KR102146826B1 - Content recommendation method, and apparatas and system for providing the method - Google Patents

Abstract

The content recommendation apparatus according to the present invention includes a recommendation information collection unit that collects a plurality of recommendation target contents and generates a recommendation list, a first weight according to a purchase rate for each category of a user, and a priority among contents having the same recommendation reason. A recommendation information adjustment unit that calculates a second weight according to the result, and reflects the first weight and the second weight, adjusts the priority of the recommended content on the recommendation list, and provides the adjusted recommendation list to the content receiving device It includes a recommended information providing unit.

Description

Content recommendation method and devices and systems that provide it {CONTENT RECOMMENDATION METHOD, AND APPARATAS AND SYSTEM FOR PROVIDING THE METHOD}

The present invention relates to a content recommendation method, a device and a system providing the same, and more particularly, to implement an algorithm for recommending content optimized for a content receiving device such as a smartphone or a tablet PC, and recommendation information generated accordingly. The present invention relates to a content recommendation method capable of providing the content receiving device to the content receiving device, and an apparatus and system providing the same.

With the recent change in the communication environment represented by the development of mobile communication technology and the spread of smartphones, the demands of users for various high-quality multimedia services using mobile terminals such as smartphones and tablet PCs are increasing. Various techniques are being developed to provide content that can correspond to

The most appropriate method for inducing a user of a mobile terminal to purchase content is to recommend content that the user is expected to use or view.

However, in providing such content recommendation information, an algorithm that can determine the optimal content by appropriately considering the profile, purchase history, and current situation of each mobile terminal user is required. It is required to properly adjust according to the exposure order and provide it to the user.

As an algorithm for selecting the optimal content to be recommended to users, representatively, there are a Collaborate Filtering Algorithm and an Associate Rule method.

A phenomenon in which content corresponding to a specific category that an individual prefers is blocked and continuously exposed when a list is created by selecting the content to be recommended according to this method, and the recommended products are selected according to the individual's preferred category ratio. This happens often.

In addition, a phenomenon in which the recommendation reason and the same reason are continuously exposed may occur due to some popular contents that the user has already purchased.

Since the personalized content recommendation generates a content recommendation list optimized for each individual according to a certain algorithm, users often qualitatively and sensitively evaluate the suitability of the list recommended from the content recommendation device.

In other words, as in the prior art, when a number of content that one does not prefer is recommended or only similar content is blocked and exposed, the reliability and interest in the recommendation information decrease, and the case that the provision of the recommendation information leads to direct purchase will decrease. I have no choice but to.

In the content recommendation method, the present invention prevents a blockage phenomenon in which content belonging to a specific category in a content recommendation list or content selected for a specific recommendation reason is continuously exposed, so that various content can be recommended according to the actual preference of the user. It is an object of the present invention to provide a content recommendation method and an apparatus and system for performing the same.

According to a first aspect of the present invention, there is provided a content recommendation apparatus comprising: a recommendation information collection unit configured to collect a plurality of recommendation target contents and generate a recommendation list; Calculate a first weight according to a user's purchase rate for each category and a second weight according to a recommendation ranking within a content group having the same recommendation reason, and reflect the first weight and the second weight, and then on the recommendation list. A recommendation information adjustment unit that adjusts the priority of the content to be recommended; And a recommendation information providing unit that provides the adjusted recommendation list to the content receiving device.

Here, the recommendation information adjusting unit may adjust the exposure order of the content on the recommendation list according to the priority of the content.

Here, the recommendation information adjustment unit is the following Equation 1-

SCORE(x)는 컨텐츠 x의 스코어, S₀(x)는 상기 추천정보 수집부에서 최초 결정된 컨텐츠 x의 스코어, W₁(c,x)는 컨텐츠 x가 속하는 카테고리 c의 제1가중치, W₂(r,x)는 컨텐츠 x와 동일한 추천 사유에 의한 컨텐츠의 그룹 중 순위를 반영한 제2가중치 - 에 따라 상기 다수의 컨텐츠 각각의 스코어를 산출하고, 산출된 스코어의 순으로 상기 컨텐츠의 우선 순위를 조정할 수 있다.

