Domestic service platform order sending system based on Internet technology

Abstract

The invention discloses a housekeeping platform order sending system based on the Internet technology, and relates to the technical field of housekeeping management systems, the housekeeping platform order sending system comprises a user side and a server side which are used for users and service personnel to place orders and receive orders, and a multi-dimensional portrayal module which is based on the relevant information of the service personnel and is used for sending the orders to the user side and the server side. And the multi-dimensional order sending module is used for processing and analyzing service requirements input by the user by adopting a natural language, extracting implicit requirements except for basic service types, time and places, and generating requirement feature vectors of the user, and is used for carrying out side portrait evaluation on service personnel and carrying out side portrait evaluation on the user according to housekeeping order information of the user as the same, and the multi-dimensional order sending module is used for processing and analyzing the service requirements input by the user by adopting a natural language, and can be used for extracting implicit requirements and generating requirement feature vectors of the user. Through the multi-dimensional portrait module, side portrait evaluation is carried out on service personnel from multiple aspects such as personality, capability, service evaluation, order completion degree, skill and education background, the problem of inaccurate matching caused by single index in a traditional order dispatching mode is avoided, and the order dispatching efficiency and the service quality are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of housekeeping management systems, and in particular to a housekeeping platform dispatching system based on Internet technology. Background Art

[0002] A housekeeping platform refers to a service model that uses the Internet to match housekeeping service needs with providers online. It covers various service needs in the family, such as nannies, confinement nannies, cleaners, repairmen, movers, pet care, etc. The housekeeping platform dispatching system is a service platform based on Internet technology, which aims to connect housekeeping service providers and demanders efficiently and accurately. The system collects, organizes and analyzes housekeeping service supply and demand information to optimize the allocation of service resources, improve service efficiency, and provide users with a more convenient and reliable housekeeping service experience.

[0003] The current housekeeping platform dispatches orders after the user places an order. If the corresponding service personnel are not specified, the dispatch system will match the order with the service personnel based on the order service type, demand and distance. This dispatch method mainly relies on the geographical location and whether the service personnel have the label of the relevant ability, making the quantitative indicators of dispatch relatively simple, ignoring deep factors such as the attitude of the service personnel, the emergency response capability of the service, the professional skills and the dynamic changes of the service situation, making the dispatch process not flexible enough. Summary of the Invention

[0004] In view of the shortcomings of the existing technology, the present invention provides a housekeeping platform dispatching system based on Internet technology, which solves the problems mentioned in the background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: The housekeeping platform dispatching system includes:

[0006] User and Service Side: used by users and service personnel to place and receive orders;

[0007] Multi-dimensional portrait module: Based on the relevant information of service personnel, a profile evaluation of service personnel is conducted. Similarly, a profile of users is conducted based on their housekeeping order information.

[0008] Multi-dimensional dispatch module: This module uses natural language processing to analyze user input service requirements. In addition to basic service type, time, and location, it can extract implicit requirements and generate user demand feature vectors.

[0009] Full process management module: Service personnel record service tracks through GPS positioning and timestamps, and communicate with service personnel in real time, with automatic warnings for abnormal situations;

[0010] Scenario evaluation module: build a virtual service scenario library, service personnel regularly participate in scenario simulation training, regularly update scenarios to match emerging needs;

[0011] Cooperative scheduling module: integrate real-time traffic data, service personnel load rate and order space-time constraints to achieve multi-objective optimization of order dispatching;

[0012] Service blockchain storage module: chain storage of key service data, support cross-chain verification of service personnel's identity and health information, realize "person, certificate and health" three-source verification of service personnel;

[0013] Service rating optimization module: update service personnel ratings based on user service evaluations, and conduct evaluation credibility analysis to ensure the timeliness and fairness of the rating system.

[0014] Preferably, the multi-dimensional portrait module evaluates the side portrait of the service personnel, specifically including the following: personality, ability, service evaluation, order completion, skill and education background;

[0015] The multi-dimensional portrait module evaluates the side portrait of the user, specifically including the following: service demand history data, special requirements and constraint conditions.

