KR101958555B1 - Apparatus and method for providing search result - Google Patents

Abstract

To a device and method for providing search results, and more particularly to an apparatus and method for providing search results in response to a user's propensity. According to an embodiment of the present invention, a search result providing apparatus for providing a search result that reflects a tendency of a user, the search result providing apparatus collecting a first user's social activity, analyzing a social activity, A determination unit for calculating a frequency of words and extracting at least one or more words of interest detected with a frequency equal to or greater than a predetermined threshold value, a classifying unit for classifying the interest words into the related topics, And calculating a tendency score of a first user for any one of the topics by assigning a time weight and a frequency weight to at least one interest word related to the query, A first week having a high tendency score among at least one topic corresponding to a word included in the first week The content related to the subject can be provided as a top search result.

Description

[0001] APPARATUS AND METHOD FOR PROVIDING SEARCH RESULT [0002]

To a device and method for providing search results, and more particularly to an apparatus and method for providing search results in response to a user's propensity.

With the development of digital and information technology, the importance of web search is increasing. However, most Web search results do not provide users with tailored information that is optimized. Often, providing universal search results does not provide the search results that users require.

In recent years, there has been an increasing demand for personalized search that considers personal tendencies rather than providing a universal search result. The personalized search is used to expose the information desired by the user to the top of the search result based on the user's information.

According to one aspect of the present invention, there is provided a search result providing apparatus for providing a search result that reflects a tendency of a user, the search result providing apparatus collecting social activities of a first user and analyzing the social activities to detect at least one word A determination unit for calculating a frequency of the word and extracting at least one or more words of interest detected with a frequency greater than or equal to a predetermined threshold value; a classification unit for constructing an ontology by classifying the interest word by a related topic; A first calculation unit for calculating a tendency score of a first user for any one of the topics by assigning a time weight and a frequency weight to at least one or more interest words related to any one of the first and second topics, At least one or more words corresponding to the words included in the query that the user searches It is possible to provide contents related to the first topic higher than the predetermined threshold value as the upper search result.

According to an embodiment, the first operation unit can give the time weight less as the current time and the past time interval in which the interesting word is input into the social activity are longer.

According to another embodiment, the first calculation unit may assign the higher frequency weight to the extracted interest word as the frequency calculated by the determination unit becomes higher.

According to still another embodiment, a difference between the propensity score of the first user and the propensity score of the user other than the first user is calculated, and a second user having a difference of less than a predetermined threshold is referred to as the first user And a second calculation unit for determining a tendency of the second user to be similar to the first user, and the tendency score of the second user among the at least one topic corresponding to the word included in the query retrieved by the first user is determined Content related to the second topic higher than the threshold value can be provided as an upper search result.

In accordance with another embodiment, for at least one or more users who perform the social activity on any one of the topics, taking into consideration the participation of other users in recommending, commenting, and sharing other users' And a third arithmetic unit for calculating the expertise score of each user on any one of the topics.

According to yet another embodiment, the third calculation unit may calculate a third user whose expertise score is equal to or greater than a predetermined threshold, and a third user whose expertise score is equal to or greater than the threshold, The second user may be determined as an expert for the first user, and if the third topic is included among at least one topic corresponding to a word included in the query retrieved by the first user, Of the at least one topic corresponding to the topic, the content of the topic related to the topic whose tendency score of the third user is higher than a predetermined threshold, as an upper search result.

According to still another embodiment, the mobile communication terminal may further include a fourth operation unit for calculating a network connection score by weighting according to the social activity relationship between the first user and the user other than the first user.

According to another embodiment, the fourth calculation unit may determine that a user other than the first user whose network score is equal to or higher than a predetermined threshold is a fourth user, and the fourth user is included in the query It is possible to provide contents related to the fourth subject having the tendency score of the fourth user higher than the predetermined threshold among the at least one topic corresponding to the word as an upper search result.

According to another embodiment, the content related to the first subject may be provided as a search result higher than the contents related to the second subject.

