KR20190109144A - Text Recommendation Device and Text Recommendation Method Using Multiple Text Recommendation Engines - Google Patents

Abstract

Disclosed are a device and a method of recommending a text by using a plurality text recommendation engines. According to one aspect of the present invention, the device of recommending a text by using a plurality of text recommendation engines comprises: the plurality of text recommendation engines; an adapter logically disposed between the text recommendation engines and an administer terminal such that the adapter controls the device of recommending a text; a text recommendation engine setting storing unit storing setting information of the text recommendation engines; and a text recommendation engine attribute storing unit storing attribution information of the text recommendation engines. The method of recommending a text by using a plurality of text recommendation engines comprises: a step of allowing the device of recommending a text to receive at least one of first information, which is identification information about a text recommendation engine to be used by the administer terminal, and second information, which is information about the number of the text recommendation engines to be used by the administer terminal, or third information which is attribute information of each text recommendation engine; a step of allowing the adaptor to store at least one of the first information and second information in the text recommendation engine setting unit or the third information in the text recommendation engine attribute storing unit; a step of allowing the device of recommending a text to select the text recommendation engine by using at least one of the first information and the second information or determine an attribute value of each text recommendation engine by using the third information; a step of allowing the selected text recommendation engine to select a recommended text; a step of allowing the device of recommending a text to store the selected recommended text.

Description

Text Recommendation Device and Text Recommendation Method Using Multiple Text Recommendation Engines}

The present invention relates to a text recommendation apparatus using a text recommendation engine and a text recommendation method.

There has been a service of recording and managing books read from the past, and sharing the evaluation of the books. In addition, the book recommendation service associated with it is also being implemented, and the service is being formed in Korea, but book recommendation service is firmly established in overseas countries, especially in the United States. This book recommendation service is currently in service with a smartphone application as well as an internet website.

When reviewing patents related to performing book recommendation services in Korea, there is 'Book SNS system and its provision method' (No. 10-2014-0038017, published date 2014.03.28). The present disclosure discloses a concept of sharing another user's content based on SNS (Social Networking Service) for book recommendation. Readers who read the book also share information about the book in 'Methods and Systems for Providing Social Network Services through Sharing Reading Information' (No. 10-2014-0133647, Publication Date 2104.11.20) Only the concept of sharing with other readers is disclosed.

All of the preceding book recommendation methods are limited to recommending books to readers using SNS or personalized information, and there is no clear description on how to analyze, classify, and recommend the actual contents of books.

In addition, most of these recommendation services have a limitation in recommending text or content through a single recommendation engine. In other words, there are many criteria for recommending text or content, such as sales volume, personal signs, trend trends, and social issues. However, conventional recommendation services do not respond to this situation and use text and content through a recommendation engine with rigid logic. Recommended situation.

Republic of Korea Patent Publication No. 10-2014-0038017 (2014.03.28.) Republic of Korea Patent Publication No. 10-2014-0133647 (2104.11.20.)

An object of the present invention is to provide a text recommendation apparatus utilizing a text recommendation engine.

It is also an object of the present invention to provide a text recommendation method utilizing a text recommendation engine.

More specifically, the present invention provides a text recommendation apparatus capable of recommending an optimal text or content by adjusting a plurality of text recommendation engines that can be recommended from various viewpoints according to a user's request.

According to an aspect of the invention, the text recommendation method is a plurality of text recommendation engine, logically located between the text recommendation engine and the administrator terminal, the text recommendation device for controlling the text recommendation device, text recommendation for storing the setting information of the text recommendation engine A text recommendation apparatus comprising an engine setting storage unit and a text recommendation engine attribute storage unit for storing attribute information of a text recommendation engine, the text recommendation apparatus comprising: first information which is text recommendation engine identification information to be used from an administrator terminal or text recommendation to be used Receiving at least one of the second information as the number information of the engine or third information as attribute information of each text recommendation engine; Storing, by the adapter, at least one of the first information and the second information in the text recommendation engine setting storage unit or storing the third information in the text recommendation engine attribute storage unit; Selecting, by the text recommendation apparatus, a text recommendation engine using at least one of the first information and the second information, or determining an attribute value of each text recommendation engine using the third information; Selecting the recommended text by the selected text recommendation engine; And storing, by the text recommendation device, the selected recommendation text.