[0016] Preferably, the multi-dimensional order dispatching module extracts the user's implicit demand, including: skill demand, service scenario demand, special qualification demand, time demand, service quality demand, service detail demand and risk control demand.

[0017] Preferably, the service personnel monitoring of the full-process management module is realized by obtaining the device permissions of the service personnel's mobile intelligent device through the service end, and the authorized permissions include but are not limited to GPS permissions, time permissions, communication permissions, camera permissions and recording permissions.

[0018] Preferably, the virtual service scenario generated by the scenario evaluation module is simulated by the service personnel through an external VR device.

[0019] The main content of the virtual service scenario is the simulation training of emergency situations and service content, and the service personnel are scored after the training, and the score is taken into account in the order dispatching weight of the service personnel.

[0020] Preferably, the scheduling method of the cooperative scheduling module is as follows:

[0021] Candidate set: after the user places an order, the cooperative scheduling module filters the service personnel's state, filters out service personnel with service distance <5km, demand skill matching and load rate <70%;

[0022] Priority: the qualified service personnel are further calculated by the comprehensive score, which is composed of distance score*0.3, scenario simulation score*0.4 and historical praise rate*0.3;

[0023] The service personnel with the top three comprehensive scores are the high-weighted order dispatch personnel, and the service personnel with the top three matching degrees are sent to the user end as recommended items. If the user does not specify the service personnel, the order is preferentially dispatched to the service personnel with the highest comprehensive score to complete the order;

[0024] The load rate refers to the number of orders received by the service personnel on the same day and the working time on the same day. When the number of consecutive orders received by the service personnel and the service time exceed the working load threshold, the order dispatch weight of the corresponding service personnel is automatically reduced to avoid fatigue work.

[0025] Preferably, the data types of the key service data uploaded to the service blockchain storage module include service trajectory, service node evidence, biological feature verification and service monitoring. The service data of the service blockchain storage module is collected and uploaded through the whole-process management module.

[0026] Preferably, the service trajectory is the positioning and timestamp data of the service personnel, and the data collection interval is every 5 minutes;

[0027] The service node evidence is the photographing and storage data of the service personnel on the service scene before, during and after the service at the user address;

[0028] The biological feature verification is the biological verification of the service personnel within a specified time after receiving the order, including face, fingerprint, palm vein and voiceprint. The service personnel must complete the biological verification before starting the order. Otherwise, the order dispatch is cancelled, and the order dispatch weight of the service personnel is limited for 2 hours in the future. After completing the service, the service personnel must be biologically verified again to complete the order and record the service order data of this time;

[0029] The service monitoring is the real-time collection of sound and video images in the service environment during the order service of the service personnel, and the watermark is added to the voiceprint and video content, and then the data is uploaded and stored. The data is retained for 90 days.

[0030] Preferably, the service rating optimization module generates the service score of the corresponding service personnel according to the order evaluation of the user on the service personnel. The service rating optimization module generates the service score of the corresponding service personnel according to the decay model of the evaluation in the last 30 days and the total historical evaluation of the service personnel. The score is used as the order dispatch weight of the service personnel.

[0031] Preferably, the segmented weight optimization method of the decay model is as follows:

[0032] To balance the score stability and timeliness, the time axis is divided into 3 decay intervals, and different decay coefficients are used in each interval:

[0033] The user service evaluation in the last 7 days is a high-sensitive index, the decay coefficient is λ=0.05 / day, and the weight calculation formula is as follows:

[0034]

[0035] The user service evaluation of 830 days is a medium-sensitive index, the decay coefficient is λ=0.02 / day, and the weight calculation formula is as follows:

[0036]

[0037] The user service evaluation of 31 days and above is a low-sensitive index, the decay coefficient is λ=0.01 / day, and the weight calculation formula is as follows:

[0038]

[0039] In the above calculation formula, S(t) is the evaluation weight at the current time, and (t-t0) is the number of days from the evaluation generation time to the current time.

