KR102207368B1 - Method for recommendation of private educational institute and apparatus for performing the method - Google Patents

Abstract

The present invention relates to a method for recommendation of a private educational institute and a device for performing the method. The method includes the steps of: receiving user information by a school recommendation server; and generating school recommendation information based on user information by the school recommendation server. According to the present invention, a private educational institute may be recommended to a user based on user location and input student information.

Description

TECHNICAL FIELD [Method for recommendation of private educational institute and apparatus for performing the method]

The present invention relates to a school recommendation method and an apparatus for performing the method. In more detail, it relates to a method for recommending a customized academy to a student in consideration of the characteristics of the student and an apparatus for performing this method.

There are parents and students who want to find a school, but information about the school is currently inevitable through some flyers, evaluation of the academy existing in some areas, and private networks. When offline meetings are difficult, it is even more difficult to share information about academy, and information that can be obtained through networks between some parents is also limited.

Due to this information exchange method, evaluation information about academy is inevitably biased, and information written by a specific person indiscriminately becomes an evaluation of academy.

Therefore, there is a need to provide information on recommending academies suitable for the characteristics of students more conveniently through more objectively collected and analyzed data.
As prior art, there are Registered Patent Publication No. 10-2018674 (2019.09.05.), Registered Patent Publication No. 10-1917383 (2018.11.09.), and Registered Patent Publication No. 10-2083257 (2020.03.02.) .

An object of the present invention is to solve all of the above-described problems.

In addition, an object of the present invention is to provide a method for recommending a school to a user based on the user's location and input student information.

Another object of the present invention is to provide an apparatus for performing a method for recommending a school to a user based on the user's location and input student information.

In addition, it aims to provide a school recommendation result to users through machine learning-based learning based on counseling record information, school information, and student information.

A typical configuration of the present invention for achieving the above object is as follows.

According to an embodiment of the present invention, a method for recommending a school may include receiving, by a school recommending server, user information, and generating, by the school recommending server, school recommendation information based on the user information.

Meanwhile, the academy recommendation information may include information on a recommended academy determined based on a candidate recommended academy (target student) and a candidate recommended academy (machine learning).

In addition, the candidate recommended academy (target student) and the candidate recommended academy (machine learning) may be academies within a user interest area.

According to another embodiment of the present invention, a school recommendation server for performing a school recommendation includes a communication unit implemented to receive user information and a process operably connected to the communication unit, wherein the processor receives user information, and the It may be implemented to generate school recommendation information based on user information.

Meanwhile, the academy recommendation information may include information on a recommended academy determined based on a candidate recommended academy (target student) and a candidate recommended academy (machine learning).

In addition, the candidate recommended academy (target student) and the candidate recommended academy (machine learning) may be academies within a user interest area.

According to the present invention, a school may be recommended to a user based on the user's location and input student information.

In addition, according to the present invention, an apparatus for performing a method for recommending a school to a user based on the user's location and input student information may be provided.

In addition, according to the present invention, a school recommendation result may be provided to a user through machine learning-based learning based on counseling record information, school information, and student information.

1 is a conceptual diagram showing a school recommendation server according to an embodiment of the present invention.
2 is a conceptual diagram showing a method for recommending a school according to an embodiment of the present invention.
3 is a conceptual diagram showing a method of recommending a private institute within a user interest area according to an embodiment of the present invention.
4 is a conceptual diagram showing a method for determining a candidate recommended academy (machine learning) according to an embodiment of the present invention.
5 is a conceptual diagram showing a method of quantifying evaluation information according to an embodiment of the present invention.
6 is a conceptual diagram showing a method for providing recommended academy information according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE INVENTION The detailed description of the present invention described below refers to the accompanying drawings, which illustrate specific embodiments in which the present invention may be practiced. These embodiments are described in detail sufficient to enable those skilled in the art to practice the present invention. It is to be understood that the various embodiments of the present invention are different from each other but need not be mutually exclusive. For example, specific shapes, structures, and characteristics described herein may be changed from one embodiment to another and implemented without departing from the spirit and scope of the present invention. In addition, it should be understood that the positions or arrangements of individual elements in each embodiment may be changed without departing from the spirit and scope of the present invention. Therefore, the detailed description to be described later is not intended to be limited, and the scope of the present invention should be taken as encompassing the scope claimed by the claims of the claims and all scopes equivalent thereto. Like reference numerals in the drawings indicate the same or similar elements over several aspects.

Hereinafter, various preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings in order to enable those skilled in the art to easily implement the present invention.

1 is a conceptual diagram showing a school recommendation server according to an embodiment of the present invention.