According to another aspect of the present invention, a text recommendation apparatus includes: a plurality of text recommendation engines for selecting recommended texts; A text recommendation engine setting storage unit configured to store setting information of the text recommendation engine; A text recommendation engine attribute storage unit for storing attribute information of the text recommendation engine; A data collection unit receiving one or more of first information, which is text recommendation engine identification information to be used from the administrator terminal, second information which is information on the number of text recommendation engines to be used, or third information, which is attribute information of each text recommendation engine; And logically located between the text recommendation engine and a manager terminal to control a text recommendation device and store at least one of the first information and the second information in the text recommendation engine setting storage unit or store the third information in the text. A text recommendation apparatus including an adapter for storing in the recommendation engine attribute storage unit, wherein the text recommendation apparatus selects a text recommendation engine using at least one of the first information and the second information, or selects the third information. To determine an attribute value of each text recommendation engine and to store the selected recommendation text.

The text recommendation apparatus may further include receiving fourth information that is information about a user who uses or recommends the recommended text.

The text recommendation apparatus may further include outputting the stored recommendation text to the user terminal.

The text recommendation apparatus may further include receiving fifth information regarding a user's response to the recommendation text output to the user terminal; The adapter storing the fifth information in a feedback storage; Modifying, by the text recommendation device, the first information or the third information based on the fifth information; And allowing the text recommendation apparatus to select the recommendation text by the text recommendation engine.

In addition, the text recommendation apparatus inputs at least one of the first information which is the text recommendation engine identification information to be used from the administrator terminal or the second information which is the number information of the text recommendation engine to be used, or the third information which is the attribute information of each text recommendation engine. The receiving of the text recommendation device may include one or more of first information, which is text recommendation engine identification information to be used from an administrator terminal, or second information, which is information on the number of text recommendation engines to be used, according to a web page output to the user terminal. Receiving the third information, which is attribute information of the text recommendation engine, and selecting the recommendation text by the selected text recommendation engine, selects the recommendation text according to the web page output by the selected text recommendation engine to the user terminal. And outputting the stored recommendation text to the user terminal. Step may be a step of outputting the stored text like to the user terminal according to the output of the web page that the user terminal.

The text recommendation apparatus may be further configured to select a text recommendation engine using at least one of the first information and the second information, or to determine an attribute value of each text recommendation engine using the third information. The text recommendation apparatus selects a text recommendation engine for each user by using one or more of the first information or the second information and uses the fourth information, or uses the third information to attribute the text recommendation engine. Determining the value.

In addition, the text recommendation apparatus inputs at least one of the first information which is the text recommendation engine identification information to be used from the administrator terminal or the second information which is the number information of the text recommendation engine to be used, or the third information which is the attribute information of each text recommendation engine. The receiving of the text recommendation device may include: first information, which is text recommendation engine identification information to be used from an administrator terminal, or second information, which is information about the number of text recommendation engines to be used, according to the divided section of the web page output to the user terminal. Receiving at least one third information, which is attribute information of each text recommendation engine, and selecting the recommendation text by the selected text recommendation engine comprises: classifying a web page in which the selected text recommendation engine is output to the user terminal. Selecting the recommendation text according to the partition, and the stored recommendation. And outputting the host to the user terminal may be, for outputting the stored text like to the user terminal in accordance with the separated section of the web page that is output in the user terminal.

In addition, the text recommendation apparatus receives fourth information that is information about a user who uses or uses the recommendation text, and the text recommendation apparatus outputs the stored recommendation text to the user terminal, and the recommendation text output to the user terminal. Receiving the fifth information regarding the user's response to the first information, the adapter stores the fifth information in a feedback storage unit, and the text recommendation apparatus based on the fifth information, the first information or the third information. It may be to modify the, and the text recommendation engine to select the recommended text.

