A smart property city management system and method

Abstract

The present invention discloses a smart property city management system, including a client and a service platform, wherein the client and the service platform are connected through Internet communication, and the service platform includes a server, a feedback module, a publishing module, an order grabbing module and an audit module. The feedback module is connected through communication with the publishing module, the publishing module is connected through communication with the server, the publishing module is connected through communication with the order grabbing module, the order grabbing module is connected through communication with the audit module, the feedback module is connected through communication with the server through the publishing module, the server is connected through communication with the order grabbing module through the publishing module, and the order grabbing module is connected through communication with the server through the audit module.

Description

Technical Field

[0001] The present invention relates to the technical field of urban management, and in particular to a smart property urban management system and method. Background Art

[0002] Urban management refers to the decision-making guidance, standardization and coordination, service and management behavior around the operation and development of the city, which is based on the open and complex giant system of the city, the basic information flow of the city, and the use of a series of mechanisms such as decision-making, planning, organization and command, and the use of legal, economic, administrative and technical means, through the interaction between the government, the market and the society. Urban management in a broad sense refers to the management of all urban activities, including political, economic, social and municipal management. Urban management in a narrow sense usually refers to municipal management, that is, the management of urban infrastructure, public service facilities and social public affairs related to urban planning, urban construction and urban operation. The object of general urban management research is mainly aimed at urban management in a narrow sense, that is, municipal management, but at this stage, urban management has the following shortcomings: chaotic management: caused by the system, many subjects, and difficult cooperation; few people and many things: unsatisfactory staffing; insufficient specialization: insufficient specialization of the personnel themselves; unprofessional management level; unspecialized scientific and technological means; insufficient refinement: backward means and overly macro planning. Summary of the invention

[0003] The purpose of the present invention is to provide a smart property city management system and method, which has the advantages of technical basis and conditions for cross-regional promotion.

[0004] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a smart property city management system, including a client and a service platform, the client and the service platform are connected through the Internet, the service platform includes a server, a feedback module, a publishing module, an order grabbing module and an auditing module; the feedback module is connected to the publishing module, the publishing module is connected to the server, the publishing module is connected to the order grabbing module, and the order grabbing module is connected to the auditing module;

[0005] The feedback module includes an identity authentication unit, a category selection unit, a feedback unit, and a judgment generation unit, wherein the output end of the identity authentication unit is connected to the input end of the category selection unit, the output end of the category selection unit is connected to the input end of the feedback unit, the output end of the feedback unit is connected to the input end of the judgment generation unit, and the output end of the judgment generation unit is connected to the publishing module; wherein the judgment generation unit uses a keyword extraction algorithm to extract keywords, arrange and combine them, and then output them to the publishing module.

[0006] Furthermore, the order grabbing module includes an identification unit, a keyword matching unit, a reminder unit and an execution unit, wherein the output end of the identification unit and the output end of the publishing module are simultaneously connected to the input end of the keyword matching unit, the output end of the keyword matching unit is connected to the input end of the reminder unit, and the output end of the reminder unit is connected to the input end of the execution unit.

[0007] Furthermore, the category selection unit, feedback unit, and judgment generation unit respectively include M one-to-one corresponding category selection sub-units, feedback sub-units, and judgment generation sub-units; M is an integer greater than 0.

[0008] Furthermore, the client includes an input client, an auxiliary client, a reward client and an execution client.

[0009] A smart property city management method comprises the following steps:

[0010] S01: The input end user selects the problem feedback interface in the input client, and the identity authentication unit authenticates the identity of the input end user; the authenticated input end user selects the category of the problem to be fed back in the category selection unit; and enters the feedback sub-unit under the category;

[0011] S02: The input end user inputs feedback question information into the feedback subunit, and the feedback subunit outputs the received feedback question information to the corresponding judgment generation subunit;

[0012] S03: the judgment generation subunit uses a keyword extraction algorithm to extract keywords, arrange and combine them, and then output them to the publishing module;

[0013] S04: the publishing module publishes the feedback question information containing the arranged and combined keywords;

[0014] S05: The identification unit authenticates the identity of the stored auxiliary end user and / or execution end user; and outputs the authenticated auxiliary end user and / or execution end user and their corresponding introduction to the keyword matching unit; and the publishing module synchronously outputs the feedback question information to the keyword matching unit;

[0015] S06: The keyword matching unit extracts keywords from the feedback question information and the assistant end user and / or the execution end user introduction respectively, matches the keywords, and obtains the assistant end user and / or the execution end user sequence according to the matching degree;

[0016] S07: the reminder unit sends reminder information to each person in the assistant end user and / or execution end user sequence in sequence, and the frequency of the reminder information decreases with the order of the persons in the assistant end user and / or execution end user sequence;

[0017] S08: The auxiliary end user and / or the execution end user chooses whether to execute the task on the auxiliary client and / or the execution client. If execution is chosen, the execution unit displays the feedback problem information and outputs the execution result to the review module after executing the task.