[0040] When the service personnel receives ≥3 negative evaluations within 7 days, the decay coefficient of the last 7 days is temporarily increased to accelerate the decay of historical negative evaluations, reflecting the shortcomings of the current service state, and the increase of the decay coefficient corresponds to the decrease of the dispatch weight, encouraging the service personnel to improve their performance, and when the service personnel's negative evaluation rate decreases, the default dispatch weight is restored.

[0041] The present application provides a kind of housekeeping platform dispatch system based on internet technology.It has the following beneficial effects:

[0042] (1) Through multi-dimensional portrait module, service personnel is evaluated from character, ability, service evaluation, order completion, skill and educational background etc., while according to the user's housekeeping order information, including service demand history data, special requirements and constraint conditions etc., user is side-portrayed, then through multi-dimensional dispatch module, not only can the basic service type, time and place etc. demand input by user be analyzed, but also skill demand, service scene demand, special qualification demand etc. implicit demand can be extracted, the demand feature vector of user is generated, through this multi-dimensional portrait and demand extraction, accurate matching of service personnel and user demand is realized, the problem of inaccurate matching caused by single index in traditional dispatch mode is avoided, so as to improve dispatch efficiency and service quality.

[0043] (2) The cooperative scheduling module fuses real-time traffic data, service personnel load rate and order space-time constraints to realize multi-objective optimization of order dispatching. After a user places an order, the service personnel who meet the conditions are filtered out as high-weight order dispatch personnel. When the number of consecutive orders and service time of the service personnel exceeds the workload threshold, the order dispatch weight is automatically reduced to avoid fatigue work. This scheduling mechanism fully considers the working state, skill level and historical performance of the service personnel, and improves the scientific nature of order dispatching and service efficiency while ensuring reasonable working load of the service personnel.

[0044] (3) The service blockchain storage module stores key service data such as service trajectory, service node evidence, biometric verification and service monitoring on the chain, supports cross-chain verification of the identity and health information of the service personnel, realizes "person, certificate and health" three-source verification, ensures the authenticity and non-tamperability of the service data, provides reliable service guarantee for the user, and the service rating optimization module generates a service score according to the user evaluation, according to the decay model of the service personnel's evaluation in the past 30 days and the total evaluation in the past, this rating system not only considers the timeliness of the evaluation, but also balances the score stability through the decay model, at the same time, the processing mechanism of negative evaluation encourages the service personnel to improve their performance, ensures the timeliness and fairness of the scoring system, and further ensures that the service personnel can be better confirmed in the order dispatching process. The weight of the order dispatched, while monitoring the service personnel before, during and after the service, ensuring the safety of the user while also storing evidence for the service process of the service personnel, protecting the safety of the service personnel, and also providing proof support for disputes. BRIEF DESCRIPTION OF DRAWINGS

[0045] Fig. 1 The system framework diagram of the housekeeping platform order dispatching system based on internet technology;

[0046] Fig. 2 The order dispatching flowchart of the housekeeping platform order dispatching system based on internet technology;

[0047] Fig. 3 The service personnel order dispatching weight adjustment flowchart of the housekeeping platform order dispatching system based on internet technology. DETAILED DESCRIPTION

[0048] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of the present application.