According to an exemplary embodiment of the present invention, a text recommendation apparatus using a recommendation engine may be provided.

According to an embodiment of the present invention, a text recommendation method using a text recommendation engine may be provided.

1 and 2 are flow charts shown in accordance with one embodiment of the present invention.
3 is a block diagram according to an embodiment of the present invention.

As the invention allows for various changes and numerous embodiments, particular embodiments will be illustrated in the drawings and described in detail in the written description. However, this is not intended to limit the present invention to specific embodiments, it should be understood to include all transformations, equivalents, and substitutes included in the spirit and scope of the present invention. In the following description of the present invention, if it is determined that the detailed description of the related known technology may obscure the gist of the present invention, the detailed description thereof will be omitted.

Terms such as first and second may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another.

The terminology used herein is for the purpose of describing particular example embodiments only and is not intended to be limiting of the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this application, the terms "comprise" or "have" are intended to indicate that there is a feature, number, step, operation, component, part, or combination thereof described in the specification, and one or more other features. It is to be understood that the present invention does not exclude the possibility of the presence or the addition of numbers, steps, operations, components, components, or a combination thereof.

Hereinafter, embodiments according to the present invention will be described in detail with reference to the accompanying drawings, and in the following description with reference to the accompanying drawings, the same or corresponding components are given the same reference numerals and redundant description thereof will be omitted. Shall be.

In addition, terms such as first and second used below are merely identification symbols for distinguishing the same or corresponding components, and the same or corresponding components are limited by terms such as the first and second components. no.

In addition, the coupling does not only mean the case where the physical contact is directly between the components in the contact relationship between the components, other components are interposed between the components, the components in the other components Use it as a comprehensive concept until each contact.

Further, in connection with a computer program, the suffixes "module" and / or "unit" for a component may be part of a computer program for functioning the computer as a means of a specific function, or may be a part of a computer program for realizing a specific function in the computer. It can be part. For example, the module A can be interpreted as a computer program for operating the computer as the means A or a computer program for realizing the function A in the computer. As a method "steps" may be implemented as computer programs on a computer and executed.

On the other hand, an application refers to a set of computer programs designed to perform a specific task, also called an application program. A user may add related functions by installing an application according to an embodiment of the present invention to his or her electronic device.

The electronic device of the user where the application is installed is composed of CPU, RAM, ROM, and storage devices such as computers, tablet PCs, and smartphones, and is an environment in which the entire system is controlled by graphic operating systems such as Windows, ios, Android, and Linux. It is preferable, and in particular, a smart phone that can send and receive calls and texts to the registered contact.

For reference, in the present invention, content or text are terms that may be used interchangeably. Recommendation engines and text recommendation engines are terms that can be used interchangeably.

1 is a view showing a flowchart according to the present invention. Referring to FIG. 1, the text recommendation apparatus may include at least one of first information which is text recommendation engine identification information to be used from the administrator terminal, second information which is information about the number of text recommendation engines to be used, or third attribute information of each text recommendation engine. Receiving information (S110); Collecting all related data (S120); The adapter storing engine settings, attributes, a database to be used, and the like in a storage unit (S130); Storing, by the text recommendation apparatus, the event record collected by the data collector in the event server (S140); And selecting the recommendation book under the attribute value by the selected recommendation engine based on the engine setting, the attribute, and the event record (S150). It will be described in detail below.