In FIG. 1, a school recommendation server for recommending a school based on user information is disclosed.

Referring to FIG. 1, a school recommendation server may include a user information input unit 100, a school information input unit 120, and a school recommendation unit 140.

The user information input unit 100 may be implemented to receive user information. The user information is information about a user used for recommendation of a school, and may include information on a user location, a user school, a class level when taking a user academy, a user grade, and a user academy available time. The user may be a parent or a student, and the input user information may be information based on a student.

The institute information input unit 120 may be implemented to input information about the institute. The institute information is information about the institute used for the institute recommendation, and may include information about the institute location information, the institute professor subject information, the institute teacher information, and the institute evaluation information. The institute information may be information input by the institute itself, or information input by a user who attends the institute.

The school recommendation unit 140 may be implemented to recommend a school to a user based on user information and school information. The school recommendation unit 140 may recommend a school suitable for user information based on a preset recommendation algorithm and machine learning for recommendation.

In other words, the academy recommendation server may include a communication unit implemented to receive user information and a processor operably connected to the communication unit. The processor may be implemented to perform an operation of a school recommendation server to be described later. The processor may be implemented to receive user information and generate school recommendation information based on the user information. The processor may be implemented to control the operation of the user information input unit 100, the institute information input unit 120, and the institute recommendation unit 140.

A specific method of recommending academy will be described later.

2 is a conceptual diagram showing a method for recommending a school according to an embodiment of the present invention.

In FIG. 2, a method of determining a user interest area for recommending a school based on user information is disclosed.

Referring to FIG. 2, a user interest area may be set based on user location information and user school information input as user input information.

The user location information 200 may be current location information of a user determined based on a global positioning system (GPS), or location information of a user separately input or selected by the user for a school search at a specific location.

When the user location information 200 is determined, the reference extended area 210 may be set around the user location information 200. The reference extended area 210 may be a reference area for determining a user interest area based on the user location information 200. The reference extended area 210 may be determined in consideration of location information of another user having similar characteristics to the user.

In order to determine the reference extension area 210, the school recommendation server may search for the school identity information of another user located within a critical distance around the user's location information.

A clustering group including the user location information 200 may be generated by clustering the user location information 200 and location information of other users (hereinafter, information sharing user 260) having a record of the existing institute. Thereafter, the reference extension area 210 may be determined by expanding the area so that the number of information sharing users who attend the same school as the user or a school adjacent to the user's school in the clustering group is equal to or greater than the threshold value.

Thereafter, the academy recommendation server considers the information of the institute that the information sharing users 260 in the reference extended area 210 attend, the operating range of the shuttle operated by the institute, and information about other institutes that send shuttles within the reference extended area. You can set the area.

In more detail, the primary user interest area 220 may be set through the primary extension based on the reference extension area 210. The primary user interest area 220 may be set by expanding the reference extended area to include the location of the institute where information sharing users attend and the location of another institute sending the shuttle into the reference extended area 210.

After the setting of the primary user interest area 220, the secondary user interest area 230 may be additionally set by performing clustering on the school in consideration of the location information of the school in the primary user interest area 220.

Although not included in the primary user interest area 220, the secondary user interest area 230 is formed through clustering of the school location information in order to additionally include a school included in the primary user interest area 220 and an approved school. Can be set. An area including the primary user interest area 220 and the secondary user interest area 230 determined as described above may be set as the user interest area. Candidate recommended academy (target student) and candidate recommended academy (machine learning) to be described later may be an academy located within a user interest area.

3 is a conceptual diagram showing a method of recommending a private institute within a user interest area according to an embodiment of the present invention.

In FIG. 3, a method for determining a recommended academy for recommending to a user from among the academies located in the user interest area is disclosed.

Referring to FIG. 3, a method for extracting a target student 310 and determining a candidate recommended academy (target student) 320 based on the target student 310 is disclosed.

The target student 310 may be an information sharing user having a relatively high similarity to user information. User school included in the user information 330, subjects required for user academy teaching, and user grades. When taking a user academy, a target student 310 may be extracted based on information on a class level, and the like. As the number of target students 310 increases, the standard for similarity may be set relatively high.

When at least one target student 310 exists, information on at least one candidate recommended academy (target student) 320 attended by the at least one target student 310 may be extracted.

The candidate recommended academy (target student) 320 may be prioritized in consideration of information on target student accuracy, academy location, and academy evaluation. The higher the accuracy of the target student, the higher the priority of the candidate recommended academy (target student) 320 may be relatively high, and the shorter the distance to the candidate recommended academy (target student) 320, the shorter the candidate recommended academy ( The priority of the target student) 320 may be relatively higher, and the higher the evaluation information of the candidate recommended academy (target student) 320 is, the higher the priority of the candidate recommended academy (target student) 320 is relatively It can be high.