[0049] Embodiment 1

[0050] Please refer toFigs. 1-3 The application provides an Internet technology-based housekeeping platform order dispatching system, and the above object is achieved by the following technical scheme: the housekeeping platform order dispatching system comprises:

[0051] User terminal and service terminal: used for users and service personnel to place and accept orders, the user terminal can realize the following operations of the user on the housekeeping platform: placing an order, modifying an order, canceling an order, inquiring about the progress, viewing the details of a recommended service personnel (including a score, a skill label, and a service case), communicating with the service personnel in text and voice, and service evaluation;

[0052] Multi-dimensional portrait module: based on the relevant information of a service personnel, a side portrait of the service personnel is evaluated, and based on the housekeeping order information of a user, a side portrait of the user is evaluated;

[0053] The relevant information of the service personnel specifically includes: basic information is obtained by real-name authentication, including a scanned copy of an ID card, a health certificate, and a skill certificate, and the health status is verified by connecting to a health commission system; service data includes the collection of historical order completion time, customer evaluation text, and service trajectory deviation rate; and additional information includes a personality label generated by a psychological test questionnaire and a professional training record;

[0054] For example, the portrait label of a service personnel A is: patient and meticulous personality, baby care service ability, certified lactation consultant professional skill, service order completion degree and service evaluation, and the system automatically recommends the service personnel A to a "newborn care" order.

[0055] Multi-dimensional order dispatching module: natural language processing is used to analyze the service demand input by a user, and in addition to the basic service type, time and location, the implicit demand can be extracted to generate a demand feature vector of the user;

[0056] Full-process management module: a service personnel records a service trajectory by GPS positioning and a time stamp, and communicates with the service personnel in real time, and an abnormal situation is automatically warned;

[0057] Scenario evaluation module: a virtual service scenario library is constructed, a service personnel regularly participates in scenario simulation training, and the scenarios are regularly updated to match emerging demands;

[0058] Collaborative scheduling module: real-time traffic data, service personnel load rate and order space-time constraints are integrated to realize multi-objective optimization of order dispatching;

[0059] Service blockchain storage module: key service data is stored on a chain, the identity and health information of a service personnel is verified across chains, and "identity, certificate and health" of the service personnel is verified;

[0060] Service rating optimization module: update service staff rating according to user service evaluation, and conduct evaluation credibility analysis to ensure the timeliness and fairness of the rating system.

[0061] Multi-dimensional portrait module for side portrait evaluation of service personnel, including the following: personality, ability, service evaluation, order completion, skill and education background;

[0062] Multi-dimensional portrait module for side portrait evaluation of users, including the following: service demand history data, special requirements and constraint conditions.

[0063] Scenario evaluation module generates virtual service scenarios, and service personnel perform virtual simulation operations through external VR devices;

[0064] The main content of the virtual service scenario is the simulation training of emergency situations and service content, and the service personnel are scored after the training, which is included in the consideration of the service personnel's order dispatch weight.

[0065] Embodiment 2

[0066] Specifically, the extraction of the implicit needs of users in the multi-dimensional order dispatch module includes skill demand, service scenario demand, special qualification demand, time demand, service quality demand, service detail demand and risk control demand;

[0067] In this embodiment,

[0068] Skill demand: implicit requirements in user demand that are not directly listed but require specific professional skills, for example:

[0069] "Clean central air conditioner" is parsed as "refrigeration equipment maintenance qualification and high-altitude operation safety certification";

[0070] "Guide primary school students to do homework" is parsed as "primary school teacher qualification certificate and subject knowledge matching degree (Chinese / mathematics weight increase)";

[0071] "Care for allergic baby" is parsed as "allergen avoidance knowledge and infant care experience weighting";

[0072] Automatically trigger skill library matching to generate "required skills" and "priority skills" tags, and include them in the "skill matching degree" dimension of the dynamic scoring model;

[0073] Service scenario demand: implicit scenario demand based on service object, environment or special conditions, for example:

[0074] "Villa deep cleaning" is parsed as "large residential service experience and equipment carrying requirements (such as industrial vacuum cleaner)";

[0075] “Hospital nurse” is parsed as “hospital environment service experience and hospital infection prevention and control knowledge weight promotion”;

[0076] “Pet companion” is parsed as “basic pet behavior and anti-bite emergency handling ability”;

[0077] Generate scenario-specific screening conditions, exclude personnel without corresponding scenario service records, and adjust distance weight, special scenarios allow cross-zone order allocation;

