A method for building a campus service platform system based on mobile terminals

Abstract

The present invention relates to a method for building a campus service platform based on a mobile terminal, and belongs to the technical field of teaching methods. The technical scheme of the present invention is: to establish a logistic regression model, to establish an information search engine model, to use the bandpass filtering theory of microwave conversion to generate a time series, to extract campus data information, and to use the extraction results as the basis for designing a campus news platform. The beneficial effects of the present invention are: by establishing a campus news platform based on information crawling and information mining technology, personalized information transmission for individuals or groups of people is realized, different transmission methods are adopted according to different scheduling requirements, and on the premise of ensuring that information is delivered to users in a timely and effective manner, the network connection time is reduced, and the optimal control of power loss and network traffic is achieved. The system runs smoothly, the user interface is friendly, and each functional module of the system runs normally, and the operation results are correct, which can meet the daily needs of system use.

Description

Technical Field

[0001] The invention relates to a method for building a campus service platform system based on a mobile terminal, and belongs to the technical field of teaching methods. Background Art

[0002] At present, most colleges and universities have independent campus portal websites. The functions of the campus websites of various colleges and universities are relatively simple, mainly including the promotion of school characteristics, major introduction, and teacher introduction. The campus websites of some colleges and universities have more comprehensive functions, adding data collection and teaching management systems. However, there is a lack of interaction between teachers and students, and they cannot meet the multi-faceted application needs of college students. Summary of the invention

[0003] The purpose of the present invention is to provide a method for building a campus service platform system based on a mobile terminal. By establishing a campus news platform based on information crawling and information mining technology, personalized information transmission for individuals or groups of people can be realized. Different transmission methods are adopted according to different scheduling requirements. Under the premise of ensuring that information is delivered to users in a timely and effective manner, the network connection time is reduced, and optimal control of power loss and network traffic is achieved. The system runs smoothly, the user interface is friendly, the various functional modules of the system run normally, the operating results are correct, and the daily needs of system use can be met, effectively solving the above-mentioned problems existing in the background technology.

[0004] The technical solution of the present invention is: a method for building a campus service platform system based on a mobile terminal, comprising the following steps:

[0005] (1) Establish a logistic regression model;

[0006] (2) Establishing an information search engine model, setting input and output data information samples of the campus theme part, adapting the data information samples to the campus theme part and the structure and parameters of the fuzzy neural network, and establishing an information search engine model using the campus theme based on the fuzzy neural network;

[0007] (3) Use the bandpass filtering theory of microwave conversion to generate time series, extract campus data information, and use the extraction results as the basis for designing a campus news platform.

[0008] In step (1), web information is represented as a feature vector of an independent variable, and the logistic regression model is described by the following formula:

[0009]

[0010] Among them, g(x)=β0+β1x1+β2x2+…+β n x nIt is a category scalar of data information, which combines campus themes and targets to find the parameter estimate of the maximum likelihood function.

[0011] The step (2) comprises the following steps:

[0012] (a) Starting from the first sample (x1, y1) of campus theme data information, a cluster center x is established based on x1 -1 , and set A1 = y1, B1 = 1; select a radius r, in which B l (l=1,2,…,m) records the number of input and output samples of data information cluster l;

[0013] (b) Consider the kth data information sample (x k ,y k ), if there are clusters m, the centers are x -1 ~x -m , we get the following formula:

[0014]

[0015] The following formula is used to obtain the data information of the smallest cluster sample of campus topics:

[0016]

[0017] In the above formula, cluster p is the possible nearest neighbor cluster of campus theme data information sample k;

[0018] (c) Use the following formula to make the data information sample (x k ,y k ) as a new data information clustering:

[0019]

[0020] Divide the data information sample k of the campus theme into cluster p, and then the following formula needs to be satisfied:

[0021]

[0022] In the formula, represents the total input value of the data information samples of campus topics belonging to cluster p;

[0023] (d) The learning sample data information is input into the fuzzy neural network, and the total error of the data information sample of the fuzzy neural network is obtained. The following formula is used to describe it:

[0024]

[0025] In the above formula, s represents the total number of data information samples of campus topics. When the learning sample input is x k When f k is the output value of the neural network;

[0026] (e) The following formula is used to derive ERR and to adjust the fuzzy neural network structure and parameter σ adaptively:

[0027]

[0028] In the above formula, q is the learning law of data information with campus as the theme;

[0029] (f) Use the following formula to establish a search engine model for campus subject data information based on fuzzy neural network:

[0030]

[0031] In the formula, and is the parameter of the neural network, n is the number of components contained in the data information sample represented by the campus theme, and m is the sum of the clusters of data information with campus as the theme.

