M generates a probability distribution for each word in the vocabulary. This probability distribution represents the probability that each word in the vocabulary is the next word in the current context.

[0044] The final output encoding is a vector where each element corresponds to the probability of a word in the vocabulary. This vector can be used as a tool to predict the next word. Typically, the word with the highest probability will be selected as the predicted output of the large language model M.

[0045] The embodiment of the present invention also proposes an intelligent agent planning algorithm based on reverse thinking chain, which aims to optimize the traditional forward thinking strategy. Towards the target state The strategy of advancement (denoted as ), however, this approach is often inefficient, especially when there is a decision bottleneck. Forward planning can easily lead to an exponential growth in path exploration and increase computational redundancy. This embodiment uses a reverse planning method, denoted as ,The core idea is to reduce the search space by constraining the intermediate steps through the target state.

[0046] Define the planned path P as a sequence of state transitions, that is, The reverse planning algorithm starts from Start from, and reverse to , the reversibility of state transition is expressed by the following formula: ,in represents the inverse operation of the state transition, yes In addition, a "self-verification" mechanism is introduced, denoted as , to ensure the validity of the planned path. The mechanism verifies each state transition through the following logic: ,in, is a verification function that asserts that the state transition is valid. This process involves checking each step in P Logical reasoning and verification are performed to ensure that each operation can correctly achieve state transition, thereby improving the overall accuracy and efficiency of planning. In this way, the planning ability of the property manager agent in complex environments is significantly improved.

[0047] S104, performing reasoning analysis on the property management data by the property manager agent to generate a management decision; In this embodiment, the property manager agent conducts in-depth analysis of the collected property management data, including but not limited to resident complaints, facility maintenance records, security monitoring data, etc., to identify problems and potential trends. Based on the results of the reasoning analysis, the property manager agent will generate a series of management decision suggestions, which may involve resource allocation, service improvement, emergency plans, etc., aiming to optimize the property management process and improve service quality.

[0048] In one embodiment, if Figure 8 As shown, the step S104 includes steps S701 to S704: S701, output the inference result according to the following formula: , where R represents the preset rules generated by RAG technology, and a represents the analysis algorithm; S702. Based on the inference result, the property behavior is divided into standardized behavior and non-standardized behavior according to the following formula: , where J is the classification function, B is the standardized behavior, and N is the unstandardized behavior; S703, automatically generating a work order through the property manager agent according to the standardized behavior, and assigning the task through integration with the work order system; S704: For non-standardized behaviors, the property manager agent sends daily reports to management personnel regularly according to a preset schedule.

[0049] In this embodiment, the property manager agent conducts an in-depth analysis of the project business status, obtains reasoning results, and divides the reasoning results into two categories: standardized behavior and non-standardized behavior.

[0050] For standardized behaviors, the property manager agent has direct decision-making capabilities and can seamlessly connect to the work order system. The property manager agent automatically dispatches work orders to responsible personnel based on the preset decision logic to ensure that various standardized operations are executed in a timely and accurate manner.

[0051] For non-standardized behaviors, the property manager agent adopts a notification and reminder mechanism. The present invention designs a timed push function. The property manager agent regularly sends detailed situation descriptions and necessary reminder information to relevant personnel according to the specific circumstances of non-standardized actions, ensuring that key information can be paid attention to and processed in a timely manner.

[0052] In one example, the method for generating management decisions based on a property manager agent further includes steps S801 to S803: S801. Construct a dictionary including key output results and corresponding timestamps, wherein the dictionary is expressed as follows: ,in, Indicates the output result, Indicates the corresponding timestamp; S802: According to the current target and historical information, the relevance of the behavior information is determined according to the following formula: ,in, Indicates the current target. Represents historical information. is the filter function; In this embodiment, in order to ensure the relevance of the behavior information, a screening function is first defined , then based on the current target and historical information Select the relevant behavioral information.

[0053] S803: Update the cache of the dictionary, add the new key output result to the dictionary, and remove irrelevant behavior information according to the following formula: ,in, is the cache update function, Indicates new key output results.

[0054] In this implementation, in each update iteration, the cache update function Output the new key Add to , while removing behavioral information that is no longer relevant.

[0055] In summary, in steps S801 to S803, when processing a large amount of data, directly inputting it into the large language model M will inevitably lead to increased token (the smallest unit of text) consumption and increased costs. In addition, the input of irrelevant information may reduce the accuracy of the output of the large language model M. In order to improve the efficiency of tokens use and ensure the reliability of the large language model M, this embodiment proposes a cache management strategy.

[0056] The strategy involves aggregating the output results of each step into a tool message dictionary tools_message(tm), which acts as an information intermediary and caches only progress information directly related to the current planned goal achievement. This approach significantly reduces the consumption of tokens by the large language model M when the property manager agent performs complex tasks, and can effectively review and utilize historical action data without missing important steps due to information overload, while maintaining the relevance of information and the accuracy of model output. It not only reduces operating costs, but also improves the efficiency and effectiveness of intelligent property managers when dealing with complex business scenarios.

