KR20230025373A - Method and apparatus for providing an online integrated repair platform for recommending repairers through scoring for repairers - Google Patents

Abstract

Disclosed are a method and apparatus for providing a repair shop for a user through scoring on repair shops in an integrated online repair platform. The apparatus for providing a repair shop fit for a user through scoring on repair shops in an integrated online repair platform comprises: at least one processor; and a memory which stores instructions for the at least one processor to execute at least one operation. The at least one operation comprises the steps of: obtaining information on a target user, a type of target to be repaired and location information of the target user from a user terminal; and based on information on a plurality of users and repair histories, clustering repair shops indicated by the repair histories into a plurality of repairing target type clusters, selecting a type cluster to be repaired based on the information on the target user and the type of target to be repaired among the plurality of pre-obtained repairing type clusters, checking information on a plurality of repair shops contained in the selected type cluster to be repaired, and selecting a repair shop among the plurality of checked repair shops based on the location information of the target user, and generating repair shop recommending information containing the selected repair shop.

Description

Method and apparatus for providing an online integrated repair platform for recommending repair companies through scoring of repair companies

The present invention relates to a technology for providing a repair company to a user, and more particularly, to a method and apparatus for providing a repair company suitable for a user through an online integrated repair platform through scoring of the repair company.

Unless otherwise indicated herein, material described in this section is not prior art to the claims in this application, and inclusion in this section is not an admission that it is prior art.

Today, O2O (Online to Offline) business is growing in all industries. This is because as products and services in each field become globalized and supply and demand diversify, suppliers and consumers can conveniently purchase or supply through the O2O platform.

The same is true in the field of providing repair services. Consumers who want to receive repair services have a problem of incurring human costs, such as searching Internet sites or making inquiries in order to receive repair services suitable for them.

Of course, a platform that provides repair services for a specific product already exists, but a platform for not only a specific product but also all products subject to repair (including people/companies) does not exist.

Korean Patent Registration No. 10-2285861 (published date: 2021.01.19.)

An object of the present invention to solve the above problems is to provide a method and apparatus for providing a repair company suitable for a user in an integrated online repair platform.

One aspect of the present invention for achieving the above object provides a device for providing a repair company suitable for a user in an integrated online repair platform.

An apparatus for providing a repair company suitable for a user in an integrated online repair platform includes at least one processor; and a memory for storing instructions instructing the at least one processor to perform at least one operation.

The at least one operation may include obtaining information about a target user, a target repair target type, and location information of the target user from a user terminal; and clustering repair companies indicated by the repair history into a plurality of repair target type clusters based on information about a plurality of users and repair records, and the target among a plurality of repair type target clusters previously obtained. A repair target type cluster is selected based on information about the user and the target repair target type, information on a plurality of repair companies included in the selected repair target type cluster is checked, and the plurality of repair companies identified The method includes selecting a repair company based on the location information of the target user and generating repair company recommendation information including the selected repair company.

The at least one operation may include obtaining information and repair history of a plurality of users from an external server; clustering the users into a plurality of user clusters based on the information about the plurality of users; and clustering, for each of the user clusters, repair companies indicated by repair records of users belonging to the user clusters into a plurality of repair target type clusters.

The generating of the repair company recommendation information may include selecting a user cluster matching the information on the target user from among the plurality of user clusters; selecting a repair target type cluster matching the target repair target type from among a plurality of repair target type clusters corresponding to the selected user cluster; and checking information on a plurality of repair companies included in the selected repair target type cluster, selecting a repair company from among the identified plurality of repair companies based on the location information of the target user, and selecting the selected repair company. It may include generating repair company recommendation information that includes.