[0018] Furthermore, in step S03 and step S06, a TextRank algorithm model trained by a neural network is used to extract keywords.

[0019] Furthermore, the TextRank algorithm model shown in the figure specifically includes the following steps to extract keywords:

[0020] B01: Segment the feedback information into complete sentences;

[0021] B02: Perform word segmentation and magnetic tagging on each sentence, and filter out stop words;

[0022] B03: Construct a candidate keyword graph G = (V, E), where V is a node set consisting of sentences with stop words filtered out, and then use the co-occurrence relationship to construct an edge between any two points. An edge exists between two nodes only if their corresponding words co-occur in a window of length K, where K represents the window size.

[0023] B04: Iterate and propagate the weight of each node until the TextRank algorithm model converges;

[0024] B05: Sort the node weights in reverse order to obtain keywords.

[0025] Furthermore, it also includes step S09: the review module rates the execution task and awards points based on the execution results input by the auxiliary end user and / or the execution end user and the feedback of the input end user on the execution results. At the same time, the review module extracts keywords from the execution results and adds them to the introduction of the auxiliary end user and / or the execution end user.

[0026] Further, it includes step S010: the reward end user publishes product information and corresponding points redemption amount on the "Points Mall" platform through the reward client, and the auxiliary end user and / or the execution end user can redeem the products through the client.

[0027] Furthermore, in step S08, if the assistant end user and / or the execution end user chooses not to execute on the assistant client and / or the execution client, the reminder unit stops reminding the assistant end user and / or the execution end user.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] 1. Wide range of participants: Ordinary citizens can participate. They can not only complain, report and reflect problems, but also take orders to participate in the handling and resolution of problems, solving the problem that the masses were just "spectators" in urban governance in the past, allowing the masses to participate conveniently and become the masters; people with professional knowledge can also participate and become professional volunteers to play a role in related professional fields; merchants can participate in the entire system by exchanging public welfare points, changing the previous status of being supervised and free, and participating in the urban governance system with honest and high-quality goods and services;

[0030] 2. Rich work types: Avoid the situation of buck-passing between departments, and solve the problem of industry fragmentation, lack of information exchange, and lack of interaction in actions;

[0031] 3. Quick and convenient rewards: Online rewards and instant payments can be realized, and traditional problems can be solved with Internet thinking, which conforms to the living habits of modern people, making it more convenient for the masses to participate and optimizing the channels;

[0032] 4. Solve the manpower problem: The traditional approach must be changed. We should advocate the concept of "citizen governance", let the masses participate in urban governance, let professional volunteers, professional companies and other social forces participate, and solve the serious shortage of public human resources;

[0033] 5. Establish a new "ecosystem": Through the APP system, citizens, volunteers, professional companies, businesses, law enforcement officers, etc. will form a new "ecosystem". Each group will play its own role, from the outside to the inside, promoting and supervising each other, allowing more social forces to participate in resolving social problems, solving the problem of a sharp increase in law enforcement business volume and a serious shortage of human resources, and promoting citizens to participate in and supervise urban governance throughout the process, so as to achieve the greatest work effectiveness with the smallest public expenditure and the broadest social participation, and form a vivid working situation of "coming from the masses and going to the masses". BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a schematic diagram of the service platform framework of the present invention;

[0035] Figure 2 This is a schematic diagram of the specific framework of the order grabbing module and the feedback module of the present invention;

[0036] Figure 3 The flowchart of the TextRank algorithm model of the present invention for extracting keywords. DETAILED DESCRIPTION

[0037] 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 only 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.

[0038] As attached Figure 1 As shown, a smart property city management system provided by the present invention includes a client and a service platform. The client and the service platform are connected through Internet communication. The service platform includes a server, a feedback module, a publishing module, an order grabbing module and an audit module; the feedback module is connected to the publishing module, the publishing module is connected to the server, the publishing module is connected to the order grabbing module, and the order grabbing module is connected to the audit module.