[0078] Special qualification requirements: implicit qualifications or certifications that users do not explicitly indicate but are required by the industry, such as:

[0079] “Maternal and infant care” is parsed as “maternal care certificate and hepatitis B five-item test report (mandatory verification item)”;

[0080] “Maintenance of certain brand home appliances” is parsed as “brand authorized maintenance certificate and relevant brand maintenance experience”;

[0081] “Old age care” is parsed as “care special training certificate”;

[0082] As a “hard filter condition” in the pre-screening module, those without corresponding qualifications are directly excluded;

[0083] Time requirements: implicit flexibility requirements of users for service time, such as:

[0084] “Clean up on weekends, but time can be negotiated” → parsed as “specific time is uncertain”;

[0085] “Urgently need home service tonight” is parsed as “urgent order label”;

[0086] “Monthly regular cleaning” is parsed as “periodic service requirement (generate long-term order matching strategy)”;

[0087] Convert to “time adaptation dimension” scoring parameters, flexible requirements reduce “distance” weight, and preferentially match time-flexible service personnel;

[0088] Service quality requirements: implicit preferences of users for service personnel's comprehensive ability, such as:

[0089] “Served five-star hotels” is parsed as “service standard evaluation weight increased to 40%”;

[0090] “Few negative reviews” is parsed as “prefer those with zero negative reviews in the past 30 days”;

[0091] “Good communication skills” is parsed as “historical evaluation ‘communication score’ is separately weighted”;

[0092] Dynamically adjust the weight parameters in the scoring model, such as differentiating and increasing the weight of dispatch orders based on the segmented indicators under the "historical evaluation" dimension, such as communication, efficiency, and professionalism;

[0093] Service detail requirements: non-standard service steps or special requirements extracted through semantic analysis, such as:

[0094] "Wear shoe covers when cleaning windows" is interpreted as "service process compliance inspection item";

[0095] "Use of environmentally friendly cleaning agents" is interpreted as "Requirements for carrying equipment / consumables";

[0096] “Communicating in dialect” is interpreted as “language ability matching”;

[0097] Generate a "service task list" and push it to the service personnel as a checkpoint for process monitoring;

[0098] Risk control requirements: Implicit requirements that users do not explicitly specify but need to avoid service risks, such as:

[0099] "Female user at home alone" is interpreted as "Service staff gender preference (optional setting: female priority)"

[0100] "High-value item area" is interpreted as "Background check level increased (mandatory verification of no bad record)"

[0101] "Night service" is interpreted as "safety measures (such as real-time location sharing during service)"

[0102] Trigger the risk control strategy of the pre-screening module, prioritize matching people who have passed advanced background checks, and initiate abnormal warnings during the service process.

[0103] Example 3

[0104] Specifically: The full-process management module monitors service personnel by using the server to obtain device permissions of the service personnel's mobile smart devices. The authorized permissions include but are not limited to GPS permissions, time permissions, communication permissions, camera permissions, and recording permissions;

[0105] The types of key service data stored on the chain in the service blockchain evidence module include: service trajectory, service node evidence, biometric verification and service monitoring. The service data of the service blockchain evidence module is collected and uploaded through the full-process management module.

[0106] The service trajectory is the location and time stamp data of the service personnel, and the data collection interval is once every 5 minutes;

[0107] The service node evidence is the photographic evidence taken by the service personnel upon arriving at the user's address of the service scene before, during, and after the service.

[0108] Biometric verification is that the service personnel performs face, fingerprint, palm vein and voiceprint biometric verification within a specified time after accepting an order. The service personnel must complete the biometric verification before starting the order. Otherwise, the order is canceled, and the service personnel is limited in the order weight for 2 hours in the future. After the service is completed, the service personnel must be biometrically verified again to complete the order and record the service order data of this time;

[0109] Service monitoring is that the service personnel collects the sound and video image in the service environment in real time during the order service process, adds a watermark to the voiceprint and video content, and then uploads the data for storage. The data is retained for 90 days;

[0110] The scheduling method of the collaborative scheduling module is as follows:

[0111] Candidate set: After the user places an order, the collaborative scheduling module filters the service personnel whose service distance is less than 5 km, whose demand skill matches, and whose load rate is less than 70%.