Selecting a user cluster matching the information on the target user may include converting the information on the target user into a vector representing characteristics of each item included in the information, and converting the converted vector to the plurality of users. Calculating distances of the clusters from center points, respectively, matching a user cluster corresponding to the center point of the user cluster corresponding to the smallest distance among the calculated distances with the target user, and selecting the matched user cluster. can

The generating of the repair company recommendation information may include selecting repair target type clusters that do not match a target repair target type among the plurality of repair target type clusters, and selecting the repair target type clusters according to the number of data belonging to each repair target type cluster. Calculating a score of each of the repair target type clusters, selecting a repair target type cluster having a score equal to or greater than a preset value, and including the repair target type of the selected repair target type cluster in repair company recommendation information. can

The information about the above user includes age, gender, occupation, hobbies, marital status, children, vehicle ownership, disability, academic background, special skills, language skills, pets, qualifications obtained, future wishes, height, weight, It may include at least one of visual acuity and overseas travel experience.

The repair history may include at least one of a repair company, a repair target type, a service providing area, a service providing date, the number of purchases of the same repair service, and repair contents.

The clustering of the users into a plurality of user clusters based on the information about the plurality of users may include assigning information about each user included in the information about the plurality of users to each item. is converted into a user vector representing the characteristics of a user vector, and among all N user vectors (N is a natural number greater than or equal to 1, which may be equal to the number of users), k pre-set (k is a natural number greater than or equal to 2) to be selected as the center point of the cluster. Randomly select user vectors of (step (1)), set coordinates pointed by each of the selected k user vectors to k cluster center points, and overlap between each of the total N user vectors and the center points. A degree coefficient is calculated, each of the user vectors is assigned to one of the k clusters based on the calculated k multiplicity coefficients (step (2)), and the user vectors assigned to each of the k clusters are used. to reset the cluster center points in each of the k clusters (step (3)), and repeat steps (2) and (3) until the cluster center point set in step (3) does not differ from the existing cluster center point and reallocating the user vectors to each of the k clusters (step (4)), and obtaining an end user cluster based on the iterative result.

The redundancy coefficient may be determined based on a distance between the vector and the center points, and the redundancy coefficient may be set to increase as the distance is closer.

The redundancy coefficient may be determined based on the following equation.

Here, A means the user vector, B is the vector pointing to the center point, Euc is a function that calculates the Euclidean distance, and the multiplicity coefficient is the dot product between the user vector and the vector pointing to the center point. It can be defined as a value divided by the Euclidean distance between the user vector and the vector pointing to the center point.

Selecting a repair company based on the location information of the target user may include calculating a score according to a difference between the location of the repair company included in the extracted repair target type cluster and the location of the target user; and assigning a priority to the repair company using the method, and selecting a repair company having the highest priority.

The score may be determined by the following equation.

는 선정된 수리 대상 유형 클러스터에 포함된 수리 업체를 지시하는 데이터의 총 개수, P_x는 수리 업체 x의 서비스 제공 가격의 평균을 의미하고, L_x는 사용자 단말과 수리 업체의 직선 거리를 의미할 수 있다. k1, k2, 및 k3는 각 항에 대한 가중치를 의미할 수 있다.Here, Score _x means the score for repair company x, N _s,x is the number of data indicating x repair company included in the selected repair target type cluster,

is the total number of data indicating repair companies included in the selected repair target type cluster, P _x means the average service price of repair company x, and L _x means the straight-line distance between the user terminal and the repair company. can k1, k2, and k3 may mean weights for each term.

In the selecting of a repair company based on the location information of the target user, repair review information of a repair company recommended to the target user is obtained, and a user's rating included in the obtained repair review information is less than a preset value. In this case, a step of deducting a predetermined value when calculating the score of the repair company may be included.

The step of selecting a repair company based on the location information of the target user may include obtaining the number of revisits of a repair company recommended to the target user, and when the obtained number of revisits is equal to or greater than a preset value, the repair company A step of increasing or subtracting a predetermined value may be included when calculating the score.

According to the present invention, an effect of providing a repair company suitable for a user in an integrated online repair platform can be expected.