SCORE (x) is the score of the content x, S ₀ (x) is the score of the content x initially determined by the recommendation information collection unit, W ₁ (c, x) is the first weight of the category c to which the content x belongs, W ₂ (r,x) is a second weight that reflects the ranking among the groups of contents based on the same recommendation reason as the contents x-according to the score of each of the plurality of contents, and the priority of the contents in the order of the calculated scores Can be adjusted.

Here, the recommendation information adjusting unit may generate the recommendation list including only content in which the score calculated according to Equation 1 is included in a preset ranking.

Here, a purchase ratio analysis unit for analyzing a purchase ratio for each content category of the user may be further included, and the recommendation information adjustment unit may calculate the first weight based on the analyzed purchase ratio for each category.

Here, the recommendation information collection unit may generate the recommendation list by collecting content to be recommended based on the user's purchase history or personal information.

Here, it may further include a content providing unit for providing the content selected on the basis of the list of the provided recommended target content to the content receiving device.

In addition, according to a second aspect of the present invention, a method for recommending content, comprising: collecting a plurality of content to be recommended, and generating a recommendation list; Calculating a first weight value according to a purchase rate for each category of a user; Calculating a second weight according to a recommendation ranking within a content group having the same recommendation reason; Adjusting the priority of the recommendation target content on the recommendation list by reflecting the calculated first weight value and the second weight value; And providing the adjusted recommendation list to a content receiving device.

SCORE(x)는 컨텐츠 x의 스코어, S₀(x)는 최초 결정된 컨텐츠 x의 스코어, W₁(c,x)는 컨텐츠 x가 속하는 카테고리 c의 제1가중치, W₂(r,x)는 컨텐츠 x와 동일한 추천 사유에 의한 컨텐츠의 그룹 중 순위를 반영한 제2가중치 - 에 따라 상기 다수의 컨텐츠 각각의 스코어를 산출하고, 산출된 스코어의 순으로 상기 컨텐츠의 우선 순위를 조정할 수 있다.Here, the step of adjusting the priority, the following Equation 1-

SCORE(x) is the score of the content x, S ₀ (x) is the score of the initially determined content x, W ₁ (c,x) is the first weight of the category c to which the content x belongs, and W ₂ (r, x) is A score of each of the plurality of contents may be calculated according to a second weight value reflecting a ranking among a group of contents based on the same recommendation reason as the contents x, and the priority of the contents may be adjusted in the order of the calculated scores.

Here, in the step of adjusting the priority, the recommendation list may be adjusted by including only contents in which the score calculated according to Equation 1 is included in a preset priority.

Here, in the step of generating the recommendation list, the recommendation list may be generated by collecting content to be recommended based on a user's purchase history or personal information.

Here, the method may further include providing the content selected on the basis of the provided recommendation list to the content receiving device.

In addition, according to the third aspect of the present invention, in a content recommendation system, a recommendation list is generated by collecting a plurality of content to be recommended, and a first weight according to a purchase rate for each category of a user and the same recommendation reason Calculate a second weight according to a recommendation order within a content group having, reflecting the first weight and the second weight, adjust the priority of the recommended content on the list, and add the adjusted recommendation list to the content A content recommendation device provided to the receiving device; And a content receiving device that receives the recommendation list from the content recommendation device and provides the recommendation list to a user according to the priority.

According to the present invention described above, a content recommendation method capable of recommending various contents according to the user's actual preference by preventing a blockage phenomenon in which contents belonging to a specific category in the contents recommendation list or contents selected for a specific recommendation reason are continuously exposed. And it is possible to provide an apparatus and system for performing this.

