KR102203048B1 - Method for recommending education program in education service based on the simirality and system thereof - Google Patents

Abstract

The present invention relates to a method for recommending an education program in an education service providing network system including an education service providing terminal, an education demand terminal, and an education recommendation server. The method includes: a step of receiving item profile information from the education service providing terminal, in which the item profile information includes one or both of education service information and education tool information; a step of receiving education request information from the education demand terminal, in which the education request information includes user profile information and search term information; and a step in which the education recommendation server drives an artificial neural network recommendation model function based on the education service information and the education request information, selects some of the education service information and the education tool information, selects the information as the recommendation education program and the recommendation education tool information, and transmits the same to the education demand terminal.

Description

Method of recommending education programs using similarity and its system {METHOD FOR RECOMMENDING EDUCATION PROGRAM IN EDUCATION SERVICE BASED ON THE SIMIRALITY AND SYSTEM THEREOF}

The present invention relates to a method for recommending an educational program.

In recent years, public educational institutions such as elementary schools, junior high schools, and high schools, and private educational institutions such as academies, not only general learning management, but also various experiential education activities are being performed. Accordingly, it is difficult to provide services for these various experiential education activities with only education contents provided by educational institutions, and as a result, a number of small and medium-sized education contents providers, including startups, propose new education programs and operate businesses. have.
In addition, as the development of learning tools is rapidly accelerating, it is essential for each educational institution to purchase appropriate teaching tools suitable for the learning programs they are preparing.
Under these circumstances, it has been very difficult for educational institutions to adequately provide educational programs to students to find appropriate educational content providers or appropriate teaching materials to provide their desired programs.
Korean Unexamined Patent Publication No. 10-2017-0021465 is a prior patent related to background technology.

The present invention was conceived to solve the above-described problems, and a method and system thereof for recommending educational programs using similarity in which educational institutions can select appropriate educational programs and teaching materials because they can quickly search for educational contents that change according to trends. It is to provide.

As an embodiment of the present invention conceived to solve the above-described problem, an education service providing network system using similarity including an education service providing terminal, an education demanding terminal, and an education recommendation server includes item profile information from the education service providing terminal. Receiving, wherein the item profile information includes at least one of educational service information and teaching aid information; Receiving training request information from the training demand terminal, the training request information including user profile information and search word information; And the education recommendation server, by determining a similarity between the item profile information and the education request information, and selecting some of the education service information and the teaching aid information corresponding to the item profile information having a high similarity, and recommending a recommended education program and recommendation. It may include the step of selecting as teaching aid information and transmitting it to the education demand terminal.
Here, the transmitting to the education demand terminal includes converting text data included in the item profile information into vector data, converting text data included in the education request information into vector data, and the item profile information Deriving the similarity of the vector data of the education request information and the vector data of the education request information, extracting vector data of the item profile information having a high similarity to the vector data of the education request information to derive an item profile information candidate group, and derivation Converting the vector data of the converted item profile information candidate group into a text file, filtering the converted item profile information candidate group text file, and recommending training programs and recommended teaching aid information corresponding to the filtered item profile information It may include the step of selecting and transmitting it to the education demand terminal.
Here, the education recommendation server may further include generating accumulation selection information by receiving selection education information selected from among the recommended education program and the recommended teaching aid information from the education demand terminal information.
Here, the accumulation selection information may include information selection process information.
Here, the education service information may include education service company name information, education subject range information, service time range information, service area range information, education program information, and teacher information.
Here, the user profile information may include user company name information, user area information, user target information, user gender ratio information, user person in charge personal information, and user range information.
Here, the personal information of the person in charge of the user may include information on the gender of the person in charge, information on the age of the person in charge, and information on the person in charge of the person in charge.
Here, the artificial neural network recommendation model function may be a function of selecting as the recommended education program and recommended teaching aid information by performing a CBC filtering function based on the accumulation selection information.
Here, the artificial neural network recommendation model function may be a convolutiaon neural network function.
In another embodiment of the present invention, an education service providing network system using similarity including an education service providing terminal, an education demanding terminal, and an education recommendation server, the education service providing terminal generates item profile information and transmits it to the education recommendation server. , The item profile information includes at least one of educational service information and teaching aid information; The education demand terminal generates education request information and transmits it to the education recommendation server, the education request information including user profile information and search word information; And determining the similarity of the item profile information and the education request information, selecting some of the education service information and the teaching aid information corresponding to the item profile information having a high similarity, and selecting it as a recommended education program and recommended teaching aid information. It can be transmitted to the education demand terminal.
The education recommendation server converts text data included in the item profile information into vector data, converts text data included in the education request information into vector data, and a vector of vector data of the item profile information and the education request information Derive the similarity of the data, extract the vector data of the item profile information with high similarity to the vector data of the training request information to derive the item profile information candidate group, convert the vector data of the derived item profile information candidate group into a text file , Filtering the text file of the converted item profile information candidate group may be performed, selected as the recommended training program and recommended teaching aid information corresponding to the filtered item profile information, and transmitted to the training demand terminal.