[0112] Priority: Then, the service personnel who meet the conditions are further calculated for a comprehensive score, which is composed of distance score*0.3, scenario simulation score*0.4, and historical praise rate*0.3.

[0113] The service personnel with the top three comprehensive scores are assigned as high-weight service personnel. Then, the top three service personnel are sent to the user end as recommended items. If the user does not specify the service personnel, the order is preferentially assigned to the service personnel with the highest comprehensive score to complete the order.

[0114] The load rate refers to the number of orders accepted by the service personnel in a day and the working time in a day. When the number of consecutive orders accepted by the service personnel and the service time exceed the working load threshold, the order weight of the corresponding service personnel is automatically reduced to avoid fatigue work.

[0115] In this embodiment, the service trajectory evidence includes GPS coordinates (longitude and latitude), timestamp, and moving speed. A record is generated every 5 minutes to prevent the service personnel from leaving work in the middle of the service. If the trajectory interruption exceeds 10 minutes, the service personnel is automatically marked as abnormal.

[0116] Service node evidence: The environment in front of the user's home and the initial state of the service area are photographed before the service. The key steps in the service are photographed. The service results and the user's signature confirmation screen are photographed after the service. The photographing evidence format is a photo, a timestamp, and a geographical location watermark to prevent tampering.

[0117] Biometric verification: within 30 minutes after receiving the order, complete face and fingerprint verification through the APP, if not completed, cancel the order, and lock the order authority for 2 hours, after the service is completed, perform face and voiceprint verification again to confirm the consistency of the service personnel's identity, use live detection technology to prevent photo or video verification fraud, and store the biometric verification data on the chain;

[0118] Service monitoring data: record the communication voice during the service process after user authorization, store the voice after removing sensitive information during recording, and use fixed angle to shoot the service area, generate a segment of video with watermark every 10 minutes, use a buckle type motion camera to fix on the service personnel's clothes, the motion camera has a communication module to upload the shooting video, the uploaded video and sound data are encrypted and stored for 90 days, only authorized by the platform to retrieve in disputes, and automatically deleted after expiration.

[0119] Example 4

[0120] Specifically: the service rating optimization module generates the service score of the corresponding service personnel according to the user's order evaluation of the service personnel, and the service rating optimization module generates the service score of the corresponding service personnel according to the decay model of the service personnel's evaluation in the past 30 days and the total historical evaluation, which is used as the order weight of the service personnel.

[0121] The segmented weight optimization method of the decay model is as follows:

[0122] In order to balance the stability and timeliness of the score, the time axis is divided into three decay intervals, and different decay coefficients are used in each interval:

[0123] The user service evaluation in the past 7 days is used as a high-sensitive index, and the decay coefficient is λ=0.05 / day, and the weight calculation formula is as follows:

[0124]

[0125] 830 days of user service evaluation is used as a medium-sensitive index, and the decay coefficient is λ=0.02 / day, and the weight calculation formula is as follows:

[0126]

[0127] The user service evaluation of 31 days and above is used as a low-sensitive index, and the decay coefficient is λ=0.01 / day, and the weight calculation formula is as follows:

[0128]

[0129] In the above calculation formula, S(t) is the evaluation weight at the current time, and (t-t0) is the number of days from the evaluation time to the current time;

[0130] When the service personnel receives ≥3 negative reviews within 7 days, the attenuation coefficient of the past 7 days is temporarily increased, the attenuation of historical negative reviews is accelerated, the current service state is reflected, the increase of the attenuation coefficient corresponds to the decrease of the order weight, the performance of the service personnel is encouraged to improve, and the default order weight is restored when the negative review rate of the service personnel decreases;