1 is a block diagram of a content recommendation system according to the present invention.
2A to 2C illustrate an example in which the priority of a recommendation list is adjusted.
3 is a block diagram of a content recommendation method according to the present invention.

It should be noted that the technical terms used in the present specification are only used to describe specific embodiments, and are not intended to limit the present invention. In addition, the technical terms used in the present specification should be interpreted as generally understood by those of ordinary skill in the technical field to which the present invention belongs, unless otherwise defined in the present specification, and excessively comprehensive It should not be construed as a human meaning or an excessively reduced meaning. In addition, when a technical term used in the present specification is an incorrect technical term that does not accurately express the spirit of the present invention, it should be replaced with a technical term that can be correctly understood by those skilled in the art. In addition, general terms used in the present invention should be interpreted as defined in the dictionary or according to the context before and after, and should not be interpreted as an excessively reduced meaning.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings, but the same or similar components are assigned the same reference numerals regardless of the reference numerals, and redundant descriptions thereof will be omitted. In addition, in describing the present invention, when it is determined that a detailed description of a related known technology may obscure the subject matter of the present invention, a detailed description thereof will be omitted. In addition, it should be noted that the accompanying drawings are intended to make the spirit of the present invention easier to understand, and should not be construed as limiting the spirit of the present invention by the accompanying drawings. The spirit of the present invention should be construed as extending to all changes, equivalents, or substitutes in addition to the accompanying drawings.

1 is a block diagram of a content recommendation system according to the present invention.

As shown, the content recommendation system according to the present invention may include a content recommendation device 100 and a plurality of content reception devices 200.

The content receiving device 200 is a device capable of receiving and executing content such as applications, broadcasts, movies, music, and e-books, and is implemented as a smartphone, tablet PC, desktop PC, laptop PC, PDA, digital TV, etc. Can be.

The content receiving device 200 can access the web server of the content recommending device 100 using a wired/wireless communication network based on an IP (internet protocol), receives recommendation information to be described later, and a recommendation list of recommendation information is displayed. When the user selects to purchase at least one content included in the recommendation list, the content may be received from the content recommendation device 100.

The content recommendation device 100 is implemented as a server that provides content recommendation information and content to a plurality of content receiving devices 200 connected through an IP network, and collects and analyzes information for generating content recommendation information. Perform the algorithm.

Hereinafter, a detailed configuration of the content recommendation apparatus 100 will be described in detail.

As shown in FIG. 1, the content recommendation device 100 includes a recommendation information collecting unit 110, a recommendation information adjusting unit 120, a recommendation information providing unit 130, a database unit 140, and a purchase rate analysis unit 150. ), may include a content provider.

In this case, according to a specific embodiment of the present invention, even if some of the components shown in FIG. 1 are omitted, there is no problem in implementing the content recommendation apparatus 100 according to the present invention.

The recommendation information collection unit 110 generates a recommendation list by collecting a plurality of recommendation target contents.

The recommendation information collection unit 110 may generate a personalized recommendation list for the user of each content receiving device 200 by collecting recommended content based on the user's purchase history or personal information. As an algorithm that collects, for example, a Collaborate Filtering Algorithm based on personal information and an Associate Rule based on purchase history are examples.

More specifically, the collaborative filtering algorithm is a method of identifying a user's disposition through data mining and recommending a content product by referring to content preferred by other users having a disposition similar to the identified user's disposition, The purchase association method is a method of collecting the user's purchase history and recommending content products based on the fact that there is a high possibility that the product that was preferred in the past is likely to be preferred in the future.

The recommendation list generated by the recommendation information collection unit 110 includes information on the title, category, and reason for recommendation of the content, and an initial score may be determined according to an algorithm used initially.

For example, as shown in FIG. 2A, if eight contents (comedy, action, doll, raccoon, number, cartoon, novel, poetry collection) are collected through a specific algorithm, the content receiving device ( 200) may be converted into a score and expressed as an initial score.