[0032] In step (3), the function ψ is set to a function cluster {ψ ab}, expressed by the following formula:

[0033]

[0034] In the above formula, ψ represents the basic fundamental wave, which comes from the function family {ψ ab} is called a microwave, a is the scale parameter of the campus data information, and b represents the location parameter of the campus data information;

[0035] Set the microwave generating function {ψ ab}, the expansion and shift factors are {a,b} respectively, and the continuous microwave function of the data signal f(t) in the smart campus is:

[0036]

[0037] In the above formula, It is the time-frequency characteristic of microwaves;

[0038] The equivalent frequency domain of the data information in the above formula is divided into the following formula:

[0039]

[0040] In the above formula, F(w) and ψ(aw) are f(t) and ψab The Fourier transform of (t) is given by W f The equivalent frequency domain represented by (a, b) is defined as the time series of data information.

[0041] The beneficial effects of the present invention are as follows: by establishing a campus news platform based on information crawling and information mining technology, personalized information transmission for individuals or groups is realized, different transmission methods are adopted according to different scheduling requirements, and the network connection time is reduced while ensuring that information is delivered to users in a timely and effective manner, and optimal control of power loss and network traffic is achieved. The system runs smoothly, the user interface is friendly, the various functional modules of the system run normally, the operating results are correct, and the daily needs of system use can be met. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a functional schematic diagram of the campus service platform system of the present invention;

[0043] Figure 2 This is a functional side architecture diagram of the campus service platform system of the present invention;

[0044] Figure 3 This is the main interface diagram of the campus service platform system of the present invention;

[0045] Figure 4 This is a campus news browsing page diagram of the campus service platform system of the present invention;

[0046] Figure 5 It is a flow chart of browsing news in the campus service platform system of the present invention. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solutions and advantages of the invention implementation cases clearer, the technical solutions in the implementation cases of the present invention will be clearly and completely described below in conjunction with the drawings in the implementation cases. Obviously, the implementation cases described are only a small part of the implementation cases of the present invention, rather than all the implementation cases. Based on the implementation cases in the present invention, all other implementation cases obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0048] A method for building a campus service platform system based on a mobile terminal comprises the following steps:

[0049] (1) Establish a logistic regression model;

[0050] (2) Establishing an information search engine model, setting input and output data information samples of the campus theme part, adapting the data information samples to the campus theme part and the structure and parameters of the fuzzy neural network, and establishing an information search engine model using the campus theme based on the fuzzy neural network;

[0051] (3) Use the bandpass filtering theory of microwave conversion to generate time series, extract campus data information, and use the extraction results as the basis for designing a campus news platform.

[0052] In step (1), web information is represented as a feature vector of an independent variable, and the logistic regression model is described by the following formula:

[0053]

[0054] Among them, g(x)=β0+β1x1+β2x2+…+β n x n It is a category scalar of data information, which combines campus themes and targets to find the parameter estimate of the maximum likelihood function.

[0055] The step (2) comprises the following steps:

[0056] (a) Starting from the first sample (x1, y1) of campus theme data information, a cluster center x is established based on x1 -1 , and set A1 = y1, B1 = 1; select a radius r, in which B l (l=1,2,…,m) records the number of input and output samples of data information cluster l;

[0057] (b) Consider the kth data information sample (x k ,y k ), if there are clusters m, the centers are x -1 ~x -m , we get the following formula:

[0058]

[0059] The following formula is used to obtain the data information of the smallest cluster sample of campus topics:

[0060]

[0061] In the above formula, cluster p is the possible nearest neighbor cluster of campus theme data information sample k;

[0062] (c) Use the following formula to make the data information sample (x k ,y k ) as a new data information clustering:

[0063]

[0064] Divide the data information sample k of the campus theme into cluster p, and then the following formula needs to be satisfied:

[0065]

[0066] In the formula, represents the total input value of the data information samples of campus topics belonging to cluster p;

[0067] (d) The learning sample data information is input into the fuzzy neural network, and the total error of the data information sample of the fuzzy neural network is obtained. The following formula is used to describe it:

[0068]

[0069] In the above formula, s represents the total number of data information samples of campus topics. When the learning sample input is x k When f k is the output value of the neural network;

[0070] (e) The following formula is used to derive ERR and to adjust the fuzzy neural network structure and parameter σ adaptively:

[0071]

[0072] In the above formula, q is the learning law of data information with campus as the theme;

[0073] (f) Use the following formula to establish a search engine model for campus subject data information based on fuzzy neural network:

[0074]

[0075] In the formula, and is the parameter of the neural network, n is the number of components contained in the data information sample represented by the campus theme, and m is the sum of the clusters of data information with campus as the theme.

