Management method for educational institution management system and storage medium

Abstract

The invention relates to the field of educational institution management, in particular to a management method for an educational institution management system and a storage medium. The method comprises the steps that positioning data and pre-filling information of parents are collected through a promotion link, a management system executes primary matching and secondary matching according to the positioning data and the pre-filling information, the primary matching is matching based on the positioning data, and the secondary matching is matching based on the pre-filling information. Weight distribution of primary matching and secondary matching is adjusted through a recommendation strategy optimization algorithm according to historical recommendation samples and payment conversion success rates, an education institution recommendation list is generated and pushed to a parent user side, and parents go to an education institution according to the education institution recommendation list to complete sign-in and payment. The closed-loop management from the promotion link to the recommendation mechanism to the payment conversion is realized, the management efficiency of the management system of the education mechanism is improved, and the success rate of the payment conversion is improved through a matching mechanism and an optimization algorithm.

Description

Technical Field

[0001] The present application relates to the field of educational institution management, and in particular to a management method and storage medium for an educational institution management system. Background Art

[0002] In the education industry, most businesses manage parents and students through social software, paper documents, and back-end systems. This management method has the problem of data isolation, which easily leads to information confusion, low efficiency, and requires high manpower costs.

[0003] Existing management systems for educational institutions can manage the entire process from promotion to payment to course management. However, these management systems are often migrated from management systems or recommendation systems in the e-commerce field. They are not well adapted for educational institutions, cannot support the complete management cycle of educational institutions, and cannot meet the educational institutions' needs for management efficiency and payment conversion success rate. In summary, there are no existing management systems and related management methods suitable for educational institutions. Summary of the Invention

[0004] In order to solve the problem of the lack of management systems and management methods applicable to educational institutions in the prior art, the present application provides a management method and storage medium for an educational institution management system. The method provided in the present application includes the following steps: The parent's location data and pre-filled information are collected through the promotion link, and the management system performs a primary match and a secondary match based on the location data and the pre-filled information, wherein the primary match is based on the location data and the secondary match is based on the pre-filled information; The recommendation strategy optimization algorithm adjusts the weight distribution of the first match and the second match according to historical recommendation samples and payment conversion success rate, generates an educational institution recommendation list and pushes it to the parent user end, and the parents go to the educational institution to complete the sign-in and payment according to the educational institution recommendation list. The recommendation strategy optimization algorithm regards parents who go to the educational institution to sign in and pay as the first category of samples, parents who go to the educational institution but do not pay as the second category of samples, and parents who do not go to the educational institution as the third category of samples, and optimizes the weight distribution strategy of the first match and the second match based on the first category of samples, the second category of samples and the third category of samples.

[0005] Specifically, when performing the first matching, the coordinates of the educational institution registered in the management system are obtained, the parent's positioning data and the coordinates are matched, and the parent's positioning data is corrected using a hidden Markov model during matching. The observation probability of the hidden Markov model is calculated based on the Euclidean distance between the trajectory of the parent's positioning data and the topological nodes in the road network. The transition probability of the hidden Markov model is obtained based on path connectivity training and is input into the recommendation strategy optimization algorithm as a reliability index of the first matching. When the confidence level is lower than the threshold, the weight allocation ratio of the first matching is reduced, and the weight ratio of the second matching is increased.

[0006] Specifically, when the hidden Markov model is used to modify the road network, the transition probability satisfies: , in, s i 、 s j are adjacent road segment nodes, d ij is the actual distance between road sections, m is the average distance of the road network, obtained by training with historical trajectory data, and distinguishing between two training sets during the peak and non-peak hours after school. The corresponding training set is selected according to the time when the parent user terminal accesses the promotion link. m value.

[0007] Specifically, the pre-filled information includes the student's age and course preferences. When performing the secondary matching, the course settings and age adaptation range in the educational institution registration information are obtained, the student's age is matched with the age adaptation range, and the course preferences are matched with the course settings.

[0008] Specifically, when performing the secondary matching, a first deviation value is calculated between the student's age and the age adaptation range, a second deviation value is calculated between the course preference and the course setting, and the service rating of the educational institution is obtained, and a matching degree is calculated for the first deviation value, the second deviation value and the service rating allocation weight.