Next, a candidate recommended academy (machine learning) 330 based on the machine learning result may be determined. A method of determining a candidate recommended academy (machine learning) 330 based on machine learning will be described in detail later.

Information on the recommended academy 340 may be finally provided to the user based on the candidate recommended academy (target student) 320 and the candidate recommended academy (machine learning) 330. The recommended academy 340, which is finally determined based on the candidate recommended academy (target student) 320 and a candidate recommended academy (machine learning) 330, is a candidate recommended academy (target student) 320 and a candidate recommended academy (machine learning). It may be determined in consideration of the reliability of the running) 330.

Candidate recommended academy (target student) 320 and candidate recommended academy (machine learning) 330 are set as academies within the user interest area described above in FIGS. 1 and 2 to limit unnecessary noise information outside a certain range. have.

Alternatively, when the user interest area restriction is released by setting an option for releasing the user interest area restriction, the candidate recommended academy (target student) 320 and the candidate recommended academy (machine learning) may include private institutes other than the user interest area. .

4 is a conceptual diagram showing a method for determining a candidate recommended academy (machine learning) according to an embodiment of the present invention.

In FIG. 4, a method for recommending a school for determining a candidate recommended academy (machine learning) based on data preprocessing and machine learning is disclosed.

Referring to FIG. 4, data input to determine a candidate recommended academy (machine learning) according to an embodiment of the present invention may be externally collected data collected from outside the academy recommendation system. For example, the externally collected data may include existing school search record information and existing school evaluation information input and collected through an external SNS, an external portal site, or the like. The externally collected data may also be obtained from the chatbot and the user's counseling record, the counselor's and the user's counseling record provided by a server such as a school recommendation server or another external service server.

Among the externally collected data, the school information including the school name information and the school location information is converted into a school location code determined through clustering based on the school identification code and school location to identify individual school through the primary data processing 410 Can be.

Specifically, the location information of the institute may be coded for each cluster unit determined based on clustering in consideration of the distance between the institutes. The first cluster may be coded as a school location code 1, the second cluster as a school location code 2 and ??, and the n-th cluster may be coded as a school location code n. In an embodiment of the present invention, the code of the nearest cluster may be used for individual academies not included in the cluster through clustering.

Data including (academy identification code, academy location code, academy evaluation information) may be generated through this primary data processing 410.

Through the secondary data processing 420 based on the learning model 415, the school evaluation information may be numerically converted into a school evaluation score.

The institute evaluation information may be quantified as an institute evaluation score based on the existing institute search record information and the existing institute evaluation information.

Based on the existing school search record information, an hourly school search amount may be extracted, and a school evaluation correction value may be determined based on the timely school search amount. The amount of school search by time may be divided into a school search amount (hour unit 1), ??, and a school search amount (time unit n) according to the set time unit (search amount) 430. The time unit (search amount) 430 may be adaptively adjusted for each school according to the school search volume. The time unit (search amount) 430 may be adjusted to be relatively short as the amount of academy search is relatively large, and the time unit (search amount) 430 may be adjusted to be relatively long as the amount of search for the academy is relatively small. The time unit (search amount) 430 may be adjusted in order to reflect the latest evaluation results as the information on the institute is rapidly updated.

In addition, on the learning model, based on the existing school evaluation information, the time-based positive evaluation index and the time-wise negative evaluation index may be determined. The hourly positive evaluation index and the hourly negative evaluation index of the academy may be obtained through a learning model 415 separately modeled based on the existing school evaluation information.

The learning model 415 may be a supervised learning model based on (academy positive evaluation sentences, positive) and (academy negative evaluation sentences, negative). In other words, if a specific evaluation sentence is entered through a supervised learning method that inputs data with a positive tag on a sentence expressing positively about the academy and data with a negative tag attached to a sentence expressing negatively with the academy, whether the evaluation sentence is positive A determination as to whether or not it is negative may be performed.

For each school, the existing school evaluation information is classified by time unit (evaluation information) 440 to determine whether it is positive or negative, so that the school evaluation score (hour unit 1), ??, and school evaluation score (time unit n) are generated. I can.

The time unit (evaluation information) 440 may be adjusted relatively short as the existing school evaluation information is relatively larger, and the time unit (evaluation information) 440 may be adjusted relatively longer as the existing school evaluation information is relatively small. .