In addition, the present invention selects the most similar user by clustering a pre-registered user pool based on user information, and provides a repair company to which the selected user has received repair service in the past based on the user's location information. By doing so, there is an effect of providing the user and a suitable repair company in consideration of the user's location.

Effects obtainable from the embodiments are not limited to the effects mentioned above, and other effects not mentioned are clearly derived and understood by those skilled in the art based on the detailed description below. It can be.

BRIEF DESCRIPTION OF THE DRAWINGS Included as part of the detailed description to aid understanding of the embodiments, the accompanying drawings provide various embodiments and, together with the detailed description, describe technical features of the various embodiments.
1 is a schematic diagram of a method and apparatus for providing a repair company suitable for a user in an integrated online repair platform according to an embodiment of the present invention.
FIG. 2 is a block diagram showing functional modules of the recommendation server according to FIG. 1 by way of example.
3 is a diagram showing an example of data on user information and repair history.
4 is a diagram showing a conceptual diagram in which a user cluster acquisition unit and a repair target type cluster acquisition unit perform clustering.
5 is a conceptual diagram showing a repair target type cluster acquiring unit generating a repair target type cluster.
6 is a diagram showing a hardware configuration of a recommendation server according to an embodiment of the present invention by way of example.

Since the present invention can make various changes and have various embodiments, specific embodiments will be illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, and should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention. Like reference numerals have been used for like elements throughout the description of each figure.

Terms such as first, second, A, and B may be used to describe various components, but the components should not be limited by the terms. These terms are only used for the purpose of distinguishing one component from another. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element, without departing from the scope of the present invention. The terms and/or include any combination of a plurality of related recited items or any of a plurality of related recited items.

It is understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, but other elements may exist in the middle. It should be. On the other hand, when an element is referred to as “directly connected” or “directly connected” to another element, it should be understood that no other element exists in the middle.

Terms used in this application are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, the terms "include" or "have" are intended to designate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, but one or more other features It should be understood that it does not preclude the possibility of addition or existence of numbers, steps, operations, components, parts, or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in the present application, they should not be interpreted in an ideal or excessively formal meaning. don't

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a schematic diagram of a method and apparatus for providing a repair company suitable for a user in an integrated online repair platform according to an embodiment of the present invention.

Referring to FIG. 1, a method of providing a repair company suitable for a user in an integrated online repair platform is a recommendation server (100, which may be referred to as a device that provides a repair company suitable for a user in an integrated online repair platform), It can be performed using the external server 200 and the user terminal 300 . In this case, the recommendation server 100, the external server 200, and the user terminal 300 may be referred to as a recommendation system.

The external server 200 is a server in which data for the recommendation server 100 to acquire clusters is stored, and may transmit data for acquiring clusters to the recommendation server 100 . For example, the external server 200 may be a server that operates an online integrated repair platform, and the external server 200 provides users with online integrated repair platform services while providing accumulated information and repair history of a plurality of users. data can be stored. In this case, the integrated online repair platform may be a platform that provides repair services not only for products requiring repair, but also for people and/or companion animals that can be healed by expanding the concept of repair. In other words, the online integrated repair platform can handle anything in the world that can be fixed, repaired, or cured.

The user terminal 300 may be provided with a recommended repair company from the recommendation server 100 . In detail, the user terminal 300 inputs, to the recommendation server 100, information about a user (hereinafter, referred to as a target user) for whom a repair company is recommended, a target repair target type to be repaired, and location information. and a recommended repair company may be provided from the recommendation server 100 . In this case, the recommendation server 100 may recommend a repair company used by a user similar to the user to the target user based on the pre-obtained cluster. A method of recommending a repair company by the recommendation server 100 will be described in detail in another drawing below.

For example, the user terminal 300 may include a communicable desktop computer, a laptop computer, a notebook, a smart phone, a tablet PC, and a mobile phone. ), smart watch, smart glass, e-book reader, portable multimedia player (PMP), portable game device, navigation device, digital camera, digital multimedia broadcasting (DMB) It may be a player, a digital audio recorder, a digital audio player, a digital video recorder, a digital video player, a personal digital assistant (PDA), and the like.