[0131] In this embodiment, the negative review dynamic adjustment mechanism: the trigger condition is ≥3 negative reviews within 7 days, the score ≤2 or the comment content contains negative words, or the complaint content received by the platform, are all considered as negative review indicators. When the negative review condition is met, the attenuation coefficient of the past 7 days is temporarily increased to 0.1 / day, and the old negative review weight is accelerated to decay, and the current negative review content is given priority to adjust the weight;

[0132] The negative review rate is <5% within 15 consecutive days, the default attenuation coefficient is automatically restored, the order weight of the service personnel is restored, the service score is reduced by 0.5 points, the order priority is reduced by 10%, if the current comprehensive score is <3.5 points (10 points), the load rate threshold is reduced to 50% when the candidate set is screened, that is, the order quantity is more strictly limited, the historical negative review record is permanently stored on the chain, but the negative review weight of more than 6 months is attenuated to <10%, to avoid long-term impact, using such order weight judgment standard, the service content quality of the service personnel can be constrained, the service quality of the housekeeping platform is higher, and the service personnel with poor service quality is eliminated, so that the internal structure of the housekeeping platform can present a virtuous circle, thereby increasing the users and income of the housekeeping platform.

[0133] Although the embodiments of the present application have been shown and described, it can be understood by those skilled in the art that various changes, modifications, replacements and variations can be made to these embodiments without departing from the principles and spirits of the present application.

Claims

1. A housekeeping platform dispatching system based on Internet technology, characterized in that: The housekeeping platform dispatching system includes: User and Service Side: used by users and service personnel to place and receive orders; Multi-dimensional portrait module: Based on the relevant information of service personnel, a profile evaluation of service personnel is conducted. Similarly, a profile of users is conducted based on their housekeeping order information. Multi-dimensional dispatch module: This module uses natural language processing to analyze user input service requirements. In addition to basic service type, time, and location, it can extract implicit requirements and generate user demand feature vectors. Full process management module: Service personnel record service tracks through GPS positioning and timestamps, and communicate with service personnel in real time, with automatic warnings for abnormal situations; Scenario Assessment Module: Build a virtual service scenario library, have service personnel regularly participate in scenario simulation training, and regularly update scenarios to match emerging needs; Collaborative dispatch module: Integrates real-time traffic data, service personnel load rate, and order time and space constraints to achieve multi-objective optimized dispatching; Service blockchain evidence storage module: This module stores key service data on-chain, supports cross-chain verification of service personnel's identity and health information, and implements three-source verification of service personnel's "personality, evidence, and health"; Service rating optimization module: Updates service staff ratings based on user service evaluations, and conducts evaluation credibility analysis to ensure the timeliness and fairness of the rating system.

2. The Internet-based housekeeping platform dispatching system according to claim 1, characterized in that: The multi-dimensional portrait module evaluates the service personnel's profile, specifically including the following: Personality, abilities, service evaluation, order completion, skills and educational background; The multi-dimensional portrait module evaluates the user's side portrait, specifically including the following contents: historical data of service demand, special requirements and constraints.

3. The Internet-based housekeeping platform dispatching system according to claim 1, characterized in that: The implicit requirements extracted from users in the multi-dimensional dispatching module specifically include: skill requirements, service scenario requirements, special qualification requirements, time requirements, service quality requirements, service detail requirements and risk control requirements.

4. The Internet-based housekeeping platform dispatching system according to claim 1, characterized in that: The full-process management module monitors service personnel by using the server to obtain device permissions of the service personnel's mobile smart devices. Authorized permissions include but are not limited to GPS permissions, time permissions, communication permissions, camera permissions and recording permissions.

5. The Internet-based housekeeping platform dispatching system according to claim 1 is characterized by: The virtual service scene generated by the scenario assessment module is used by service personnel to perform virtual simulation operations through external VR equipment; The main content of the virtual service scenario is simulation training of emergencies and service content, and the service personnel will be scored after the training is completed, and the score will be included in the weight consideration of dispatching orders for the service personnel.