First, in connection with step S110, the identification information of the text recommendation engine indicates which recommendation engine to use and receives the identification information from the administrator terminal. For example, suppose there are various types of recommendation engines that recommend text, and more specifically, books. Each recommendation engine may be an engine designed to timely recommend a book related to a recent news, or an engine designed to recommend a book most suitable for the behavior by analyzing a reader's behavior (e.g., behavior on SNS). It may be. A detailed text recommendation engine will be described later. As such, there can be various recommendation engines. The text recommendation device receives information about which engine to use. In the case of the second information, information on how many engines to use is used instead of a specific engine. An administrator, that is, an administrator who uses the text recommendation apparatus (for example, an administrator of a web service using the present invention) may not be aware of the characteristics of each recommendation engine. Instead, the administrator can provide the recommended text to the user by utilizing the checked number of engines by checking the number of recommended engines provided by the text recommendation device. For example, when an operator who operates a text recommendation device and an administrator using the text recommendation contract, the administrator may select how many engines of the text recommendation engines to use. Meanwhile, as more engines are used, administrators can check the feedback of various users as feedback. Specific recommendation engines include keywords or content searched by the first user, more specifically content or text that other users with similar inclinations may recommend, in more detail, recommending various categories of content or text for a particular user. Recommended user-to-user text recommendation engine, buy together text or other content that performs a variety of actions, such as buying or reading certain content or text at the same time. Or content that users who have purchased, read, or subscribe to works of interest, such as text, or content that have purchased, read, or registered works of interest, recommend an item-to-item text recommendation engine, entire content, or Popular content that recommends bestsellers, soaring content, or text within specific categories Popular text recommendation engines, current real-time social topics, keywords related to content or news related to the topic at issue, real-time search queries on portal sites, social trend text recommendation engines using SNS, It may be a user profile (demographic) text recommendation engine that recommends the content or text in consideration of personal information such as gender, age, location, and nationality of the user who will consume the text or content.

Next, the third information may include information about properties of each engine. Each recommendation engine has different performances and attributes depending on the recommendation engine. Therefore, the administrator can set to recommend the content of the desired result by changing the attribute value of each recommendation engine. For example, the attribute value may be various attributes such as a category to be recommended and a category to exclude the recommendation.

Meanwhile, the text recommendation apparatus may receive the first information or the second information, and the third information may be additionally input when the first information is input. However, the present invention is not limited thereto. The text recommendation device may receive only the attribute value, and the text recommendation device may recommend text by selecting a recommendation engine that can use the attribute value automatically. For example, various recommendation engines may exist, and each recommendation engine may have a first text recommendation engine that strongly recommends trends such as domestic news, and a second text recommendation engine that reflects moderately. There may be a third text recommendation engine that may or may not be reflective. On the other hand, there may be a fourth text recommendation engine that recommends strongly reflecting the trend of overseas news, etc., there may be a fifth text recommendation engine that recommends reflecting moderately or the sixth text recommendation engine that weakens the reflection There may be. In such a case, when the administrator sets the attribute value to reflect the trend tendency, in this case, when the administrator does not select the text engine and sets the attribute value to be recommended to simply reflect the trend tendency, the text recommendation apparatus may set the first text. The recommendation engine and the fourth text recommendation engine automatically increase the attribute value to recommend the text including the trendy trend.

Next, the text recommendation apparatus collects related data (S120). The related data may be performed by the data collector in the text recommendation apparatus. The data collector collects data about the administrator who will use the text recommendation device. For example, if the administrator is an operator of a content providing server, such as a web story, the user's activity on the content providing server (e.g., purchased web novel, browsed web novel, view time, view time, view to read) Information such as time taken, user activity on social network services, user terminal location information, and user terminal status information (eg, monitor screen size, operating system information, etc.) can be collected. The text recommender does not necessarily need the S120 for text recommendation, and may recommend text except for the above steps.

Next, the adapter in the text recommendation apparatus stores any one or more of the first information or the second information in the text recommendation engine setting storage unit or the third information in the text recommendation engine attribute storage unit (S130). ). The adapter exists between the user terminal and the text recommendation device, and serves to connect both. The adapter stores the above-mentioned first information to third information in a storage unit, and also stores the text recommended by the text recommendation engine again. Outputs to In more detail, the first information to the third information may be managed and stored in one command set for each terminal of the administrator or according to the degree desired by the administrator. In more detail, the administrator terminal may generate a command set using the first text recommendation engine and using the 3-1 attribute information for the first text recommendation engine. In addition, a command set using the second text recommendation engine and using the 3-2 attribute information for the second text recommendation engine may be generated. In addition, a command set using the third-3 attribute information may be generated in the first text recommendation engine. As described above, various types of command sets may exist for each manager or even if the managers are the same.