According to an embodiment of the present invention having the above-described configuration, a marketplace is provided between the education service provider and the education service consumer to smoothly provide the supply and demand for various education programs newly created according to the changing market according to the market trend. can do.
According to the present invention, an optimal service and teaching aid information is provided to an educational service consumer through deep learning, so that an educational service consumer can easily select the content they want.

1 is a conceptual diagram of an education service providing network system according to an embodiment of the present invention.
2 is a block diagram of an education service providing network system according to an embodiment of the present invention.
3 is a flow chart illustrating a method for recommending an education program in an education service providing network system according to an embodiment of the present invention.
Figure 4 is a flow chart for specifically explaining the education recommendation server according to an embodiment of the present invention.
5 is an Excel diagram for explaining an example of training request information used in an education service providing network system according to an embodiment of the present invention.
6 is an image of a search screen according to a method for recommending an education program in an education service providing network system according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the accompanying drawings. The embodiments of the present invention may be modified in various forms, and the scope of the present invention should not be construed as being limited to the embodiments described below. The present embodiments are provided to explain the present invention in more detail to those of ordinary skill in the art to which the present invention pertains.
1 is a conceptual diagram of an education service providing network system according to an embodiment of the present invention. As shown in FIG. 1, the education service providing network system according to the present invention may include an education service providing terminal 100, an education demand terminal 200, and a “gyu recommendation server 300”.
The education service providing terminal 100 is a terminal of a company that develops various educational programs, for example, a terminal of a company that can provide external education to elementary, middle and high schools, which are regular educational institutions. The education service providing terminal 100 is Item profile information is generated and transmitted to the education recommendation server 300, where the item profile information may include at least one of education service information and teaching aid information. Education service information may include search keyword information, company and education program information, and teacher information. In other words, the education service information may include at least one of education service company name information, education subject range information, service time range information, service area range information, education program information, and teacher information.
The education demand terminal 200 is a terminal of a company that requests education programs or teaching aid information provided by the education service providing terminal 100 and may be, for example, an elementary, middle and high school terminal. The training receiving terminal generates training request information, and the training request information may include user profile information and search word information. More specifically, it may include at least one of user company name information, user region information, user target information, user male and female ratio information, user person in charge personal information, and user range information, and the user person in charge personal information may include the person in charge gender information, It can include age information and job information in charge of the person in charge.
The education recommendation server 300 drives an artificial neural network recommendation model function based on the education service information and the education request information, and selects some of the education service information and the teaching aid information as a recommended education program and recommended teaching aid information. It functions to select and transmit it to the education demand terminal 200. The artificial neural network recommendation model function may be a convolutiaon neural network function. A convolutional neural network is a type of deep neural network (DNN), consisting of one or several convolutional layers, a pooling layer, and a fully connected layer. As a neural network, CNN has a structure suitable for learning 2D data, and can be trained through a backpropagation algorithm.
The education recommendation server 300 determines the similarity of the item profile information and the education request information, and selects some of the education service information and the teaching aid information corresponding to the item profile information having a high similarity to provide a recommended education program and a recommended teaching aid. It may be selected as information and transmitted to the education demand terminal.
For example, the education recommendation server 300 converts text data included in item profile information into vector data, converts text data included in the education request information into vector data, and converts vector data of the item profile information and Derive the similarity of the vector data of the training request information, extract vector data of the item profile information with high similarity to the vector data of the training request information to derive an item profile information candidate group, and the vector data of the derived item profile information candidate group May be converted into a text file, filtered on the text file of the converted item profile information candidate group, selected as the recommended training program and recommended teaching aid information corresponding to the filtered item profile information, and transmitted to the training demand terminal. .
Hereinafter, a network system for providing an education service according to an embodiment of the present invention will be described in more detail with reference to FIG. 2.
2 is a block diagram of an education service providing network system according to an embodiment of the present invention. Here, reference numeral 100 denotes an education service providing terminal 100, 200 denotes an education demand terminal 200, and 300 denotes an education recommendation server 300.
As illustrated, the education recommendation server 300 may include a service provider DE, a service consumer DB 320, a CNN recommendation module 330, and an accumulation selection information DB 340.
As described above, the service provider DB 310 is a database storing item profile information and search word information (search word information selected by the service provider) received from the education service providing terminal 100. As described above, the ICHEM profile information includes at least one of education service information and parish information, and in the case of education service information, education service company name information, education target range information, service time range information, service area range information, and education Program information and teacher information can be included.
The service consumer DB 320 is a database that accumulates training request information transmitted from the training consumer terminal, and the training request information includes user profile information and search word information, and the user profile information includes user company name information, user region information, It may include user target information, user gender ratio information, user person in charge personal information, and user range information.
The accumulation selection information DB 340 stores all the courses selected by the education service consumer through the education program recommendation method in the education service providing network system to be described later, and deep learning training is performed through this DB so that more accurate recommendations are made. It is possible to provide educational programs and recommended teaching aid information to the education demand terminal 200.