The recommendation server 100 obtains information on a target user, a target repair target type, and location information of the target user from the user terminal 300 to whom a repair company is recommended, and based on the information on a plurality of users and the repair history. Accordingly, a repair company may be selected based on target user personal information, target repair target type, and location information of the target user in the pre-obtained cluster, and the selected repair company may be provided to the user terminal 300 . At this time, the recommendation server 100 may be a server that operates the above-described integrated online repair platform. In addition, the recommendation server 100 may be implemented in the form of a central server, a management server, a cloud server, a web server, a client server, and the like.

The recommendation server 100 according to an embodiment of the present invention obtains information and repair history of a plurality of users from an external server 300, and selects the users based on the information about the plurality of users. Repair vendors by clustering into user clusters, and clustering, for each of the user clusters, repair vendors indicated by a repair history for a user belonging to the user clusters into a plurality of repair target type clusters. Clustering may be performed in advance on user information and repair history to recommend. In addition, the recommendation server 100 matches information on a target user with a plurality of user clusters clustered in advance, and matches a target repair target type with a plurality of repair target type clusters corresponding to the matched user clusters to find a repair company. By recommending, it is possible to recommend a repair company used by a user having information similar to that of the target user to the target user.

Accordingly, since it is possible to provide a repair company suitable for the user through clustering, it is possible to recommend a repair company suitable for the user while reducing the cost for the user to find a repair company.

FIG. 2 is a block diagram showing functional modules of the recommendation server according to FIG. 1 by way of example.

Referring to FIG. 2 , the recommendation server 100 may include an information management unit 101 , a cluster acquisition unit 102 , and a repair company recommendation unit 103 .

The information management unit 101 performs a function of acquiring information to be used in the recommendation server 100 . The information management unit 101 may obtain personal information and repair records of a plurality of users from the external server 300 . Also, the information management unit 101 may obtain personal information about the target user, a target repair target type, and location information of the target user from the user terminal.

For example, the information about the user is information indicating the characteristics of the user in order to distinguish the user from other users, such as age, gender, occupation, hobby, marital status, presence or absence of children, whether or not a vehicle is owned, disability, and educational background. , special skills, language ability, whether or not there is a pet, license acquisition matters, future hopes, height, weight, eyesight, overseas travel experience, and various other information can be included.

In addition, the repair history is information on all the details of repair (concept including treatment) service provided by a specific company or person in the past for each user, and may be information that can be mapped to correspond to information about each user, Various information such as information on the repair company used by the user (or natural person who provided the repair service, that is, a craftsman in a specific repair field), type of repair target, service area, date of service delivery, number of purchases of the same repair service, contents of repair, etc. It may be information classified into a plurality of categories based on criteria.

3 is a diagram showing an example of data on user information and repair history.

Referring to FIG. 3 , information about users includes information on items such as gender, age, marital status, ownership of a vehicle, and overseas travel experience for each user. In addition, the repair history shows information on items such as the type of repair target, the address of the repair company, and the price. In this case, the repair history means the user repair history for each user. That is, the repair history is information about all details of repair services that a user has received in the past. There is a corresponding repair history for each user, and in FIG. 3 , it can be seen that the repair history indicates the repair history for user 2.

The cluster acquisition unit 102 performs a function of clustering the information acquired by the information management unit 101 to obtain a cluster. The cluster acquisition unit 102 may include a user cluster acquisition unit 102a and a repair target type cluster acquisition unit 102b according to functions.

4 is a diagram showing a conceptual diagram in which a user cluster acquisition unit and a repair target type cluster acquisition unit perform clustering.