[0039] The feedback module includes an identity authentication unit, a category selection unit, a feedback unit, and a judgment generation unit, wherein the output of the identity authentication unit is connected to the input of the category selection unit, the output of the category selection unit is connected to the input of the feedback unit, the output of the feedback unit is connected to the input of the judgment generation unit, and the output of the judgment generation unit is connected to the release module; wherein the judgment generation unit uses a keyword extraction algorithm to extract keywords, arrange and combine them, and then output the release module. The category selection unit, the feedback unit, and the judgment generation unit respectively contain M one-to-one corresponding category selection subunits, feedback subunits, and judgment generation subunits; M is an integer greater than 0; each selection subunit corresponds to a feedback transaction, for example, all feedback issues can be divided into sections of types such as public security, construction, industry and commerce, and finance, so as to quickly and accurately match the executive personnel for solving related problems.

[0040] The order grabbing module includes an identification unit, a keyword matching unit, a reminder unit and an execution unit, wherein the output end of the identification unit and the output end of the publishing module are simultaneously connected to the input end of the keyword matching unit, the output end of the keyword matching unit is connected to the input end of the reminder unit, and the output end of the reminder unit is connected to the input end of the execution unit.

[0041] The client in the present invention may specifically include an input client, an auxiliary client, a reward client and an execution client.

[0042] Please see attached Figure 1-2 The present invention provides a smart property city management method, comprising the following steps:

[0043] S01: The input end user selects the question feedback interface in the input client, and the identity authentication unit authenticates the identity of the input end user; the authentication here mainly matches the identity of the input end user with its registration information, and then obtains the information of the input end user; if the input end user has not registered, it is necessary to register in the input client now and fill in the relevant information before entering the next step.

[0044] The authenticated input end user selects the category of the problem to be fed back in the category selection unit; and enters the feedback sub-unit under the category; the category selection unit contains M category selection sub-units, each category selection sub-unit corresponds to the affairs of a section, for example, it can correspond to sections such as public security, construction, industry and commerce, finance, etc. The input end user selects the corresponding category selection sub-unit according to the problem he needs to feedback. In the present invention, corresponding category selection sub-units are set for different sections, mainly to classify the feedback problem information more quickly and then solve it.

[0045] S02: The input end user inputs feedback question information into the feedback sub-unit, and the feedback sub-unit outputs the received feedback question information to the corresponding judgment generation sub-unit.

[0046] S03: The judgment generation subunit uses a keyword extraction algorithm to extract keywords, arrange and combine them, and then output them to the publishing module; in the present invention, the category selection subunit, the feedback subunit, and the judgment generation subunit are in a one-to-one correspondence, and each selection subunit corresponds to a feedback subunit and a judgment generation subunit, which can ensure that each feedback problem information is solved in a timely manner, and can also ensure that the feedback problem information in different category selection subunits can be processed separately without affecting each other.

[0047] In this step, the TextRank algorithm model trained by the neural network can be used to extract keywords, and the specific keywords include time, place, person and event. The specific neural network algorithm includes the following steps:

[0048] S031: Setting a training set, wherein the training set includes a plurality of training sentences, and keywords have been marked in each training sentence; the keywords may include time, place, person, and event.

[0049] S032: Input each training sentence into the TextRank algorithm model in turn, and compare the output result with the keywords marked in the training sentence, and continuously adjust the model parameters of the TextRank algorithm model until the output accuracy of the TextRank algorithm model reaches the set threshold; the TextRank algorithm model at this time is the TextRank algorithm model after neural network training.

[0050] Before outputting the TextRank algorithm model, the accuracy of the TextRank algorithm model can be further verified using a test set, wherein the test set contains training sentences that are different from the training set.

[0051] As attached Figure 3 As shown, the process of extracting keywords using the TextRank algorithm model trained by a neural network in the present invention specifically includes the following steps:

[0052] B01: Segment the feedback information into complete sentences;

[0053] B02: Perform word segmentation and magnetic tagging on each sentence, and filter out stop words; retain words with specified parts of speech, such as nouns, verbs, etc.

[0054] B03: Construct a candidate keyword graph G = (V, E), where V is a node set consisting of sentences with stop words filtered out, and then use the co-occurrence relationship to construct an edge between any two points. An edge exists between two nodes only if their corresponding words co-occur in a window of length K, where K represents the window size, that is, at most K words co-occur.