The school evaluation score (hour unit 1), ??, the existing school evaluation score (hour unit n) are matched with the school search volume (hour unit 1), ??, the school search volume (time unit n), and finally the evaluation information Through digitization, the school evaluation score can be determined.

When user information (user location information, user school required subject information, user grade information) is input based on (school identification code, school location code, school evaluation score) determined through the above method, respond to user location information The institute having the location code of the institute can be extracted first, and after the first extraction, the corresponding institute is secondarily extracted based on (user institute teaching required subject information, user grade information), and based on the institute evaluation score. By listing the institutes, candidate recommended institutes (machine learning) can be provided.

Hereinafter, a method of digitizing the evaluation information of a specific institute is disclosed.

5 is a conceptual diagram showing a method of quantifying evaluation information according to an embodiment of the present invention.

In FIG. 5, a method for determining a candidate recommended academy (machine learning) based on machine learning is disclosed.

Referring to FIG. 5, a school search volume (hour unit 1), ??, a school search volume (time unit n) and existing school evaluation information (time unit 1), ??, existing school evaluation information (hour unit) In n), final evaluation information may be quantified through matching of the school search amount (hour unit 1), ??, and the school search amount (time unit n).

School search volume (hour unit 1), ??, school search volume (hour unit n) and school evaluation score (hour unit 1), ??, school evaluation score (hour unit n) is the school search volume (hour unit 1) , ??, academy search amount (time unit n) may be matched with each other as the number of overlapping areas increases.

For example, the school search volume (hour unit 1) 510 is matched with the school evaluation score (hour unit 1) 515, and the school search volume (time unit 2) 520 is the school evaluation score (time unit 2) ) 525 and the school evaluation score (hour unit 3) 535.

After such time-based matching, the school evaluation score (hour unit 1) 515 to the existing school evaluation information (time unit n) may be corrected based on the amount of search for the school.

The school evaluation score (hour unit n) is determined in consideration of the positive and negative ratios, and an additional correction for the positive evaluation value may be performed as the amount of search for a school increases.

Through the correction procedure as described above, an evaluation value of the school evaluation score (hour unit 1) to the school evaluation score (time unit n) may be determined.

The final evaluation value 560 is given a relatively high weight as the school evaluation score is closer from the current point in time, so that the evaluation values of each school evaluation score (time unit 1) to the school evaluation score (time unit n) are determined, and the weight is applied. The evaluation value of the academy may be determined as an average value of each evaluation value from the school evaluation score (time unit 1) to the school evaluation score (time unit n).

Through this method, it is possible to more accurately reflect the changes in the evaluation of the institute by increasing the weight of the current institute evaluation rather than the weight of the past institute evaluation. In addition, the positive evaluation value is corrected by taking into account the user's interest in the academy rather than a separate evaluation of the academy through the weight of the search volume of the academy. It can be corrected to have a high evaluation value.

Each of the candidate recommended academies (machine learning) may have a reliability value set individually, and the reliability may be set higher as the number of institute searches and institute evaluation scores are relatively larger.

6 is a conceptual diagram showing a method for providing recommended academy information according to an embodiment of the present invention.

In FIG. 6, a method for finally providing information on a recommended academy to a user based on a candidate recommended academy (target student) and a candidate recommended academy (machine learning) is disclosed.

Referring to FIG. 6, in the case of an academy commonly recommended to a candidate recommended academy (target student) 600 and a candidate recommended academy (machine learning) 610, it may be a first priority group 650. When a plurality of candidate recommended academies exist in the first priority group 650, the priority of the candidate recommended academy (target student) 600 and the priority of the candidate recommended academy (machine learning) 610 are considered. 1 A candidate recommended academy in the priority group 650 may be determined.

Next, among the academies and candidate-recommended academies (machine learning) 610 included only in the candidate-recommended academy (target student) 600, a candidate-recommended academy (machine learning) 610 having a critical reliability or higher is the second priority group ( 660). Within the second priority group 660, the schools having a high priority among the schools included only in the candidate recommended academy (target student) 600, and the school having the highest evaluation value among the candidate recommended academies (machine learning) 610 The priority of the academy can be set.

The academy included only in the remaining candidate recommended academy (machine learning) 610 may be set as the third priority group 670. In the third priority group 670, the priority may be determined in the order of the institutes having a high evaluation value among candidate recommended institutes (machine learning) 610.

The embodiments according to the present invention described above may be implemented in the form of program instructions that can be executed through various computer components and recorded in a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded in the computer-readable recording medium may be specially designed and constructed for the present invention or may be known and usable to those skilled in the computer software field. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical recording media such as CD-ROMs and DVDs, and magnetic-optical media such as floptical disks. medium), and a hardware device specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of the program instructions include not only machine language codes such as those produced by a compiler but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware device can be changed to one or more software modules to perform the processing according to the present invention, and vice versa.