[0057] like Fig. 9 As shown, an embodiment of the present invention further provides a device 900 for generating management decisions based on a property manager agent, including a collection unit 901, a preprocessing unit 902, a construction unit 903, and a reasoning and analysis unit 904.

[0058] The collection unit 901 is used to collect relevant data of the property management system and obtain a long text data set; A preprocessing unit 902 is used to clean and preprocess the long text data set using RAG technology to obtain preprocessed property management data; A construction unit 903, used to construct a property manager agent; The reasoning and analysis unit 904 is used to perform reasoning and analysis on the property management data through the property manager agent to generate management decisions; Wherein, the construction unit 903 includes: Setting tasks, wherein the tasks include defining objectives, setting output constraints and resource lists; Calling a large language model to process the task; Executing the instructions output by the large language model to obtain historical processing results; wherein the instructions include parsing instructions, selecting execution tools and execution tool sequences; A cache processing mechanism is used to store the historical processing results.

[0059] The device for generating management decisions based on the property manager agent provided by the embodiment of the present invention acquires a large amount of long text data sets by collecting relevant data of the property management system, cleans and preprocesses the long text data sets by using the RAG (Retrieval-Augmented Generation) technology, obtains property management data, and then constructs a property manager agent. The property management data is analyzed by the property manager agent to generate management decisions. In terms of constructing the property manager agent, the present invention sets specific tasks and calls a large language model to process these tasks. The large language model can understand and process complex instructions, including parsing instructions, selecting execution tools and execution tool sequences, etc., and obtains historical processing results by executing the output instructions of the large language model. Finally, a cache processing mechanism is used to store historical processing results. The embodiment of the present invention significantly improves the efficiency and quality of property management, optimizes the accuracy and timeliness of decision-making, improves service efficiency, reduces labor costs, and reduces workload; at the same time, a more humane service model is realized, and the property management industry is promoted to develop in the direction of intelligence and efficiency.

[0060] The embodiment of the present invention also provides a device for generating management decisions based on a property manager agent, which can be implemented in the form of a computer program. The computer program can be used in Fig.10 Runs on the computer device shown.

[0061] See also Fig.10 , Fig.10 1 is a schematic block diagram of a computer device provided by an embodiment of the present invention. The computer device 1000 is a server, which may be an independent server or a server cluster composed of multiple servers.

[0062] See also Fig.10 The computer device 1000 includes a processor 1002 , a memory and a network interface 1005 connected via a system bus 1001 , wherein the memory may include a non-volatile storage medium 1003 and an internal memory 1004 .

[0063] The non-volatile storage medium 1003 can store an operating system 10031 and a computer program 10032. When the computer program 10032 is executed, the processor 1002 can execute a method for generating management decisions based on a property manager agent.

[0064] The processor 1002 is used to provide computing and control capabilities to support the operation of the entire computer device 1000.

[0065] The internal memory 1004 provides an environment for the operation of the computer program 10032 in the non-volatile storage medium 1003. When the computer program 10032 is executed by the processor 1002, the processor 1002 can execute a method for generating management decisions based on a property manager agent.

[0066] The network interface 1005 is used for network communication, such as providing data information transmission, etc. Those skilled in the art will understand that Fig.10 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device 1000 to which the solution of the present invention is applied. The specific computer device 1000 may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0067] Those skilled in the art will understand that Fig.10 The embodiments of the computer device shown in the figure do not constitute a limitation on the specific composition of the computer device. In other embodiments, the computer device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. For example, in some embodiments, the computer device may only include a memory and a processor. In such embodiments, the structure and function of the memory and the processor are the same as those of the embodiment of the present invention. Figure 1 The embodiments shown are consistent and will not be described again here.

[0068] It should be understood that in the embodiment of the present invention, the processor 1002 may be a central processing unit (CPU), and the processor 1002 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0069] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the above-described equipment, devices and units can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. Those of ordinary skill in the art can appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, computer software or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0070] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation, and units with the same function may be combined into one unit, for example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0071] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention.

[0072] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0073] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), disk or optical disk and other media that can store program code.

[0074] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for generating management decisions based on a property manager agent, characterized in that: include: Collect relevant data from the property management system and obtain long text data sets; Using RAG technology to clean and preprocess the long text data set to obtain preprocessed property management data; Build a property manager agent; The property manager agent performs reasoning analysis on the property management data to generate management decisions; The steps of constructing the property manager agent include: Setting tasks, wherein the tasks include defining objectives, setting output constraints and resource lists; Calling a large language model to process the task; Executing the instructions output by the large language model to obtain historical processing results; wherein the instructions include parsing instructions, selecting execution tools and execution tool sequences; A cache processing mechanism is used to store the historical processing results.