[0009] Specifically, the recommendation strategy optimization algorithm adjusts the weight ratio of the first matching and the second matching by minimizing the weighted cross entropy loss function of the positive samples and the negative samples, wherein the weighted cross entropy loss function is a function that reflects the distribution of the positive samples and the negative samples, the positive samples are the set of the first category samples, and the negative samples are the set of the second category samples and the third category samples.

[0010] Specifically, when the recommendation strategy optimization algorithm optimizes the weight allocation strategy of the first matching and the second matching, different weight values ​​in the weighted cross entropy loss function are assigned to the first category samples, the second category samples and the third category samples, and a time attenuation coefficient is added to the historical recommended samples according to the time sequence.

[0011] Specifically, the management method further includes: When parents pay, their signature is entered and the image of the signature is enhanced using the pix2pix model. The pix2pix model includes a self-attention module and a folding operation. The self-attention module is used to enhance the edge clarity of calligraphy characters, and the folding operation is used to simulate the ink diffusion effect of writing on paper.

[0012] Specifically, the management method further includes: After the parents complete the payment, the system generates a student file and stores it in association with the parent's signature when paying for student management and renewal management.

[0013] The present application also provides a computer-readable storage medium storing a computer program, which implements the steps of the method described above when executed by a processor.

[0014] This application has the following technical effects: It realizes closed-loop management from promotion links to recommended institutions and then to payment conversion, improves the management efficiency of the management system of educational institutions, and forms algorithmic support for the educational institution recommendation list through matching mechanism and optimization algorithm, thereby improving the success rate of payment conversion. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] By reading the detailed description below with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present application will become readily understood. In the accompanying drawings, several embodiments of the present application are shown in an exemplary and non-limiting manner, and the same or corresponding numbers represent the same or corresponding parts.

[0016] Figure 1 This is a flow chart of a management method for an educational institution management system in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application.

[0018] In the wave of digitalization, educational institutions often build a series of management systems in order to improve their own management level. These management systems involve multiple business links such as accessing WeChat promotion links, recommending educational institutions, and managing students after payment, to achieve management of the entire process. Among these business links, the link that is most relevant to the profitability of educational institutions is recommending educational institutions, because it will directly affect the willingness of parents to go to educational institutions and the degree of matching between parents and educational institutions, and thus affect the success rate of conversion payment. Based on this, this embodiment aims to propose a management method for a management system of an educational institution, which can not only meet the management needs of the entire process, but also effectively improve the conversion ability of parents' payments in the link of recommending educational institutions. Figure 1 As shown, the management method of this embodiment includes the following steps: The parent's location data and pre-filled information are collected through the promotional link. The management system performs primary and secondary matching based on the location data and pre-filled information. The primary matching is based on the location data, and the secondary matching is based on the pre-filled information. Through the recommendation strategy optimization algorithm, the weight distribution of the first match and the second match is adjusted according to the historical recommendation samples and the payment conversion success rate, and a recommendation list of educational institutions is generated and pushed to the parent user end. Parents go to the educational institution to complete the sign-in and payment according to the recommendation list of educational institutions. The recommendation strategy optimization algorithm regards parents who go to the educational institution to sign in and pay as the first category of samples, parents who go to the educational institution but do not pay as the second category of samples, and parents who do not go to the educational institution as the third category of samples, and optimizes the weight distribution strategy of the first match and the second match based on the first category samples, the second category samples and the third category samples.