The CNN recommendation module 330 drives an artificial neural network recommendation model function based on education service information, education request information, and accumulation selection information, selects recommended education program and recommended instructional information, and transmits it to the education demand terminal 200. Functions. In addition, when the education demand terminal 200 selects some of the education programs and teaching aid information that have been chuncheon, after receiving the selected education information from the education demand terminal 200, the accumulation selection information is generated by using the selected education information. It is stored in the DB 340. This accumulation selection information includes information selection process information (ie, all information on the above process). In other words, the artificial neural network recommendation model function can generate more accurate recommendation information by performing filtering based on the accumulation selection information.
Hereinafter, a method of recommending an education program in the education service providing network system having the above-described configuration will be described with reference to FIG. 3.
3 is a flowchart illustrating a method of recommending an education program in an education service providing network system according to an embodiment of the present invention. First, the education service providing terminal 100 generates item profile information and transmits it to the recommendation server 300 (S11). The item profile information includes at least one of educational service information such as education service company name information, education subject range information, service time range information, service area range information, education program information, teacher information, and other educational tool information. In generating education service information, the education service provider registers as a member in the recommendation server 300, and at this time, according to a general ID generation method, company name information, education subject information, and service range information are input. In addition, the education service providing terminal 100 may be configured to increase the accuracy of a recommendation search later by inputting search word information for searching for itself when registering as a member.
Then, the education demand terminal 200 transmits the education request information (S31). Before such training request information is transmitted, the training demand terminal 200 also registers as a member in the recommendation server 300, and at this time, it is configured to input basic requested program information along with various personal information. Information according to such ID generation may be included in the training request information. The training request information may include search word information input from the training demand terminal 200.
The education recommendation server 300 stores education service information and education request information in each DB and drives an artificial neural network recommendation model function to generate a recommended training program or recommended teaching aid information (recommended information) suitable for the education demand terminal 200. And it is transmitted to the education demand terminal 200 (S31, S32). The display unit of the education demand terminal 200 displays the recommended education program and the recommended teaching aid information according to the search word input by the user, and after checking the detailed information, the user selects the actual desired education program or teaching aid information. Other selection information is transmitted to the education recommendation server 300 again (S22, S23).
4 is a flow chart for specifically explaining an education recommendation server according to an embodiment of the present invention.
Referring to FIG. 4, the education recommendation server determines the similarity of item profile information and the education request information, and selects some of the education service information and the teaching aid information corresponding to the item profile information having a high similarity to provide a recommended education program. And selecting the recommended teaching aid information and transmitting it to the education demand terminal.
Transmitting the education recommendation server to the education demand terminal may convert text data included in the item profile information into vector data (400). For example, the education recommendation server uses the education request information "data_raw: ['Sexual Violence Prevention Seoul Gangnam-gu Middle School National Coed General Design 2 4 2 Chungnam Gyeryong']" as vector data "x_data: [[39 17 9 10 2]. 3 4 5 14 7 8 7 15 79 11 11 11 11 11 11 8 11 12 0 0 0 0 0]]".
The education recommendation server may convert text data included in the education request information into vector data (410).
The education recommendation server may derive a degree of similarity between vector data of the item profile information and vector data of the education request information (420). For example, the education recommendation server can derive the similarity as "batch_scores: [-59.77964 -43.46729 -53.563435 -39.96291 -39.971992 -26.947601 -56.816105 -27.615383 -48.176525 74.42727 -4.9208074 -7.525153 -43.746677 -49.84964 ]" .
The education recommendation server extracts vector data of the item profile information having a high similarity to the vector data of the training request information to derive an item profile information candidate group (430), and converts the vector data of the derived item profile information candidate group into a text file. There is (440). For example, the education recommendation server can use the vector data of the item profile information candidates derived such as "result: ['regional test','test in campus card 8','sexual violence prevention education','visiting CEO special lecture']". It can be converted to a text file.
The education recommendation server may perform filtering on the text file of the converted item profile information candidate group (450). For example, the education recommendation server may derive only the final candidate group by deleting unnecessary candidate groups based on information such as user profile information and accumulation selection information.
The education recommendation server may select a recommended education program and recommended teaching aid information corresponding to the filtered item profile information and transmit it to the education demand terminal (460).
Such selection information is generated as accumulation selection information and stored in the DB (S33).
Hereinafter, with reference to FIG. 5, training request information used in an embodiment of the present invention will be described.
5 is an Excel diagram for explaining an example of training request information used in an education service providing network system according to an embodiment of the present invention. As shown, each column represents each education demand source, and each row will be a specific example of education service information. As education service information, region information, school type information, school classification information, gender ratio information, and task information of the person in charge , Target grade information, program request information, education time information, etc. will be included. Also, keyword information and selection program information are included in this.
6 is an image diagram of a search screen according to a method for recommending an education program in an education service providing network system according to an embodiment of the present invention. The example shown in FIG. 6 is an initial screen in which the education demand terminal 200 is connected to the education recommendation server 300. When a search word is entered through the search word window 10 after logging in, the recommended program list 20 is displayed at the bottom, and when this is selected, detailed content information of each recommended program can be checked.
According to an embodiment of the present invention having the above-described configuration, a marketplace is provided between the education service provider and the education service consumer to smoothly provide the supply and demand for various education programs newly created according to the changing market according to the market trend. can do.
According to the present invention, an optimal service and teaching aid information is provided to an educational service consumer through deep learning, so that an educational service consumer can easily select the content they want.
Various modifications may be made to the described embodiments, and all or part of each of the embodiments may be selectively combined and configured. In addition, it should be noted that the embodiments described in the present specification are for the purpose of explanation and not for the limitation thereof. In addition, those skilled in the art will understand that various embodiments are possible within the scope of the technical idea of the present invention.