In this way, when the content list collected by the recommendation information collection unit 110 is provided to the content receiving device 200 as it is, it is inevitable to provide a block of content list for each category as in the present.

For example, as shown in FIG. 2A, the priority is determined in the order of contents belonging to the movie, game, and e-book category, and accordingly, the user of the content receiving device 200 receiving the recommendation information is temporal or Since information of content with high spatial priority can be checked first, it is inevitable to check recommended information blocked by category.

Accordingly, the content recommendation apparatus 100 according to the present invention may adjust the content recommendation information so that each content is evenly distributed according to a category and recommendation reason to prevent such blocking.

The priority of the content on the recommendation list refers to a temporal or spatial exposure priority when the content receiving device 200 displays the recommendation list. For example, the content with high priority is exposed first in time or the recommendation list screen It can be highlighted and displayed so that it is more easily identified by the user.

The database unit 140 performs a function of storing a database related to content recommendation in the content recommendation device 100, and a user information storage unit 142, a purchase history storage unit 144, a recommendation history storage unit 146, A content storage unit 148 may be included.

The user information storage unit 142 stores a profile of a user of the content receiving device 200 who purchases content.

For example, user information such as gender, age, access time zone, and preferred content category of the user of the content receiving apparatus 200 may be stored.

The purchase history storage unit 144 collects and stores content purchase history information of a plurality of content receiving devices 200.

Specifically, the purchase history of each content receiving device 200 may be stored by date, time, or category.

The recommendation history storage unit 146 may store a history of recommendation information provided by the content recommendation device 100 to each content receiving device 200.

Specifically, information such as whether the user actually purchased the content may be stored in correspondence with the history of the content recommended by the content category or the reason for the recommendation, and the recommendation information.

The content storage unit 148 stores content files including broadcasts, movies, music, e-books, and cartoons.

The recommendation information adjustment unit 120 adjusts the priority of the recommendation target content on the recommendation list generated by the recommendation information collection unit 110 by reflecting the first weight value and the second weight value.

In the present invention, the first weight value is determined according to a content purchase rate for each category of a user of the content receiving apparatus 200.

For example, among the contents purchased by the user of the content receiving device 200 for a certain period, the ratio of the game category is 40%, the ratio of the movie category is 35%, and the ratio of the e-book category is 25%. In this case, each purchase ratio may be a first weight.

In Figure 2a, the content collected by the recommendation information collection unit 110 is a doll, a raccoon, a number, a comedy. Priority was determined in the order of action, cartoon, novel, and poetry. When the initial score of each content is multiplied by the first weight value of 0.4, 0.35, and 0.25 according to the purchase rate for each category, the score is as shown in FIG. 2B. Change, and the priority may change accordingly.

The content recommendation device 100 according to the present invention may include a purchase rate analysis unit 150 that analyzes a purchase rate for each content category of a user of each content receiving device 200 in order to calculate a first weight. , The purchase rate analysis unit 150 may analyze the content purchase history information of the plurality of content receiving devices 200 stored in the purchase history storage unit 144 to analyze the purchase rate by category of each content receiving device 200. have.

In the present invention, the second weight may be assigned according to a recommendation ranking within a content group having the same recommendation reason.

The reason for the recommendation may be determined based on the user's purchase history or personal information. For example, content frequently purchased by users of similar gender, age, and occupation, and many other users who recently purchased the content purchased by the user One content, content having a content similar to the content purchased by the user in the vicinity, may be an example of a recommendation reason.

As described above, the recommendation list includes information on the recommendation reason of each recommendation target content, and the recommendation information adjustment unit 120 groups contents having the same recommendation reason among the recommendation target content, and the recommendation ranking within each group According to the second weight can be calculated.

For example, as shown in FIG. 2C, as an example of the reason for recommendation, the content preferred by other users who have purchased the movie X previously purchased by the user, and the doll game, action, and poetry collection based on the recommendation reason Content can be included on the recommendation list.