The user cluster acquisition unit 102a may acquire a plurality of user clusters based on information about a plurality of users. The repair target type cluster acquisition unit 102b may acquire a plurality of repair target type clusters by clustering repair companies indicated by repair records of users belonging to each user cluster.

For example, referring to FIG. 4 , the user cluster acquisition unit 102a clusters user information to obtain a user A cluster and a user B cluster, and the repair target type cluster acquisition unit 102b may obtain a plurality of repair target type clusters by clustering repair companies indicated by repair records of all users belonging to the user A cluster into a plurality of repair target type clusters.

In this regard, the cluster acquisition unit 102 may perform clustering on user information and repair history by applying a K-means clustering algorithm.

First, the user cluster acquisition unit 102a converts information about each user included in information about a plurality of users into a user vector representing characteristics of each item. In the case of an item whose characteristic is expressed as a number, the component values of the vector for that item are converted as they are as numbers representing the characteristic. is converted to a preset number according to For example, information about user 1 in FIG. 2 may be converted into a vector (1, 23, 0, 0, 1 ??). The component values of the vector representing gender, marital status, vehicle ownership, and overseas travel experience are converted to a preset value of 1 or 0 according to the characteristics of the item, and the second component value of the vector representing age is the age of user 2. It could be 23 as it is.

Thereafter, the user cluster acquisition unit 102a selects a preset number of k vectors (where N is a natural number equal to or greater than 1 and may be equal to the number of users) to be selected as the central point of a cluster among user vectors representing information on users. k is a natural number greater than or equal to 2) user vectors are randomly selected (step (1)).

Then, the user cluster acquisition unit 102a sets the coordinates pointed by each of the selected k user vectors as k cluster center points C ₀ , and sets each of the set cluster center points C ₀ to k clusters. Defined as the initial center of gravity for each.

Next, the user cluster acquisition unit 102a calculates multiplicity coefficients between each of the N user vectors and the central points C ₀ (step (2)). Accordingly, k multiplicity coefficients may be calculated in relation to k center points for each of all user vectors. That is, k multiplicity coefficients can be calculated for one user vector.

The user cluster acquisition unit 102a may allocate each of the user vectors to one of the k clusters based on the calculated k multiplicity coefficients (step (2)). For example, the user cluster acquisition unit 102a may allocate each of the user vectors to a cluster having a cluster center point corresponding to the largest value among the calculated k multiplicity coefficients. That is, the user vector is assigned to the cluster having the highest multiplicity coefficient with the central point.

In this case, the multiplicity coefficient may be determined based on the distance between the user vector and the central points, and the closer the distance is, the higher the multiplicity coefficient may be set. In one embodiment, the multiplicity coefficient (OC) may be determined based on Equation 1.

Here, A means a user vector, B is a vector pointing to the center point, and Euc is a function that calculates the Euclidean distance. That is, the multiplicity coefficient is defined as a value obtained by dividing the dot product between the user vector and the vector pointing to the center point by the Euclidean distance between the user vector and the vector pointing to the center point.

Then, the user cluster acquisition unit 102a may reset cluster center points in each of the k clusters using the user vectors assigned to each of the k clusters (step (3)). Specifically, an average vector of user vectors allocated to each of the k clusters may be set as new cluster center points for each of the k clusters. For example, an average vector of user vectors assigned to the first cluster is set as a cluster center point for the first cluster.

Thereafter, the user cluster acquisition unit 102a repeats steps (2) and (3) until the cluster center point set in step (3) does not differ from the existing cluster center point, thereby dividing the user vectors into k clusters, respectively. can be reassigned (step (4)). Then, the user cluster acquisition unit 102a acquires the end user cluster based on the iteration result.

Thereafter, the repair target type cluster acquisition unit 102b acquires end user clusters from the user cluster acquisition unit 102a, identifies users included in each user cluster, and checks repair history data of users belonging to the same cluster. . For example, the repair target type cluster acquisition unit 102b may check repair history data of all users belonging to the A user cluster.