[0055] B04: Iteratively propagate the weight of each node until the TextRank algorithm model converges; the specific method of iteratively propagating the weight of each node can adopt the existing formula.

[0056] B05: Sort the node weights in reverse order, and obtain keywords based on the set keyword targets and the sorting results.

[0057] S04: The publishing module publishes the feedback question information containing the arranged and combined keywords; the arrangement and combination in this step can be the feedback question information after combining the keywords in a specific order, for example, the combined feedback question information is time, place, person, and event.

[0058] S05: The identification unit authenticates the identity of the stored auxiliary end user and / or execution end user; and outputs the authenticated auxiliary end user and / or execution end user and its corresponding introduction to the keyword matching unit; at the same time, the publishing module synchronously outputs the feedback question information to the keyword matching unit.

[0059] In this step, the identity of the stored auxiliary end user and / or execution end user is authenticated, mainly verifying the identity of the registered auxiliary end user and execution end user, wherein each auxiliary end user or execution end user includes a personal profile, which is used to mark information such as the field that the auxiliary end user or execution end user is good at solving.

[0060] S06: The keyword matching unit extracts keywords from the feedback question information and the assistant end user and / or the execution end user introduction respectively, matches the keywords, and obtains the assistant end user and / or the execution end user sequence according to the matching degree.

[0061] The keyword matching unit uses the trained TextRank algorithm model to extract the keywords in each personal profile and the keywords in the feedback question information. It is worth noting that the keywords in the feedback question information here have been extracted in step S03 and arranged in a specific order. There is no need to extract them again here, and only the sentences in a specific order need to be split.

[0062] The method of extracting keywords of the personal profile may be as shown in step S03, which will not be described in detail here.

[0063] At this time, the keyword matching unit matches the keywords in the personal profile and the keywords in the feedback problem information. Specifically, the matching can be performed according to the number of identical or similar keywords. The more identical or similar keywords there are, the higher the matching degree between the auxiliary end user or the execution end user and the feedback problem information. According to the matching degree, the auxiliary end user and / or execution end user sequence is obtained. Note: The sequence here is for each feedback problem information. The auxiliary end user and / or execution end user sequence obtained from different feedback problem information is different. The person who is closer to the front in the sequence is more suitable for solving the feedback problem. In order to speed up the matching speed and the efficiency of solving feedback problems, a threshold matching degree can be set. People with matching degrees below the threshold can be directly filtered out and no longer appear in the auxiliary end user and / or execution end user sequence.

[0064] S07: The reminder unit sends reminder information to each person in the auxiliary end user and / or execution end user sequence in turn, and the frequency of the reminder information decreases with the order of the people in the auxiliary end user and / or execution end user sequence. The purpose of this step is to first send reminder information to the auxiliary end user and / or execution end user with a high matching degree, and then send reminder information to the auxiliary end user and / or execution end user with a low matching degree, and the reminder frequency for the auxiliary end user and / or execution end user with a high matching degree is higher. These settings are all to ensure that the auxiliary end user and / or execution end user with a high matching degree has priority in receiving orders and processing tasks.

[0065] S08: The assistant end user and / or the execution end user chooses whether to execute the task on the assistant client and / or the execution client until someone accepts the order, and the reminder unit stops reminding; at this time, the execution unit displays the feedback problem information, and the person who accepts the order outputs the execution result to the review module after executing the task. If the assistant end user and / or the execution end user chooses not to execute the task on the assistant client and / or the execution client, the reminder unit continues to remind the assistant end user and / or the execution end user.

[0066] S09: The review module rates the execution task and awards points based on the execution results input by the auxiliary-end user and / or the execution-end user and the input-end user's feedback on the execution results. At the same time, the review module extracts keywords from the execution results and adds them to the introduction of the auxiliary-end user and / or the execution-end user; this is to further improve the task experience of the auxiliary-end user and / or the law enforcement officer, and update their introduction as the task is executed.

[0067] S010: The reward-end user publishes product information and the corresponding points redemption amount on the "Points Mall" platform through the reward client, and the auxiliary-end user and / or execution-end user can redeem products through the client.

[0068] The feedback problem information in the present invention can be daily clerical work such as discovering problems, patrolling, educating, persuading, and handling simple problems, so that these clerical work can be handed over to social forces for intervention and handling, and the form of integrating volunteer services and effective incentives can be used to stimulate the sense of responsibility and honor of input-end users as the masters of the city, truly achieving "the people's city is managed by the people."