At this time, in the state of reflecting the first weight, a comedy, a raccoon dog, and a number corresponding to popular content in the 30s with a recommendation reason are generated as a group, and the second weight is 1, 0.8, respectively, in the order of the highest score. 0.6 can be given.

Here, the method of ranking the content groups having the same recommendation reason and the ratio of the second weight are not limited to those described above, and may be changed according to the settings of the content recommendation apparatus 100.

As described above, the method of adjusting the priority of the content to be recommended by reflecting the first weight value and the second weight value by the recommendation information adjustment unit 120 calculates a score for each content according to Equation 1 below, and the calculated score Priority can be determined in the order of.

Equation 1

Here, SCORE(x) is the score of the content x, S ₀ (x) is the score of the content x initially determined by the recommendation information collecting unit 110, and W ₁ (c,x) is the first category c to which the content x belongs. The 1 weight value, W ₂ (r,x) means the second weight value reflecting the ranking among the groups of content based on the same recommendation reason as the content x.

As described above, FIG. 2A shows the initial score S ₀ (x) of the content to be recommended collected by the recommendation information collection unit 110, and FIG. 2B shows the first weight W ₁ (c,) to S ₀ (x). The score adjusted by multiplying by x) is shown, and FIG. 2A shows the score finally adjusted by multiplying by the second weight W ₂ (r,x).

Referring to FIG. 2A, the priority of the content to be recommended for each category is initially blocked, but referring to FIG. 2C in which the first weight and the second weight are reflected by the recommendation information adjustment unit 120, the priority can be referred to. At this time, it can be confirmed that the categories and recommendation reasons of each content are evenly distributed.

Meanwhile, the recommendation information adjustment unit 120 may generate the recommendation list including only contents in which the score calculated according to Equation 1 is included in a preset ranking.

For example, in FIG. 2C, a comedy, doll, action, raccoon dog, and cartoon having the highest priority among eight contents are selected, and a recommendation list including the five contents is generated, and the contents receiving device 200 Can provide.

The recommendation information providing unit 130 provides the recommendation list adjusted by the recommendation information adjustment unit 120 to the content receiving apparatus 200 through an IP network.

The content receiving device 200 receiving the recommended information displays a user interface through which content can be purchased based on the received content list when executing a content-related application, and the user requests content purchase through the user interface. In this case, it is possible to request the content recommendation device 100 to transmit the content.

As described above, the content list received by the content receiving device 200 includes information on at least one content, and may include the priority of each content, according to the setting of the content receiving device 200 Only contents corresponding to a predetermined number of higher priorities may be displayed through the user interface.

In addition, the content receiving device 200 may induce a user to purchase a content by displaying information such as a name, a category, a brief content, the number of downloads, and evaluation of other users according to the recommendation information.

The content providing unit transmits the requested content to the content receiving device 200 according to the request of the content receiving device 200.

The content provider may transmit content stored in the content storage unit 148 or content received through an IP network from an external content server to the content receiving device 200.

The content recommendation apparatus 100 according to the present invention described above includes a first weight according to a purchase rate for each category of a user and a second weight according to a recommendation ranking within a content group having the same recommendation reason. By adjusting the priority, it is possible to recommend various contents according to the user's actual preference by preventing a blockage phenomenon in which contents belonging to a specific category on the contents recommendation list or contents selected according to a specific recommendation reason are continuously exposed.

3 is a flowchart of a content recommendation method according to the present invention.

Hereinafter, descriptions of technical features identical to those of the present invention described with reference to FIG. 1 will be omitted.

The content receiving device 200 according to the present invention is a device capable of receiving and executing content, and may be implemented as a smartphone, tablet PC, desktop PC, laptop PC, PDA, digital TV, etc., and content recommendation device 100 May be implemented as a server that serves to provide content recommendation information and content to a plurality of content receiving devices 200 connected through an IP network.

The content recommendation device 100 collects a plurality of content to be recommended and generates a recommendation list (S110).