2. The method for generating management decisions based on a property manager agent according to claim 1, characterized in that: The calling of the large language model to process the task includes: Encode the input sequence according to the following formula to obtain the input code: ,in, Embedding represents embedding, Positional Encoding represents the positional encoding, and t represents the input sequence; The input code is linearly transformed according to the following formula to obtain the query matrix, key matrix and value matrix: ,in , , is a learnable weight matrix, Q represents the query matrix, K represents the key matrix, and V represents the value matrix; The self-attention weight is calculated by the query matrix, key matrix and value matrix, and the output result of the self-attention layer is obtained according to the following formula: ,in, is the vector dimension of the key matrix K, is the transposed matrix of the key matrix K.

3. The method for generating management decisions based on a property manager agent according to claim 2, characterized in that: The calling of the large language model to process the task further includes: The multi-head attention mechanism is used to send the input code to multiple self-attention layers respectively, and then the output results of the self-attention layers are spliced ​​and linearly transformed through the weight matrix. The linear transformation result of the multi-head attention mechanism is obtained according to the following formula: ,in, represents the weight matrix, Represents the i-th output result of the self-attention layer; The linear transformation result of the multi-head attention mechanism is further transformed nonlinearly by a feedforward neural network. The nonlinear transformation result is expressed as follows: ,in, and is the weight matrix in the feedforward neural network, and is the bias term in the feedforward neural network; The features of different layers of the large language model are integrated as follows: ; The probability distribution is generated by the linear layer and the softmax function, and the output encoding is obtained according to the following formula: ,in, is the weight matrix of the output embedding layer, is the predicted next word.

4. The method for generating management decisions based on a property manager agent according to claim 1, characterized in that: The method of performing reasoning and analysis on the property management data by the property manager agent to generate management decisions includes: Output the inference results according to the following formula: , where R represents the preset rules generated by RAG technology, and a represents the analysis algorithm; According to the above reasoning results, property behaviors are divided into standardized behaviors and non-standardized behaviors according to the following formula: , where J is the classification function, B is the standardized behavior, and N is the unstandardized behavior; According to the standardized behavior, the property manager agent automatically generates work orders and assigns tasks through integration with the work order system; For non-standardized behaviors, the property manager agent sends daily reports to management personnel regularly according to a preset schedule.

5. The method for generating management decisions based on a property manager agent according to claim 1, characterized in that: Also includes: Construct a dictionary including key output results and corresponding timestamps, and the dictionary is expressed as follows: ,in, Indicates the output result, Indicates the corresponding timestamp; Based on current goals and historical information, the relevance of behavioral information is determined as follows: ,in, Indicates the current target. Represents historical information. is the filter function; The dictionary is cached and updated, the new key output results are added to the dictionary, and irrelevant behavior information is removed according to the following formula: ,in, is the cache update function, Indicates the new key output result.

6. The method for generating management decisions based on a property manager agent according to claim 1, characterized in that: The RAG technology is used to clean and preprocess the long text data set to obtain preprocessed property management data, including: The property-related queries and documents in the long text dataset are converted into vector representations according to the following formula: , where Q and C represent property-related queries and documents, respectively. is an n-dimensional real vector space, represents the vector representation of each document in the document collection, and q represents the query vector; The vector search between the query vector and the document vector is calculated as follows: ,in, Represents the query vector q and the document vector The vector dot product of Represents the Euclidean norm of a vector; The retrieved relevant documents and property-related queries are concatenated and merged to form context information according to the following formula: ,in, Indicates related documents; The context information is input into the generation model to generate a text set according to the following formula: , where G is the generative model and m is the dimension of the context vector.

7. The method for generating management decisions based on a property manager agent according to claim 6, characterized in that: The use of RAG technology to clean and preprocess the long text data set to obtain preprocessed property management data also includes: Using the preset rules generated by RAG technology, the long text data set is hashed for deduplication, correction and completion operations; The accuracy of the data in the long text dataset is calculated according to the following formula: ,in, represents a data set, Representation dataset A single data item in Representation dataset The correct data item in The completeness rate of the data in the long text dataset is calculated according to the following formula: , where n represents the data set The total number of data fields in represents the jth field, It is an indicator function that returns 0 or 1 according to the value of the input data field; The consistency of the data in the long text dataset is calculated according to the following formula: , where z represents the data set Data items that conform to the specifications.

8. A device for generating management decisions based on a property manager agent, characterized in that: include: A collection unit is used to collect relevant data of the property management system and obtain a long text data set; A preprocessing unit, used for cleaning and preprocessing the long text data set by using RAG technology to obtain preprocessed property management data; Construction unit, used to build property manager agent; A reasoning and analysis unit, used to perform reasoning and analysis on the property management data through the property manager agent to generate management decisions; The steps of constructing the property manager agent include: Setting tasks, wherein the tasks include defining objectives, setting output constraints and resource lists; Calling a large language model to process the task; Executing the instructions output by the large language model to obtain historical processing results; wherein the instructions include parsing instructions, selecting execution tools and execution tool sequences; A cache processing mechanism is used to store the historical processing results.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, it implements the method for generating management decisions based on a property manager agent as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, causes the processor to execute the method for generating management decisions based on a property manager agent as described in any one of claims 1 to 7.

Images