Thereafter, the repair target type cluster acquisition unit 102b acquires a plurality of repair target type clusters by clustering repair companies indicated by repair histories of all users belonging to a specific user cluster into a plurality of repair target type clusters. can do.

That is, the repair target type cluster acquisition unit 102b clusters, for each of the user clusters, repair companies indicated by the repair records of users belonging to the user clusters into a plurality of repair target type clusters. , A plurality of repair target type clusters corresponding to each of the user clusters may be created. The repair target type cluster acquisition unit 102b may generate a plurality of repair target type clusters by collecting repair histories of all users belonging to the same user clusters and clustering them based on the repair target type.

5 is a conceptual diagram showing a repair target type cluster acquiring unit generating a repair target type cluster.

Referring to FIG. 5 , it can be seen that data on repair histories of users 3 and 4 belonging to the user B cluster are shown.

The repair target type cluster acquisition unit 102b may generate a laptop cluster and a car cluster by collecting repair records of notebook computers and repair records of automobiles from repair histories of users 3 and 4 belonging to the user B cluster. The repair target type cluster includes information on a plurality of repair companies. For example, information about repair companies includes addresses and prices of repair companies.

In this case, the type to be repaired may be various types such as TV, home theater, CD player, clothes, accessories, watch, bag, shoes, belt, car, electric kickboard, skateboard, dining table, table, closet, chair, shelf, etc.

In summary, the cluster acquisition unit 102 may doubly cluster information on a plurality of users and repair history data by user type and repair target type.

The repair company recommendation unit 103 performs a function of generating repair company recommendation information from the user cluster acquired by the cluster acquisition unit 102 and the repair target type cluster based on information obtained from the user terminal 300 .

The repair company recommendation unit 103 obtains information about a target user, a target repair target type, and location information of the target user from the information management unit 101, and obtains previously obtained information about a plurality of users and repair history. In the cluster, repair company recommendation information may be generated based on the target user information, the target repair target type, and location information of the target user. At this time, the pre-acquired clusters refer to clusters obtained by the above-described cluster acquisition unit 102 .

In this regard, the repair company recommendation unit 103 selects a user cluster matching the information on the target user from among the plurality of user clusters, and selects the target repair from among a plurality of repair target type clusters corresponding to the selected user cluster. A repair target type cluster matching the target type may be selected.

In detail, the repair company recommendation unit 103 converts information on a target user into a target user vector representing characteristics of each item included in the information, and converts the converted target user vector into a plurality of finally determined user clusters. Calculate the distances from the center points C of each of the calculated distances, match the user cluster corresponding to the center point of the user cluster corresponding to the smallest distance among the calculated distances with the target user, and match the matched user cluster to the target A user cluster that matches information about a user can be selected.

That is, the repair company recommendation unit 103 matches the target user with the cluster corresponding to the center point of the cluster closest to the vector representing information about the target user, thereby specifying the range and characteristics of users similar to the characteristics of the target user. There is an effect that can be grasped by

Also, the repair company recommendation unit 103 may select a repair target type cluster matching the target repair target type from among a plurality of repair target type clusters corresponding to the selected user cluster. For example, if the target repair target type is a laptop computer, a laptop cluster may be selected.

The repair company recommendation unit 102 may select a repair company based on location information of a target user in the selected repair target type cluster. The repair target type cluster includes information on a plurality of repair companies. For example, information about repair companies includes addresses and prices of repair companies.

In this regard, the repair company recommendation unit 102 may calculate a score according to the size of the difference between the location of the repair company included in the selected repair target type cluster and the location of the target user, and use the score to calculate the repair company. It is possible to prioritize, select a repair company with the highest priority, and generate repair company recommendation information including the selected repair company. That is, the higher the score, the closer the location to the user terminal may be. For example, scores of 50 points, 60 points, and 70 points may be calculated for three repair companies belonging to the same repair target type cluster, depending on the location of the user terminal. In this case, the repair company with the highest score of 70 can be given 1st priority. Accordingly, the first-order repair company may be preferentially included in the repair company recommendation information.