[0069] The following further explains the system of the present invention from the perspectives of input client, auxiliary client, execution client and reward client. It is worth noting that each input end user can provide feedback and ask for help through the input client, grab orders and execute tasks through the auxiliary client, and redeem points for merchandise through the reward client.

[0070] From the input client perspective:

[0071] Register an account: Select the input end user identity to register an account through the client, enter the corresponding information to submit the account application. The input end user can be an ordinary citizen.

[0072] Report a case: Select case classification, automatically located location, case description, voice description, video description, and image upload through the feedback module.

[0073] Feedback case: After the feedback problem information is resolved, the task executor can be evaluated, etc.

[0074] From the secondary client perspective:

[0075] Register an account: Select a professional assistant user identity through the client to register an account, enter the corresponding information and submit the account application. The assistant user can be a volunteer.

[0076] Participate in activities: View activity information through the feedback module and sign up according to your interests.

[0077] Grab the order: The case reported by the input-end user is assigned to the auxiliary-end user through the grab order module, and the auxiliary-end user grabs the case order.

[0078] Processing cases: Assist end users to process the cases they have grabbed, and upload the processed video description, voice description, picture description, text description, and text description case materials.

[0079] Scoreboard: View the user's score ranking and see your own score ranking.

[0080] Mall: Exchange goods in the mall, and use the redemption code obtained by redeeming goods to redeem goods in time.

[0081] Case reality: Through the review module, you can view the details of the closed case before and after it is reported, and you can handle the case.

[0082] From the perspective of rewarding clients:

[0083] Register an account: Select the reward end user identity to register an account through the client, enter the corresponding information to submit the account application. The reward end user can be a merchant.

[0084] Upload products: select product categories, upload product information, pictures, and product prices.

[0085] Upload coupons: upload discount price information, coupon content, and optimize money.

[0086] Verify products: Verify the product order through the redemption code obtained by the user when redeeming the product.

[0087] Verify coupons: Verify coupons using the redemption code obtained from the user's redemption of the discount price.

[0088] Points withdrawal: Enter the correct WeChat ID or the correct Alipay account to apply for withdrawal.

[0089] From the execution client perspective:

[0090] Register an account: Select the execution end user identity to register an account through the client, enter the corresponding information to submit the account application. The execution end user can be a law enforcer.

[0091] Grab an order: The case reported by the input-end user is assigned to the execution-end user, who then grabs the case through the grab order module.

[0092] Case processing: by uploading processed video description, voice description, picture description, text description, and text description case materials.

[0093] Scoreboard: View the user's score ranking and see your own score ranking.

[0094] Mall: Exchange goods in the mall, and use the redemption code obtained by redeeming goods to redeem goods in time.

[0095] Case reality: Through the review module, you can view the details of the closed case before and after it is reported, and you can handle the case.

[0096] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A smart property city management system, including a client and a service platform, characterized in that: The client is connected to the service platform through the Internet, and the service platform includes a server, a feedback module, a publishing module, an order grabbing module and an audit module; the feedback module is connected to the publishing module, the publishing module is connected to the server, the publishing module is connected to the order grabbing module, and the order grabbing module is connected to the audit module; The feedback module includes an identity authentication unit, a category selection unit, a feedback unit, and a judgment generation unit, wherein the output end of the identity authentication unit is connected to the input end of the category selection unit, the output end of the category selection unit is connected to the input end of the feedback unit, the output end of the feedback unit is connected to the input end of the judgment generation unit, and the output end of the judgment generation unit is connected to the publishing module; wherein the judgment generation unit uses a keyword extraction algorithm to extract keywords, arrange and combine them, and then output them to the publishing module; The category selection unit, feedback unit, and judgment generation unit respectively include M one-to-one corresponding category selection sub-units, feedback sub-units, and judgment generation sub-units; M is an integer greater than 0, and each category selection sub-unit corresponds to a forum transaction; The order grabbing module includes an identification unit, a keyword matching unit, a reminder unit and an execution unit, wherein the output end of the identification unit and the output end of the publishing module are simultaneously connected to the input end of the keyword matching unit, the output end of the keyword matching unit is connected to the input end of the reminder unit, and the output end of the reminder unit is connected to the input end of the execution unit; The judgment generation subunit uses a keyword extraction algorithm to extract keywords, arrange and combine them, and then output them to the publishing module; The publishing module publishes the feedback question information containing the arranged and combined keywords; The identification unit authenticates the identity of the stored auxiliary end user and / or execution end user; and outputs the authenticated auxiliary end user and / or execution end user and their corresponding introduction to the keyword matching unit; and the publishing module synchronously outputs the feedback question information to the keyword matching unit; The keyword matching unit extracts keywords from the feedback question information and the introduction of the auxiliary end user and / or the execution end user, respectively, and matches the keywords, and obtains the sequence of the auxiliary end user and / or the execution end user according to the degree of matching; a threshold matching degree is set, and the personnel with a matching degree below the threshold can be directly filtered out and no longer appear in the sequence of the auxiliary end user and / or the execution end user; The reminder unit sends reminder information to each person in the assistant end user and / or execution end user sequence in sequence, and the frequency of the reminder information decreases with the order of the person in the assistant end user and / or execution end user sequence; The assistant end user and / or the execution end user chooses whether to execute the task on the assistant client and / or the execution client. If the task is executed, the execution unit displays the feedback problem information. After executing the task, the assistant end user and / or the execution end user outputs the execution result to the review module. If the assistant end user and / or the execution end user chooses not to execute on the assistant client and / or the execution client, the reminder unit stops reminding the assistant end user and / or the execution end user.