At this time, based on the user's purchase history or personal information, it is possible to collect the content to be recommended to generate a personalized recommendation list for the user of each content receiving device 200. As an algorithm for collecting the content to be recommended Examples are the Collaborate Filtering Algorithm based on personal information and the Associate Rule based on purchase history.

An example of a recommendation list including the collected recommendation target content is as described above with reference to FIG. 2A.

The content recommendation device 100 calculates a first weight value according to a user's purchase rate for each category (S120).

As described above with reference to FIG. 2B, the first weight value is 40% of the game category, 35% of the movie category, and e-book (e-book) among the content purchased by the user of the content receiving device 200 for a certain period. When the ratio of the book) category is 25%, each purchase ratio may be the first weight, and the content recommendation device 100 is the user's content of each content receiving device 200 to calculate the first weight. You can analyze the purchase rate by category.

The content recommendation apparatus 100 calculates a second weight according to a recommendation ranking within a content group having the same recommendation reason (S130).

As described above, the recommendation list includes information on the recommendation reason for each recommendation target content, and the recommendation reason may be determined based on the user's purchase history or personal information. For example, similar gender, age, occupation Contents frequently purchased by users having a value, contents purchased by many other users who have recently purchased contents, and contents having contents similar to contents purchased by the user may be reasons for recommendation.

The content recommendation apparatus 100 may group content having the same recommendation reason among the content to be recommended, and calculate a second weight value according to a ranking within each group.

The content recommendation apparatus 100 adjusts the priority of the content to be recommended on the recommendation list by reflecting the calculated first weight value and the second weight value (S140).

In this case, as a method of adjusting the priority of the content to be recommended, a score for each content may be calculated according to Equation 1 below, and the priority may be determined in the order of the calculated scores.

Equation 1

Here, SCORE(x) is the score of the content x, S ₀ (x) is the score of the content x initially determined by the recommendation information collecting unit 110, and W ₁ (c,x) is the first category c to which the content x belongs. The 1 weight value, W ₂ (r,x) means the second weight value reflecting the ranking among the groups of content based on the same recommendation reason as the content x.

As described above with reference to FIGS. 2A to 2C, the priority of the content to be recommended for each of the same categories was initially blocked, but referring to FIG. 2C in which the first weight and the second weight are reflected, each content category and You can see that the reasons for recommendation are evenly distributed.

The content recommendation device 100 provides the adjusted recommendation list to the content receiving device 200 (S150).

Thereafter, when the content receiving device 200 requests the purchase of content based on the recommendation list, the content recommending device 100 receives the content stored in the database or the content received from an external content server. 200) can be transmitted (S160).

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

In the content recommendation method according to the present invention, the priority of the content to be recommended collected according to the first weight value according to the purchase rate for each category of the user and the second weight value according to the recommendation order within the content group having the same recommendation reason By adjusting, there is an effect of preventing a blockage phenomenon in which content belonging to a specific category on the content recommendation list or content selected for a specific recommendation reason is continuously exposed, and thus various content can be recommended according to the actual preference of the user.

The embodiments of the present invention have been described above, but those of ordinary skill in the relevant technical field can add, change, delete or add components within the scope not departing from the spirit of the present invention described in the claims. Accordingly, the present invention can be variously modified and changed, and it will be said that this is also included within the scope of the present invention.

According to the content recommendation apparatus 100 and the content recommendation method of the present invention described above, a recommendation target collected according to a first weight according to a purchase rate for each category of a user and a second weight according to a priority among contents having the same recommendation reason By adjusting the priority of the content, it is possible to prevent the blockage phenomenon in which content belonging to a specific category on the content recommendation list or content selected for a specific recommendation reason is continuously exposed. As a result, it is possible to overcome the limitations of the existing technology. It is an invention that has industrial applicability because it is not only possible to market or sell the applied device, but also to the extent that it can be implemented clearly in reality.