Meanwhile, the change of the first information to the third information may also be stored in the text recommendation engine setting storage unit and the text recommendation engine attribute storage unit. For example, it is assumed that an administrator who is a user selects to use the first to third recommendation engines on January 1, 2018, and is stored in the text recommendation engine setting storage unit. Subsequently, on February 1, 2018, when the administrator changes the first, second, and fourth recommendation engines from the first to the third recommendation engines and selects them again, the above change history is stored in the text recommendation engine setting storage unit. Will be recorded. Similarly, the text recommendation engine attribute store can also store time series changes (e.g., change the recommendation from the romance category field on January 1, 2018 to the recommendation of the mystery category field on January 1, 2018).

On the other hand, the administrator as a user can set in detail which text recommendation engine to use for each page or part within the page, and further, select which attribute to use. More specifically, for example, the homepage of the operating website may be set to use the first text recommendation engine and the detail page to use the second text recommendation engine. In addition, different text recommendation engines may be used for each part depending on the division of the web page.

Next, the text recommendation apparatus stores the event record collected by the data collection unit in the event server (S140). More specifically, all event records collected by the data collection unit, for example, the user's activities (e.g., purchased web novels, viewed web novels, viewed time, viewed time, time taken to read and read) Information, such as the number of times a web novel is viewed, is stored in a specific storage unit. The event record may be utilized by a text recommendation engine to recommend text. On the other hand, the present invention can be carried out except for the step S140.

Next, the text recommendation apparatus selecting a text recommendation engine using at least one of the first information and the second information or determining the attribute value of each text recommendation engine using the third information (S150). ) May be included.

Next, the selected text recommendation engine may select the recommended text through an operation (S160). Meanwhile, the text recommendation engine may recommend texts desired by the owner through the event records stored in the event server as well as the first to third information.

Next, the text recommending apparatus stores the selected recommendation text (not shown), dispersing machine learning with Apache spark for similarity analysis (not shown), and results from the similarity analysis. The method may further include a step of storing in an elastic search to quickly respond to an external recommendation request (not shown).

2 is a view showing a flowchart according to the present invention. Referring to FIG. 2, disposing a recommendation result (S210); Receiving a text recommendation request from the user terminal (S220); Outputting, by the text recommendation apparatus, the stored recommended text to a user terminal (S230); Receiving, by the text recommendation apparatus, fifth information regarding a user's response to the recommendation text output to the user terminal (S240); Storing (S250), by the adapter, the fifth information; Modifying, by the text recommendation apparatus, the first information or the third information based on the fifth information (S290); And allowing the text recommendation apparatus to select the recommended text by the text recommendation engine (S295). It is divided in detail below.

First, in relation to placing a recommendation result, the text recommended through the text recommendation engine is first batched as a result. Thereafter, when the recommendation request is input from the user terminal, the recommendation result is output in response to the request (S230).

Next, the text recommendation apparatus receives the fifth information about the user's response to the recommendation text output to the user terminal. More specifically, all actions related to the recommendation text are collected and recorded as feedback data on the recommendation text, such as when the user views, purchases, recommends, or shares the recommended text. The collection may be processed in the data collection unit.

Next, the first information, which is the text recommendation engine identification information, or the third information, which is attribute information of the text recommendation engine, which has been initially input, is corrected using the fifth information input as the feedback. More specifically, for example, assume that an administrator who is a user has selected the first first to third text recommendation engines and based on this, the first to third text recommendation engines recommend text. In this case, it is assumed that an administrator who is a user intensively views, purchases, or shares the text recommended by the first text recommendation engine. In this case, the text recommendation device reflects that the administrator, the user, is most active in recommending the first text recommendation engine, so that the text recommendation engine has more text than the second or third text recommendation engine. Can be controlled to recommend.