In another embodiment, the repair company recommendation unit 102 may calculate a score based on the number of data indicating the same repair company included in the selected repair target type cluster, the location of the repair company, and the price. The score may be determined based on Equation 2.

는 선정된 수리 대상 유형 클러스터에 포함된 수리 업체를 지시하는 데이터의 총 개수, P_x는 수리 업체 x의 서비스 제공 가격의 평균을 의미하고, L_x는 사용자 단말과 수리 업체의 직선 거리를 의미할 수 있다. k1, k2, 및 k3는 각 항에 대한 가중치로써, 관리자에 의해 설정될 수 있다.Here, Score _x means the score for repair company x, N _s,x is the number of data indicating x repair company included in the selected repair target type cluster,

is the total number of data indicating repair companies included in the selected repair target type cluster, P _x means the average service price of repair company x, and L _x means the straight-line distance between the user terminal and the repair company. can k1, k2, and k3 are weights for each term and may be set by a manager.

In addition, the repair company recommendation unit 102 obtains repair review information of a repair company recommended to the target user from the information management unit 101, and when the user's rating included in the obtained repair review information is less than a preset value, the When calculating the score of the repair company, a certain value may be deducted.

In addition, the repair company recommendation unit 102 obtains the number of revisits of the repair company recommended to the target user from the information management unit 101, and calculates a score of the repair company when the obtained number of revisits is equal to or greater than a preset value. A certain value can be added or subtracted at any time.

In addition, the repair company recommendation unit 102 may recommend repair companies as well as products repaired by the user having characteristics similar to those of the user. In other words, it is possible to recommend products that the user may like.

In this regard, the repair company recommendation unit 102 selects a user cluster that matches the target user information, and repair target types that do not match the target repair target type among a plurality of repair target type clusters corresponding to the selected user cluster. select a cluster

Then, the repair company recommendation unit 102 calculates a score for each of the repair target type clusters according to the number of data belonging to the repair target type cluster. At this time, the repair company recommendation unit 102 calculates a score for each repair target type cluster according to the number of data belonging to each repair target type cluster.

In this regard, the number of data belonging to each of the repair target type clusters may mean the number of repairs performed for each repair target type by a user having characteristics similar to the target user. Furthermore, it can be interpreted that the greater the number of repairs, the greater the user's interest in the repair target type, so that the greater the number of data, the greater the interest of similar users in the repair target type of the repair target type cluster to which the corresponding data belongs. .

Thereafter, the repair company recommendation unit 102 may extract repair target type clusters having the score equal to or greater than a preset value, and include the repair target type of the repair target type cluster in repair vendor recommendation information.

In summary, the repair company recommendation unit 102 may provide a repair target type that is not a target repair target type but has been repaired by a similar user together with a repair company. For example, when a user is recommended a laptop repair company, a product the user may like may also be recommended.

6 is a diagram showing a hardware configuration of a recommendation server according to an embodiment of the present invention by way of example.

Referring to FIG. 6 , the recommendation server 100 includes at least one processor 110; and a memory for storing instructions instructing the at least one processor 110 to perform at least one operation.

The at least one operation may include the operation of the recommendation server 100 described with reference to FIGS. 1 to 5 .

Here, the at least one processor 110 may mean a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor for performing methods according to embodiments of the present invention. can

The memory 120 may include at least one of volatile storage media and non-volatile storage media. For example, the memory 120 may include at least one of a read only memory (ROM) and a random access memory (RAM).

The storage device 160 may be, for example, a hard disk drive (HDD) or a solid state drive (SSD).

Also, the recommendation server 100 may include a transceiver 130 that communicates through a wireless network. Also, the recommendation server 100 may further include an input interface device 140 , an output interface device 150 , a storage device 160 , and the like. Each component included in the recommendation server 100 may be connected by a bus 170 to communicate with each other.