In the above, the present invention has been described by specific matters such as specific elements and limited embodiments and drawings, but this is provided only to help a more general understanding of the present invention, and the present invention is not limited to the above embodiments. Anyone with ordinary knowledge in the technical field to which the invention belongs can make various modifications and changes from these descriptions.

Accordingly, the spirit of the present invention is limited to the above-described embodiments and should not be defined, and all ranges equivalent to or equivalently changed from the claims to be described later as well as the claims to be described later are the scope of the spirit of the present invention. It will be said to belong to.

Claims (6)

How to recommend a school,
Receiving, by the academy recommendation server, user information;
Including the step of the school recommendation server generating school recommendation information based on the user information,
The academy recommendation information includes information on a recommended academy determined based on a candidate recommended academy (target student) and a candidate recommended academy (machine learning),
The candidate recommended academy (target student) and the candidate recommended academy (machine learning) are academies within a user interest area,
The user region of interest includes a first region of interest and a second region of interest,
The primary region of interest is determined based on a reference extended region determined based on user location information,
The reference extended area is performed based on information sharing user location information and clustering based on the user location information,
The second region of interest is determined based on clustering of academies adjacent to the academy in the first region of interest,
The candidate recommended school (target student) is recommended based on the school the target student attends,
The target student is an information sharing user having a relatively high similarity with user information,
The candidate recommended school (machine learning) is a school recommended based on machine learning,
The candidate recommended academy (machine learning) is performed based on primary data processing and secondary data processing,
The primary data processing is converted into a school identification code and a school location code determined through clustering based on the school location to generate data including the school identification code, the school location code, and school evaluation information,
In the secondary data processing, the school evaluation information is quantified as a school evaluation score based on the existing school search record information and the existing school evaluation information,
Based on the existing school search record information, the amount of school search by time is extracted,
A method, characterized in that a school evaluation correction value is determined based on the time-based school search volume, and the school evaluation score is corrected based on the school evaluation correction value. delete The method of claim 1,
The time-based school search amount includes a plurality of school search amounts (time unit n) divided according to a set time unit (search amount) (where n is a natural number),
The time unit (search volume) is adaptively adjusted for each school according to the school search volume,
The time unit (search volume) is adjusted to be relatively short as the amount of search for the academy is relatively large, and is adjusted to be relatively long as the amount of search for the academy is relatively small,
The time unit (search amount) is adjusted in consideration of the frequency of information update on the institute. The academy recommendation server that performs academy recommendation,
A communication unit implemented to receive user information; And
Including a processor operably connected to the communication unit,
The processor receives user information,
It is implemented to generate school recommendation information based on the user information,
The academy recommendation information includes information on a recommended academy determined based on a candidate recommended academy (target student) and a candidate recommended academy (machine learning),
The candidate recommended academy (target student) and the candidate recommended academy (machine learning) are academies within a user interest area,
The user region of interest includes a first region of interest and a second region of interest,
The primary region of interest is determined based on a reference extended region determined based on user location information,
The reference extended area is performed based on information sharing user location information and clustering based on the user location information,
The second region of interest is determined based on clustering of academies adjacent to the academy in the first region of interest,
The candidate recommended academy (target student) is recommended based on the academy that the target student attends,
The target student is an information sharing user having a relatively high similarity with user information,
The candidate recommended school (machine learning) is a school recommended based on machine learning,
The candidate recommended academy (machine learning) is performed based on primary data processing and secondary data processing,
The primary data processing is converted to a school identification code and a school location code determined through clustering based on the school location to generate data including the school identification code, the school location code, and school evaluation information,
In the secondary data processing, the school evaluation information is quantified as a school evaluation score based on the existing school search record information and the existing school evaluation information,
Based on the existing school search record information, the amount of school search by time is extracted,
A school recommendation server, characterized in that the school evaluation correction value is determined based on the time-based school search volume, and the school evaluation score is corrected based on the school evaluation correction value. delete The method of claim 4,
The time-based school search amount includes a plurality of school search amounts (time unit n) divided according to a set time unit (search amount) (where n is a natural number),
The time unit (search volume) is adaptively adjusted for each school according to the school search volume,
The time unit (search volume) is adjusted to be relatively short as the amount of search for the academy is relatively large, and is adjusted to be relatively long as the amount of search for the academy is relatively small,
The time unit (search amount) is a school recommendation server, characterized in that adjusted in consideration of the frequency of information update on the school.

Images