In another example, it is assumed that an administrator who is a user selects the first text recommendation engine and further sets the values of the first attribute information of the first text recommendation engine to a, b, and c, respectively. In this case, it is assumed that an administrator who is a user intensively reads, purchases, or shares the text recommendation result when the value of the attribute information is set to a. In this case, the text recommendation apparatus reflects that the attribute information is most responsive to the setting with a, so that the text recommendation apparatus may control to recommend more text when the first attribute information of the first text recommendation engine is set to a. Can be.

On the other hand, the feedback data for the recommendation of the administrator who is a user may be stored in a predetermined space. This will be described later in detail.

Next, the text recommendation apparatus may allow the text recommendation engine to select the recommended text (S295). In more detail, as described above, the feedback data, which is the response of the user or the administrator, is reflected to change the text recommendation engine or change or select the attribute value of the text recommendation engine so that the user or the administrator reacts to the recommended text more actively. do. On the other hand, the user's or administrator's response and feedback on the recommendation text may be reflected in the text recommendation device in real time, and may affect the recommendation of the text recursively.

Thereafter, the text recommendation device stores the selected recommendation text (not shown), disperses machine learning with Apache spark for similarity analysis (not shown), and performs an elasticity analysis of the similarity analysis result. Storing a) and quickly responding to an external recommendation request (not shown).

Meanwhile, the present invention may further include a step (not shown) of receiving, by the text recommendation apparatus, fourth information that is information on a user who uses or recommends the recommended text. In addition, the text recommendation apparatus uses one or more of the first information or the second information, and selects a text recommendation engine for each user using the fourth information, or uses the third information of each text recommendation engine. The attribute value can be determined. The user information may be all information such as static or dynamic information about the user (for example, static information such as age, gender, address, and occupation, text reading information, text category, text purchase history, and SNS activity information). In more detail, the aforementioned data collector may collect feedback data for each user, and the text recommendation apparatus may calculate different texts for each user and recommend different texts for each user. Through this, even if the attribute value of the same text recommendation engine is set in the same text recommendation engine, the customized text reflecting the information about the collected user may be recommended.

3 is a block diagram of a text recommendation apparatus according to the present invention. The description is duplicated except for the above description. Referring to FIG. 3, the text recommendation apparatus 1000 may include a data collection unit 210, a text recommendation unit 220, a text search unit 230, and a control layer 200 including an controller 240 and an adapter ( 310, text equalization engine layer including the first machine learning unit 320, text recommendation engine 410, text recommendation engine 410, second machine learning unit 420, event server 430 400.

The data collection unit collects various data as described above. The text recommendation unit outputs the text recommended by the text recommendation engine to the outside in response to a request of the user terminal or a request of the controller. The text search unit outputs text to the outside according to a request of the user terminal. The control unit controls the data collection unit, the text recommendation unit, the text search unit, the service layer unit, the text recommendation engine layer unit, and the like. In addition, the service layer unit may further include a big data storage unit 250 storing various data provided by a server (eg, a web novel providing server, etc.) operated by a user terminal or a user who is a user.

Next, the adapter is redundant as described above, and stores any one or more of the first information or the second information in the text recommendation engine setting storage unit or stores the third information in the text recommendation engine property storage unit. Do it. In addition, the adapter is located between the service layer unit and the text recommendation engine unit and serves as a channel for data flow between the two layers. In addition, the first machine learning unit may play a role of re-learning the text to be recommended based on the data input again by the user terminal according to the text recommendation of the text recommendation apparatus. More specifically, when the text recommendation device first recommends text to the user terminal, if the response of the user terminal (for example, reading the recommendation text or purchasing recommendation text) is received, the text recommendation device is sent to the user again. Learn and reflect on text recommendations. In addition, the equalizer layer unit may include a text recommendation engine setting storage unit and a text recommendation engine attribute storage unit. The description is replaced with the previous description.