6. The Internet-based housekeeping platform dispatching system according to claim 1, characterized in that: The scheduling method of the collaborative scheduling module is as follows: Candidate set: After the user places an order, the collaborative scheduling module filters the service personnel's status and selects service personnel with a service distance less than 5km, matching required skills, and a load rate less than 70%; Priority: The service personnel who meet the requirements are then filtered out and their comprehensive scores are calculated. The comprehensive score is composed of: distance score * 0.3, scenario simulation score * 0.4, and historical praise rate * 0.3; The top three service personnel with the highest comprehensive scores will be selected as high-weight dispatchers. The top three matching service personnel will then be sent to the user as recommendations. If the user does not specify a service personnel, the order will be assigned to the service personnel with the highest comprehensive score first. Among them, the load rate refers to the number of orders received by the service personnel on that day and the working hours on that day. When the number of consecutive orders received by the service personnel and the service time exceed the workload threshold, the corresponding service personnel's order dispatch weight will be automatically reduced to avoid fatigue work.

7. The Internet-based housekeeping platform dispatching system according to claim 1, characterized in that: The types of key service data stored on the chain in the service blockchain evidence module include: service trajectory, service node evidence, biometric verification and service monitoring. The service data of the service blockchain evidence module is collected and uploaded through the full-process management module.

8. The Internet-based housekeeping platform dispatching system according to claim 7, characterized in that: The service trajectory is the location and time stamp data of the service personnel, and the data collection interval is once every 5 minutes; The service node evidence is the photographic evidence taken by the service personnel upon arrival at the user's address of the pre-service status, service process status, and post-service status of the service scene; Biometric verification involves a service representative performing facial, fingerprint, palm vein, and voiceprint verification within a specified timeframe after accepting an order. The service representative can begin the order after completing biometric verification. Otherwise, if verification is not completed, the order will be canceled and their order weight will be limited for the next two hours. After the service is completed, the service representative must undergo biometric verification again to complete the order and record the service order data. The service monitoring is that the service personnel collect the sound and video images in the service environment in real time during the order service process, add watermarks to the voiceprint and video content, and then upload the data for storage. The data is retained for 90 days.

9. The Internet-based housekeeping platform dispatching system according to claim 1, characterized in that: The service rating optimization module generates a service score for the corresponding service personnel based on the user's order evaluation of the service personnel and the decay model of the service personnel's evaluation in the past 30 days and the historical total evaluation. The score is used as a weight consideration for dispatching orders to the service personnel.

10. The Internet-based housekeeping platform dispatching system according to claim 9, characterized in that: The specific method for optimizing the segmented weights of the attenuation model is as follows: To balance the stability and timeliness of ratings, the timeline is divided into three decay intervals, each with a different decay coefficient: The user service evaluation in the past 7 days is a highly sensitive indicator with an attenuation coefficient of λ = 0.05 / day. The weight calculation formula is as follows: The 830-day user service evaluation is a medium-sensitive indicator with an attenuation coefficient of λ = 0.02 / day. The weight calculation formula is as follows: User service evaluation for 31 days or more is used as a low-sensitivity indicator, with an attenuation coefficient of λ = 0.01 / day. The weight calculation formula is as follows: In the above calculation formula, S(t) is the evaluation weight at the current moment, and (t-t0) is the number of days from the time the evaluation was generated to the current moment; When a service staff receives ≥3 negative reviews within 7 days, the attenuation coefficient for the past 7 days will be automatically increased temporarily to accelerate the decay of historical negative reviews and reflect the shortcomings of the current service status. The increase in the attenuation coefficient will correspond to a decrease in the dispatch weight, encouraging service staff to improve their performance. Once the negative review rate of the service staff decreases, the default dispatch weight will be restored.

Images