Examples of the recommendation server 100 include a communicable desktop computer, a laptop computer, a notebook, a smart phone, a tablet PC, and a mobile phone. ), smart watch, smart glass, e-book reader, portable multimedia player (PMP), portable game device, navigation device, digital camera, digital multimedia broadcasting (DMB) It may be a player, a digital audio recorder, a digital audio player, a digital video recorder, a digital video player, a personal digital assistant (PDA), and the like.

The methods according to the present invention may be implemented in the form of program instructions that can be executed by various computer means and recorded on a computer readable medium. Computer readable media may include program instructions, data files, data structures, etc. alone or in combination. Program instructions recorded on a computer readable medium may be specially designed and configured for the present invention, or may be known and usable to those skilled in the art of computer software.

Examples of computer readable media may include hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions may include not only machine language codes generated by a compiler but also high-level language codes that can be executed by a computer using an interpreter and the like. The hardware device described above may be configured to operate with at least one software module to perform the operations of the present invention, and vice versa.

In addition, the above-described method or device may be implemented by combining all or some of its components or functions, or may be implemented separately.

Although the above has been described with reference to preferred embodiments of the present invention, those skilled in the art will variously modify and change the present invention within the scope not departing from the spirit and scope of the present invention described in the claims below. You will understand that it can be done.

100: recommendation server 101: information management unit
102: cluster acquisition unit 103a: user cluster acquisition unit 103b: repair target type cluster acquisition unit
103: repair company recommendation unit 200: external server 300: user terminal

Claims (1)

A device that provides repair companies to users through scoring of repair companies on an online integrated repair platform,
at least one processor; and
A memory for storing instructions instructing the at least one processor to perform at least one operation,
At least one operation,
obtaining, from a user terminal, information about a target user who is a user of the user terminal, a target repair target type, and location information of the target user;
The target among a plurality of repair type target clusters obtained in advance by clustering repair companies indicated by the repair history into a plurality of repair target type clusters based on information and repair records of a plurality of users. selecting a repair target type cluster based on user information and the target repair target type;
checking information on a plurality of repair companies included in the selected repair target type cluster, and selecting a repair company from among the identified repair companies based on location information of the target user; and
Generating repair company recommendation information including the selected repair company; Including,
At least one operation,
obtaining information about the plurality of users and the repair history from an external server;
clustering the users into a plurality of user clusters based on the information about the plurality of users; and
Clustering, for each of the user clusters, repair companies indicated by repair records of users belonging to the user clusters into a plurality of repair target type clusters;
Clustering the users into a plurality of user clusters,
generating a plurality of user vectors corresponding to each of the plurality of users by using information about the plurality of users;
selecting k (k is a natural number greater than or equal to 2) user vectors set in advance from among the plurality of user vectors;
setting coordinates indicated by each of the selected user vectors as center points corresponding to each of the k clusters;
calculating k multiplicity coefficients between the center points for each of the plurality of user vectors;
assigning each of the user vectors to one of the k clusters based on the calculated k multiplicity coefficients; and
Resetting the center points corresponding to each of the k clusters using user vectors assigned to each of the k clusters;
Each of the redundancy coefficients is calculated according to the following equation,

In the above equation, A is a user vector, B is a vector pointing to the center point, Euc is a function that calculates the Euclidean distance,
The step of selecting the repair company,
Calculating a score for each repair company based on the number of data indicating repair companies included in the selected repair target type cluster, the location of the repair companies, and the price according to the following equation;

Here, Scorex means the score for repair company x, Ns,x is the number of data indicating x repair companies included in the selected repair target type cluster,

is the total number of data indicating repair companies included in the selected repair target type cluster, Px means the average service price of repair company x, Lx means the straight-line distance between the user terminal and the repair company, and k1 , k2, and k3 are weights for each term, which are set by an administrator.

Images