[0019] Specifically, in this embodiment, when parents click on a promotional link released by an educational institution through a social platform such as WeChat, the system will first call the GPS module of the mobile terminal to obtain real-time positioning coordinates, and at the same time guide the parents to fill in core pre-filled information such as the student's age, desired course type, and available time period. After these data are uploaded to the cloud server through an encrypted transmission protocol, the operation process of the map matching engine is triggered. Since the parent's movement trajectory is continuous and constrained, and the GPS / base station positioning data has random noise, this embodiment uses a hidden Markov model (HMM) to express the parent's state transition probability and observation probability, match the parent's positioning data and coordinates, and use the hidden Markov model to correct the parent's positioning data during matching. The observation probability of the hidden Markov model is calculated based on the Euclidean distance between the trajectory of the parent's positioning data and the topological nodes in the road network. The transition probability of the hidden Markov model is obtained based on path connectivity training. Specifically, the HMM model models the topological structure of the urban road network as a state space, where each road section node corresponds to a hidden state, and the observation sequence is the coordinate point of the parent's movement trajectory. The transition probability matrix is ​​obtained through training of historical traffic flow data, and the transition probability satisfies: , in, s i 、 s j are adjacent road segment nodes, d ij is the actual distance between road sections, m is the average distance of the road network, which is obtained by training with historical trajectory data. During training, two training sets are distinguished: peak time after school and non-peak time. The corresponding training set is selected according to the time when the parent user accesses the promotion link. m Normally, the peak time for school dismissal is from 15:00 to 18:00 on weekdays. m The value is smaller than that obtained from the training set during off-peak hours. m value. At the same time, m The value can also be used as a reliability indicator for a single match and input into the recommendation strategy optimization algorithm. When the confidence level falls below a threshold, the weight allocation ratio for the first match is reduced, while the weight allocation ratio for the second match is increased. This effectively addresses the problem of institution matching deviation in educational scenarios caused by the complexity of urban road networks and positioning signal drift. It also distinguishes historical trajectory data during peak and off-peak hours, forming a dynamic compensation mechanism. This significantly improves the matching accuracy of educational institution recommendation lists during peak hours in densely populated areas. The improved accuracy of path correction also increases the secondary conversion rate of users who have not paid their fees, breaking through the application bottleneck of traditional recommendation algorithms in complex urban education scenarios.

[0020] After completing one match, the system enters the second matching stage. The core goal at this time is to deeply associate the student's age and course preferences in the parent's pre-filled information with the course settings, age adaptation range, and service ratings in the educational institution's registration information. For example, a parent pre-filled the student's age as 6 years old and the course preference as "English Enlightenment". The system will give priority to matching educational institutions with an age adaptation range of 5-7 years old and that offer "Children's English" courses. In order to achieve accurate matching, the system calculates the first deviation value between the student's age and the institution's age adaptation range. The specific formula is: in, A 机构中值 is the midpoint value of the institution’s age adaptation range, A 机构范围is the span of the age range. Similarly, course preference matching uses natural language processing technology to extract keywords (such as "enlightenment" and "art") and calculates cosine similarity with the labels of the institution's course settings to obtain the second deviation value Δ2. Service scoring directly uses the institution's historical scoring data (such as parent evaluations and course completion rates) to calculate its third deviation value Δ3 from the industry benchmark score. Finally, the system generates a comprehensive matching score based on a dynamic weight allocation mechanism: The initial weights are set to w 1=0.3, w 2=0.25, w 3=0.2, and the remaining weights are used to reserve expansion parameters.

[0021] In the weight allocation process, the recommendation strategy optimization algorithm uses a double forgetting factor mechanism to dynamically adjust the weight ratio of geographic matching and demand matching. In the initial stage, the geographic matching weight is set to 0.6 and the demand matching weight is 0.4, which are updated every 24 hours based on the latest conversion data. In terms of continuous optimization of the recommendation strategy, the system starts model retraining every time 500 new samples are accumulated. During training, a cost-sensitive learning method is used for the three types of samples: the weight of the first type of sample that has successfully paid is set to 1.5, the weight of the second type of sample that has visited but not paid is 0.8, and the weight of the third type of sample that has not visited is 0.3. The loss function is modified to: , in, y ∈{0, 1} represents the true label (positive sample or negative sample), p is the model prediction probability, w + and w − are the weight coefficients of positive and negative samples respectively.

[0022] When parents visit an educational institution based on the recommended list, the system automatically signs them in through interaction between a Bluetooth beacon and the parent's mobile app. The payment process is triggered the moment the sign-in is successful, and parents can complete their handwritten signatures on the electronic agreement. The signature processing module uses an improved pix2pix generative adversarial network, innovatively adding a folding operation subnet to the generator network. This subnet decomposes the input strokes into two feature channels, horizontal and vertical strokes, and simulates the force changes during real writing through a deformable convolutional layer. The discriminator network uses a multi-scale PatchGAN structure to calculate the adversarial loss in a local area of ​​256×256 pixels, effectively addressing the legal risks of handwriting distortion in traditional electronic signatures.

[0023] After payment is completed, the system automatically generates a student profile and encrypts the signature image using a hash algorithm, storing it in association with the student's information. In subsequent course management, the teacher's app can access historical check-in records, course progress, and other data from the student profile in real time.

[0024] In summary, this example integrates high-precision map matching, dynamic weighted recommendation strategies, and intelligent signature processing technology to build a complete and efficient educational institution management system. This system not only solves the data silo problem of traditional management methods, but also significantly improves payment conversion rates through algorithm-driven precision recommendations, providing reliable technical support for the digital transformation of the education industry.