2. A smart property city management system according to claim 1, characterized in that: The clients include reward clients.

3. A working method of the smart property city management system according to claim 1, characterized in that: The steps include: S01: The input end user selects a question feedback interface on the input client, and the identity authentication unit authenticates the identity of the input end user; the authenticated input end user selects a category of the question to be fed back in the category selection unit; and enter the feedback subunit under this category; S02: The input end user inputs feedback question information into the feedback subunit, and the feedback subunit outputs the received feedback question information to the corresponding judgment generation subunit; S03: the judgment generation subunit uses a keyword extraction algorithm to extract keywords, arrange and combine them, and then output them to the publishing module; S04: the publishing module publishes the feedback question information containing the arranged and combined keywords; S05: The identification unit authenticates the identity of the stored auxiliary end user and / or execution end user; and outputs the authenticated auxiliary end user and / or execution end user and their corresponding introduction to the keyword matching unit; and the publishing module synchronously outputs the feedback question information to the keyword matching unit; S06: The keyword matching unit extracts keywords from the feedback question information and the assistant end user and / or the execution end user introduction respectively, matches the keywords, and obtains the assistant end user and / or the execution end user sequence according to the matching degree; S07: the reminder unit sends reminder information to each person in the assistant end user and / or execution end user sequence in sequence, and the frequency of the reminder information decreases with the order of the persons in the assistant end user and / or execution end user sequence; S08: The auxiliary end user and / or the execution end user chooses whether to execute the task on the auxiliary client and / or the execution client. If execution is chosen, the execution unit displays the feedback problem information. After executing the task, the auxiliary end user and / or the execution end user outputs the execution result to the review module.

4. A working method of a smart property city management system according to claim 3, characterized in that: In step S03 and step S06, the TextRank algorithm model trained by a neural network is used to extract keywords.

5. The working method of a smart property city management system according to claim 4 is characterized in that: The TextRank algorithm model for extracting keywords specifically includes the following steps: B01: Segment the feedback information into complete sentences; B02: Perform word segmentation and magnetic tagging on each sentence, and filter out stop words; B03: Construct a candidate keyword graph G = (V, E), where V is a node set consisting of sentences with stop words filtered out, and then use the co-occurrence relationship to construct an edge between any two nodes. An edge exists between two nodes only if their corresponding words co-occur in a window of length K, where K represents the window size; B04: Iterate and propagate the weight of each node until the TextRank algorithm model converges; B05: Sort the node weights in reverse order to obtain keywords.

6. The working method of a smart property city management system according to claim 3 is characterized in that: It also includes step S09: the review module rates the execution task and awards points based on the execution results input by the auxiliary end user and / or the execution end user and the feedback of the input end user on the execution results. At the same time, the review module extracts keywords from the execution results and adds them to the introduction of the auxiliary end user and / or the execution end user.

7. The working method of a smart property city management system according to claim 3 is characterized in that: The process includes step S010: the reward end user publishes product information and corresponding points redemption amount on the "points mall" platform through the reward client, and the auxiliary end user and / or the execution end user can redeem products through the client.

Images