[0076] In step (3), the function ψ is set to a function cluster {ψ ab}, expressed by the following formula:

[0077]

[0078] In the above formula, ψ represents the basic fundamental wave, which comes from the function family {ψ ab} is called a microwave, a is the scale parameter of the campus data information, and b represents the location parameter of the campus data information;

[0079] Set the microwave generating function {ψ ab}, the expansion and shift factors are {a,b} respectively, and the continuous microwave function of the data signal f(t) in the smart campus is:

[0080]

[0081] In the above formula, It is the time-frequency characteristic of microwaves;

[0082] The equivalent frequency domain of the data information in the above formula is divided into the following formula:

[0083]

[0084] In the above formula, F(w) and ψ(aw) are f(t) and ψ ab The Fourier transform of (t) is given by W f The equivalent frequency domain represented by (a, b) is defined as the time series of data information.

[0085] In actual applications, the campus service platform system is built for the education industry, providing colleges and universities with comprehensive services such as classroom sign-in, teacher-student communication, etc., including servers, management terminals, teacher terminals and student terminals.

[0086] The server includes a data query and analysis unit, a data storage unit and a teaching resource editing unit. The data query and analysis unit is used to query and analyze user data resources to realize personalized push service of important information. The data storage unit is used to store accounting data and chat data, etc., to improve the interactivity of the campus service platform and the security of the campus service platform for mobile terminals. The teaching resource editing unit allows users to edit teaching resources.

[0087] The management end includes user management unit, sign-in management unit, class management unit and communication interaction unit. The teacher end includes user login unit, class management unit and user accounting unit.

[0088] The student terminal includes a user login unit, a user sign-in unit and a user accounting unit.

[0089] Explore the use of information crawlers and user behavior analysis to achieve personalized push services for important information; study how to achieve the development and positioning of campus maps and life accounting. Improve the interactivity of the campus service platform through the development of campus chat interactions, campus canteens, and campus news forums. Realize the intelligence of school teaching management through the development of mobile terminals for campus teaching affairs systems and student attendance management systems. Use unified authentication technology for digital signatures to improve the security of the campus service platform on mobile terminals. Expand the scope of use of the campus service platform by establishing an API interface.

[0090] 1) Design principles of campus news platform

[0091] In view of the scattered campus information, a campus news platform based on information crawling and information mining technology was established to realize the personalized information transmission of individuals or groups. In order to improve the transmission performance of the server, different transmission methods are adopted according to different scheduling requirements. Under the premise of ensuring that the information is delivered to the user in a timely and effective manner, the network connection time can be reduced to achieve the optimal control of power loss and network traffic. The specific steps are as follows:

[0092] If web information is represented as a feature vector of independent variables, the following formula is used to describe the logistic regression model:

[0093]

[0094] Among them, g(x)=β0+β1x1+β2x2+…+β n x n It is a categorical scalar of data information, which combines campus topics and targets to find the parameter estimate of the maximum likelihood function.

[0095] 2) Establishment of campus subject information search engine model

[0096] Set the input and output data information samples (x(k), y(k)) of the existing part of the campus subject, adapt the data information samples to the structure and parameters of the campus subject and the fuzzy neural network, and use the campus subject based on the fuzzy neural network to establish an information search engine model:

[0097] (a) Starting from the first sample (x1, y1) of campus theme data information, a cluster center x is established based on x1 -1 , and set A1 = y1, B1 = 1. A radius r is selected, in which B l (l=1,2,…,m) records the number of input and output samples of data information cluster l.

[0098] (b) Consider the kth data sample (x k ,y k ), if there are clusters m, the centers are x -1 ~x -m , we can get the following formula:

[0099]

[0100] The following formula is used to obtain the data information of the smallest cluster sample of campus topics:

[0101]

[0102] In the above formula, cluster p is the possible nearest neighbor cluster of campus-themed data information sample k.

[0103] (c) Use the following formula to make the data information sample (x k ,y k ) as a new data information clustering:

[0104]

[0105] In the optimization design process of the campus news platform, the data information samples k of the campus theme are divided into clusters p, and then the following formula needs to be satisfied:

[0106]

[0107] In the formula, Represents the total input value of the data information samples of campus topics belonging to cluster p.

[0108] (d) The learning sample data information is input into the fuzzy neural network, and the total error of the data information sample of the fuzzy neural network is obtained. The following formula can be used to describe it:

[0109]

[0110] In the above formula, s represents the total number of data information samples of campus topics. When the learning sample input is x k When f k is the output value of the neural network.

[0111] (e) The following formula is used to derive ERR and to adjust the fuzzy neural network structure and parameter σ adaptively:

[0112]

[0113] In the above formula, q is the learning law of campus-themed data, and σ is the variance of campus data samples.

[0114] (f) Use the following formula to establish a search engine model for campus subject data information based on fuzzy neural network:

[0115]

[0116] In the formula, and is the parameter of the neural network, n is the number of components contained in the data information sample represented by the campus theme, and m is the sum of the clusters of data information with campus as the theme.