[0025] Obviously, the embodiments described above are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0026] It should be understood that when the terms "first," "second," etc. are used in the claims, specification, and drawings of this application, they are only used to distinguish different objects, rather than to describe a specific order. The terms "comprise" and "comprising" used in the specification and claims of this application indicate the presence of the 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.

Claims

1. A management method for an educational institution management system, characterized in that: The following steps are involved: The parent's location data and pre-filled information are collected through the promotion link, and the management system performs a primary match and a secondary match based on the location data and the pre-filled information, wherein the primary match is based on the location data and the secondary match is based on the pre-filled information; The recommendation strategy optimization algorithm adjusts the weight distribution of the first match and the second match according to historical recommendation samples and payment conversion success rate, generates an educational institution recommendation list and pushes it to the parent user end, and the parents go to the educational institution to complete the sign-in and payment according to the educational institution recommendation list. The recommendation strategy optimization algorithm regards parents who go to the educational institution to sign in and pay as the first category of samples, parents who go to the educational institution but do not pay as the second category of samples, and parents who do not go to the educational institution as the third category of samples, and optimizes the weight distribution strategy of the first match and the second match based on the first category of samples, the second category of samples and the third category of samples.

2. The management method according to claim 1, characterized in that: When performing the primary matching, the coordinates of the educational institution registered in the management system are obtained, the parent's positioning data and the coordinates are matched, and the parent's positioning data is corrected using a hidden Markov model during matching. The observation probability of the hidden Markov model is calculated based on the Euclidean distance between the trajectory of the parent's positioning data and the topological nodes in the road network. The transition probability of the hidden Markov model is obtained based on path connectivity training and is input into the recommendation strategy optimization algorithm as a reliability indicator of the primary matching. When the confidence level is lower than a threshold, the weight allocation ratio of the primary matching is reduced, and the weight ratio of the secondary matching is increased.

3. The management method according to claim 2, characterized in that: When the hidden Markov model is used to modify the road network, the transition probability satisfies: ， in, s i 、 s j are adjacent road segment nodes, d ij is the actual distance between road sections, μ is the average distance of the road network, obtained by training with historical trajectory data, and distinguishing between two training sets during the peak and non-peak hours after school. The corresponding training set is selected according to the time when the parent user terminal accesses the promotion link. μ value.

4. The management method according to claim 1, characterized in that: The pre-filled information includes the student's age and course preferences. When performing the secondary matching, the course settings and age adaptation range in the educational institution registration information are obtained, the student's age is matched with the age adaptation range, and the course preferences are matched with the course settings.

5. The management method according to claim 4, characterized in that: When performing the secondary matching, a first deviation value is calculated between the student's age and the age adaptation range, a second deviation value is calculated between the course preference and the course setting, and the service rating of the educational institution is obtained, and a matching degree is calculated for the first deviation value, the second deviation value and the service rating allocation weight.

6. The management method according to claim 1, characterized in that: The recommendation strategy optimization algorithm adjusts the weight ratio of the first matching and the second matching by minimizing the weighted cross entropy loss function of the positive samples and the negative samples, wherein the weighted cross entropy loss function is a function reflecting the distribution of the positive samples and the negative samples, the positive samples are the set of the first category samples, and the negative samples are the set of the second category samples and the third category samples.

7. The management method according to claim 6, characterized in that: When the recommendation strategy optimization algorithm optimizes the weight allocation strategy of the first matching and the second matching, different weight values ​​in the weighted cross entropy loss function are assigned to the first category samples, the second category samples, and the third category samples, and a time decay coefficient is added to the historical recommended samples according to the time sequence.

8. The management method according to claim 1, characterized in that: The management method further comprises: When parents pay, their signature is entered and the image of the signature is enhanced using the pix2pix model. The pix2pix model includes a self-attention module and a folding operation. The self-attention module is used to enhance the edge clarity of calligraphy characters, and the folding operation is used to simulate the ink diffusion effect of writing on paper.

9. The management method according to claim 1, characterized in that: The management method further comprises: After the parents complete the payment, the system generates a student file and stores it in association with the parent's signature when paying for student management and renewal management.

10. A computer-readable storage medium storing a computer program, characterized in that: When the program is executed by a processor, the steps of the management method according to any one of claims 1 to 9 are implemented.

Images