 Next, the text recommendation engine layer unit may include a plurality of text recommendation engines. As described above, the text recommendation engine selected through the control of the adapter selects the text to be recommended according to the given attribute. In addition, the event server serves to store the event record collected by the text recommendation device in the data collection unit. Next, the second machine learning unit serves to machine learning to recommend text by reflecting an event. In addition, the event server may include an event storage unit 440 that can record the event. The event is as described above in connection with the method invention.

The methods and processes described above are, for example, instructions for execution by a processor, controller, or other processing device and may be encoded, compact disk read-only memory (CDROM), magnetic or optical disk, flash memory, random access memory. (RAM) or read only memory (ROM), erasable programmable read only memory (EPROM), or other machine-readable media.

Such a medium may be implemented as any device that contains, stores, communicates, propagates, or moves executable instructions for use by or in connection with an instruction executable system, apparatus, or device. Alternatively or additionally, as analog or digital logic using hardware such as one or more integrated circuits, or one or more processor execution instructions; Or in software of functions defined as an API (application programming interface) or DLL (Dynamic Link Library), local or remote procedure call or available in shared memory; Or as a combination of hardware and software.

In other implementations, the method may be represented by a signal or radio-signal medium. For example, the instructions for implementing the logic of any given program may take the form of an electrical, magnetic, optical, electromagnetic, infrared or other type of signal. The system described above receives these signals at a communication interface, such as a fiber optic interface, an antenna, or other analog or digital signal interface, recovers instructions from the signals, stores them in machine readable memory, and / or uses a processor. You can also run them.

In addition, the present invention may be implemented in hardware or software. The invention may also be embodied as computer readable code on a computer readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disks, optical data storage devices, and the like, which are also implemented in the form of carrier waves (for example, transmission over the Internet). Include. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion. And functional programs, codes and code segments for implementing the present invention can be easily inferred by programmers in the art to which the present invention belongs.

Embodiments of the present invention may include a carrier wave having electronically readable control signals that may be operated with a programmable computer system on which one of the methods described herein is executed. Embodiments of the present invention may be implemented as a computer program product having a program code, the program code being operated to execute one of the methods when the computer program is run on a computer. The program code may for example be stored on a machine readable carrier. One embodiment of the invention may be a computer program having program code for executing one of the methods described herein when the computer program is run on a computer. The invention may include a computer, or a programmable logic device, for performing one of the methods described above. Programmable logic devices (eg, field programmable gate arrays, complementary metal oxide semiconductor based logic circuits) may be used to perform some or all of the functions described above.

As mentioned above, although an embodiment of the present invention has been described, those of ordinary skill in the art may add, change, delete or add components within the scope not departing from the spirit of the present invention described in the claims. The present invention may be modified and changed in various ways, etc., which will also be included within the scope of the present invention.

100: terminal
200: service layer unit
300: equalizer layer portion
400: text recommendation engine hierarchy

Claims (9)