[0117] 3) Implementation of campus news platform design optimization

[0118] By scaling and translating, the function ψ can be set to a function family {ψab}, expressed by the following formula:

[0119]

[0120] In the above formula, ψ represents the basic fundamental wave, which comes from the function family {ψ ab} is called microwave, a is the scale parameter of the campus data information, and b represents the location parameter of the campus data information.

[0121] The microwave generating function {ψ ab}, the expansion and shift factors are {a,b} respectively, and the continuous microwave function of the data signal f(t) in the smart campus is:

[0122]

[0123] In the above formula, It is the time-frequency characteristic of microwave.

[0124] The equivalent frequency domain of the data information in the above formula is divided into the following formula:

[0125]

[0126] In the above formula, F(w) and ψ(aw) are f(t) and ψ ab The Fourier transform of (t) is given by W f The equivalent frequency domain represented by (a, b) is defined as the time series of data information.

[0127] The present invention uses an improved algorithm and a traditional algorithm to design a campus news platform. In different experimental stages, the two algorithms are used to compare the error rate, accuracy, stability of data information search and search time.

[0128] The improved algorithm is better than the traditional algorithm. The improved algorithm and fuzzy neural network algorithm are used to study data information samples. The structure and parameters of the fuzzy neural network are adjusted to establish the model of the campus theme information search engine. On this basis, the bandpass filter theory of microwave conversion is used to generate time series to extract campus data information. The extraction results are used as the basis for designing the campus news platform to ensure the effectiveness of the improved algorithm in the design of the campus news platform.

Claims

1. A method for building a campus service platform system based on a mobile terminal, characterized in that The following steps are involved: (1) Establish a logistic regression model; (2) Establishing an information search engine model, setting input and output data information samples of the campus theme part, adapting the data information samples to the campus theme part and the structure and parameters of the fuzzy neural network, and establishing an information search engine model using the campus theme based on the fuzzy neural network; (3) Use the bandpass filtering theory of microwave conversion to generate time series, extract campus data information, and use the extraction results as the basis for designing a campus news platform.

2. According to the method for building a campus service platform system based on a mobile terminal according to claim 1, it is characterized in that: In step (1), web information is represented as a feature vector of an independent variable, and the logistic regression model is described by the following formula: Among them, g(x)=β0+β1x1+β2x2+…+β n x n It is a category scalar of data information, which combines campus themes and targets to find the parameter estimate of the maximum likelihood function.

3. According to the method for building a campus service platform system based on a mobile terminal according to claim 1, it is characterized in that: The step (2) comprises the following steps: (a) Starting from the first sample (x1, y1) of campus theme data information, a cluster center x is established based on x1 -1 , and set A1 = y1, B1 = 1; select a radius r, in which B l (l=1,2,...,m) records the number of input and output samples of data information cluster l; (b) Consider the kth data information sample (x k ,y k ), if there are clusters m, the centers are x -1 ~x -m , and we get the following formula: The following formula is used to obtain the data information of the smallest cluster sample of campus topics: In the above formula, cluster p is the possible nearest neighbor cluster of campus theme data information sample k; (c) Use the following formula to make the data information sample (x k ,y k ) as a new data information clustering: Divide the campus theme data information sample k into clusters p, and then the following formula needs to be satisfied: In the formula, Indicates that it belongs to a cluster p The total input value of the data information sample of the campus theme; (d) The learning sample data information is input into the fuzzy neural network, and the total error of the data information sample of the fuzzy neural network is obtained. The following formula is used to describe it: In the above formula, s represents the total number of data information samples of campus topics. When the learning sample input is x k When f k is the output value of the neural network; (e) The following formula is used to derive ERR and to adjust the fuzzy neural network structure and parameter σ adaptively: In the above formula, q is the learning law of data information with campus as the theme; (f) Use the following formula to establish a search engine model for campus subject data information based on fuzzy neural network: In the formula, and is the parameter of the neural network, n is the number of components contained in the data information sample represented by the campus theme, and m is the sum of the clusters of data information with campus as the theme.

4. According to the method for building a campus service platform system based on a mobile terminal according to claim 1, it is characterized in that: In step (3), the function ψ is set to a function cluster {ψ ab }, expressed by the following formula: In the above formula, ψ represents the basic fundamental wave, which comes from the function family {ψ ab } is called a microwave, a is the scale parameter of the campus data information, and b represents the location parameter of the campus data information; Set the microwave generating function {ψ ab }, the expansion and shift factors are {a,b} respectively, and the continuous microwave function of the data signal f(t) in the smart campus is: In the above formula, It is the time-frequency characteristic of microwaves; The equivalent frequency domain of the data information in the above formula is divided into the following formula: In the above formula, F(w) and ψ(aw) are f(t) and ψ ab The Fourier transform of (t) is given by W f The equivalent frequency domain represented by (a, b) is defined as the time series of data information.

Images