100: education service providing terminal
200: education demand terminal
300: Education recommendation server
310: Service provider DB
320: Service consumer DB
330: CNN recommendation module
340: accumulation selection information DB

Claims (12)

Item profile information is generated and transmitted to an education recommendation server, and the item profile information is an education service providing terminal including at least one of education service information and teaching aid information, and education request information is generated and transmitted to the education recommendation server. , The education request information includes an education demand terminal including user profile information and search word information, and the education service information corresponding to the item profile information by determining the similarity between the item profile information and the education request information. And an education recommendation server that selects some of the parish information, selects it as a recommended training program and recommended parish information, and transmits it to the education demand terminal, and provides a marketplace between the education service provider and the education service consumer to suit the trend. It can smoothly provide the supply and demand for various educational programs that are newly created in accordance with the changing market, and provide service and teaching aid information to educational service consumers through deep learning, so that educational service consumers can choose the content they want. As a method of recommending training programs for an educational service providing network system using similarity for ease of use,
Receiving, by the education recommendation server, item profile information from the education service providing terminal, the item profile information including at least one of education service information and teaching aid information;
The above educational service information,
Includes education service company name information, education target range information, service time range information, service area range information, education program information, and teacher information,
Receiving, by the education recommendation server, education request information from the education demand terminal, the education request information including user profile information and search word information;
The user profile information,
User company name information, user region information, user target information, user gender ratio information, user person in charge personal information, user range information,
The personal information of the person in charge of the user,
Including the gender information of the person in charge, the age information of the person in charge, and the work information in charge of the person in charge,
Converting, by the education recommendation server, text data included in the item profile information into vector data expressed as numerical values;
Converting, by the education recommendation server, text data included in the education request information into vector data expressed as numerical values;
The education recommendation server deriving a similarity between vector data of the item profile information and vector data of the education request information;
The education recommendation server deriving some of the vector data of the item profile information as an item profile information candidate group by using the similarity between the vector data of the item profile information and the vector data of the education request information;
Converting, by the education recommendation server, vector data expressed as a numerical value of the derived item profile information candidate group into a text file;
The education recommendation server deleting and filtering unnecessary candidate groups based on user profile information and accumulation selection information among text files of the item profile information candidate group converted based on the education service information and the user profile information;
The education recommendation server selecting a part of the education service information and the teaching aid information corresponding to the filtered item profile information, selecting a recommended education program and recommended teaching aid information, and transmitting it to the education demand terminal; And
The education service providing terminal includes the step of transmitting the selected education program or teaching aid information to the education recommendation server,
The education recommendation server is a neural network composed of one or several convolutional layers, a pooling layer, and fully connected layers based on the education service information and the education request information. Drives the convolutiaon neural network function,
The convolutiaon neural network function,
A method for recommending an education program using similarity, which is a function of selecting as the recommended education program and recommended teaching aid information by performing a filtering function based on the accumulated selection information. delete delete delete delete delete delete delete delete delete delete delete

Images