A plurality of text recommendation engines, logically located between the text recommendation engine and the administrator terminal, an adapter controlling the text recommendation device, a text recommendation engine setting storage unit for storing the setting information of the text recommendation engine, and attribute information of the text recommendation engine. In the text recommendation apparatus including a text recommendation engine attribute storage unit for storing,
Receiving, by the text recommendation device, at least one of first information, which is identification information of text recommendation engine to be used, second information, which is information on the number of text recommendation engines to be used, or third information, which is attribute information of each text recommendation engine, from the manager terminal; ;
Storing, by the adapter, at least one of the first information and the second information in the text recommendation engine setting storage unit or storing the third information in the text recommendation engine attribute storage unit;
Selecting, by the text recommendation apparatus, a text recommendation engine using at least one of the first information and the second information, or determining an attribute value of each text recommendation engine using the third information;
Selecting the recommended text by the selected text recommendation engine; And
And storing, by the text recommendation device, the selected recommendation text. The method of claim 1,
And receiving, by the text recommendation apparatus, fourth information that is information about a user who uses or uses recommended text. The method of claim 1,
And outputting the stored recommendation text to the user terminal by the text recommendation apparatus. The method of claim 3, wherein
Receiving, by the text recommendation device, fifth information regarding a user's response to the recommendation text output to the user terminal;
The adapter storing the fifth information in a feedback storage;
Modifying, by the text recommendation device, the first information or the third information based on the fifth information; And
The text recommendation apparatus further comprises the step of allowing the text recommendation engine to select the recommendation text. The method of claim 3, wherein
Receiving, by the text recommendation device, any one or more of first information, which is text recommendation engine identification information to be used from the administrator terminal, or second information, which is information on the number of text recommendation engines to be used, or third information, which is attribute information of each text recommendation engine; Is,
At least one of first information, which is text recommendation engine identification information to be used from an administrator terminal, or second information, which is information on the number of text recommendation engines, to be used from the administrator terminal according to the web page outputted to the user terminal, or the text recommendation apparatus of the text recommendation engine. Receiving third information, which is attribute information,
The step of selecting the recommended text by the selected text recommendation engine,
Selecting the recommended text by the selected text recommendation engine according to a web page output to the user terminal;
The outputting of the stored recommendation text to the user terminal may include:
And outputting the stored recommendation text to the user terminal according to the output web page of the user terminal. The method of claim 2,
The text recommendation apparatus may select a text recommendation engine using at least one of the first information and the second information, or determine the attribute value of each text recommendation engine using the third information.
The text recommendation apparatus selects a text recommendation engine for each user by using one or more of the first information or the second information and uses the fourth information, or uses the third information to attribute the text recommendation engine. Text recommendation method using a plurality of text recommendation engine, characterized in that the step of determining the value. The method of claim 3, wherein
Receiving, by the text recommendation device, any one or more of first information, which is text recommendation engine identification information to be used from the administrator terminal, or second information, which is information on the number of text recommendation engines to be used, or third information, which is attribute information of each text recommendation engine; Is,
At least one or each of the first information, which is text recommendation engine identification information to be used from the administrator terminal, or the second information, which is information on the number of text recommendation engines to be used, according to the divided section of the web page outputted to the user terminal by the text recommendation apparatus. Receiving third information which is attribute information of a text recommendation engine,
The step of selecting the recommended text by the selected text recommendation engine,
Selecting the recommended text by the selected text recommendation engine according to the divided section of the web page output to the user terminal;
The outputting of the stored recommendation text to the user terminal may include:
And outputting the stored recommendation text to the user terminal according to the divided section of the output web page of the user terminal. A plurality of text recommendation engines for selecting recommended texts;
A text recommendation engine setting storage unit configured to store setting information of the text recommendation engine;
A text recommendation engine attribute storage unit for storing attribute information of the text recommendation engine;
A data collection unit receiving one or more of first information, which is text recommendation engine identification information to be used from the administrator terminal, second information which is information on the number of text recommendation engines to be used, or third information, which is attribute information of each text recommendation engine; And
Logically located between the text recommendation engine and the manager terminal, and controls a text recommendation device and stores any one or more of the first information or the second information in the text recommendation engine setting storage unit or stores the third information in the text recommendation. A text recommendation device that includes an adapter that stores the engine property storage,
The text recommendation apparatus selects a text recommendation engine using at least one of the first information and the second information, or determines the attribute value of each text recommendation engine using the third information, and selects the selected recommendation text. Text recommendation apparatus utilizing a plurality of text recommendation engine, characterized in that for storing. The method of claim 8,
The text recommendation device receives fourth information that is information about a user who uses or uses the recommendation text, and the text recommendation device outputs the stored recommendation text to the user terminal, and outputs the recommended recommendation text to the user terminal. Receive fifth information about a user's response,
The adapter stores the fifth information in a feedback storage,
The text recommendation apparatus modifies the first information or the third information based on the fifth information, and causes the text recommendation engine to select the recommendation text. Recommended device.

Images