100: content recommendation device 200: content receiving device
110: recommendation information collection unit 120: recommendation information adjustment unit
130: recommendation information providing unit 140: database unit
150: purchase ratio analysis unit 160: content providing unit

Claims (13)

In the content recommendation device,
A recommendation information collection unit that collects a plurality of recommendation target contents and generates a recommendation list;
Calculate a first weight according to a user's purchase rate for each category and a second weight according to a recommendation ranking within a content group having the same recommendation reason, and reflect the first weight and the second weight, and then on the recommendation list. A recommendation information adjustment unit that adjusts the priority of the content to be recommended; And
Including a recommendation information providing unit for providing the adjusted recommendation list to the content receiving device,
The recommendation information adjustment unit creates one or more content groups by grouping the plurality of recommended target contents among contents having the same recommendation reason, and according to a recommendation ranking within a content group to which the recommended target content belongs for each of the plurality of recommended target contents. Calculate the second weight,
A score for each recommendation target content included in the recommendation list is calculated using the first weight and the second weight, and the priority is adjusted in the order of the highest score, and for each of the recommendation target content according to the priority. And adjusting the exposure order on the recommendation list, and adjusting the recommendation list so that the recommendation target content having a high priority is exposed first in the content receiving device.
delete ◈ Claim 3 was abandoned upon payment of the set registration fee. The method of claim 1,
The recommendation information adjusting unit calculates scores for each of the plurality of contents according to Equation 1 below, and adjusts the priority of the contents in the order of the calculated scores.
Equation 1

SCORE (x) is the score of the content x, S ₀ (x) is the score of the content x initially determined by the recommendation information collection unit, W ₁ (c, x) is the first weight of the category c to which the content x belongs, W ₂ (r,x) is a second weight that reflects the ranking among the groups of contents based on the same recommendation reason as the contents x.
◈ Claim 4 was abandoned upon payment of the set registration fee. The method of claim 3,
The recommendation information adjustment unit generates the recommendation list by including only the content whose score calculated according to Equation 1 is included in a preset ranking.
delete delete delete In the content recommendation method,
Collecting a plurality of recommendation target contents and generating a recommendation list;
Calculating a first weight value according to a purchase rate for each category of a user;
Calculating a second weight according to a recommendation ranking within a content group having the same recommendation reason;
Adjusting the priority of the recommendation target content on the recommendation list by reflecting the calculated first weight value and the second weight value; And
Providing the adjusted recommendation list to a content receiving device
Including,
In the calculating of the second weight, one or more content groups are generated by grouping the plurality of recommended target contents among contents having the same recommendation reason, and within a content group to which the recommended target content belongs for each of the plurality of recommended target contents. Further comprising the step of calculating the second weight according to the recommendation ranking,
The step of adjusting the priority includes calculating a score for each content to be recommended included in the recommendation list using the first weight and the second weight, adjusting the priority in the order of the highest score, and the priority The method further comprising: adjusting an exposure priority on the recommendation list for each recommendation target content according to a priority, and adjusting the recommendation list so that the recommendation target content having a higher priority is first exposed in the content receiving device. How to recommend content.
◈ Claim 9 was abandoned upon payment of the set registration fee. The method of claim 8,
The step of adjusting the priority,
A content recommendation method for calculating a score of each of the plurality of contents according to Equation 1 below, and adjusting the priority of the contents in the order of the calculated scores.
Equation 1

SCORE(x) is the score of the content x, S ₀ (x) is the score of the initially determined content x, W ₁ (c,x) is the first weight of the category c to which the content x belongs, and W ₂ (r, x) is The second weight that reflects the ranking among the groups of contents based on the same recommendation reason as the contents x
◈ Claim 10 was abandoned upon payment of the set registration fee. The method of claim 9,
The step of adjusting the priority may include adjusting the recommendation list by including only content in which the score calculated according to Equation 1 is included in a preset priority.
delete delete delete

Images