According to another embodiment, the content related to the first subject may be provided as a search result higher in content than the content related to the topic higher than the predetermined threshold of the propensity score of the third user.

According to another aspect of the present invention, there is provided a search result providing method for providing a search result that reflects a tendency of a user, the method comprising: collecting a social activity of a first user; analyzing the social activity to detect at least one word Extracting at least one word of interest that is detected as a frequency greater than or equal to a predetermined threshold value; constructing an ontology by classifying the word of interest into related topics; Calculating a tendency score of a first user for any one of the topics by assigning a time weight and a frequency weight to at least one or more interest words related to one of the topics, Of the at least one topic corresponding to the word And providing the contents related to the first topic higher than the predetermined threshold as an upper search result.

According to one embodiment, the calculating step may give a smaller time weight as the current time and the past time interval in which the interest word is entered into the social activity are shorter.

According to another embodiment, the calculating step may assign the frequency weight higher as the frequency of the extracted interest words is higher.

According to another embodiment of the present invention, a difference between the propensity score of the first user and the propensity score of a user other than the first user is calculated, and a second user having a difference equal to or less than a predetermined threshold value, And determining whether the second user's tendency score is higher than a predetermined threshold value among at least one or more topics corresponding to words included in the query retrieved by the first user, And providing the contents as the upper search result.

According to another embodiment of the present invention, for a plurality of users who perform the social activity on any one of the topics, taking into consideration the participation of other users in recommending, commenting, and sharing other users with respect to the social activity of each user, And calculating the expertise score of each user for any one of the topics.

According to yet another embodiment, determining a third user whose expertise score is equal to or greater than the threshold value as an expert for the third topic, for a third topic whose expertise score of the first user is equal to or greater than a predetermined threshold And when the third topic is included in at least one topic corresponding to a word included in a query retrieved by the first user, the content of the third user's inclination score is related to a topic higher than a predetermined threshold value As a result of the upper search.

According to another embodiment of the present invention, the method may further include calculating a network score by weighting the network based on the social activity relationship between the first user and the user other than the first user.

According to another embodiment of the present invention, there is provided a method of searching for a user, the method comprising the steps of: determining a user other than the first user whose threshold value is equal to or greater than a predetermined threshold as a fourth user; And providing the contents related to the fourth topic, which is higher than the predetermined threshold, among the one or more topics as the upper search result.

According to another embodiment, the content related to the first subject may be provided as a search result higher than the contents related to the second subject.

According to another embodiment, the content related to the first subject may be provided as a search result higher than the content related to the topic higher than the predetermined threshold of the tendency score of the third user.

FIG. 1 is a block diagram including a configuration of a search result providing apparatus according to an embodiment.
2 shows a block diagram including a configuration of a search result providing apparatus according to another embodiment.
FIG. 3 illustrates a keyword ontology building method for determining a user's propensity according to an embodiment.
FIG. 4 illustrates a method for determining a user with a similar tendency to a user, according to one embodiment.
5 shows a ranking algorithm according to one embodiment.
6 is a flowchart illustrating an overall method of providing a search result according to an embodiment.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. However, the scope of the rights is not limited or limited by these embodiments. Like reference symbols in the drawings denote like elements.

The terms used in the following description are chosen to be generic and universal in the art to which they are related, but other terms may exist depending on the development and / or change in technology, customs, preferences of the technician, and the like. Accordingly, the terminology used in the following description should not be construed as limiting the technical thought, but should be understood in the exemplary language used to describe the embodiments.

Also, in certain cases, there may be a term chosen arbitrarily by the applicant, in which case the meaning of the detailed description in the corresponding description section. Therefore, the term used in the following description should be understood based on the meaning of the term, not the name of a simple term, and the contents throughout the specification.

With the development of Internet technology and mobile devices, Social Media Service also developed rapidly. Internet technology enables quick access to social media services through mobile devices or the web to generate information quickly and conveniently. Social media services are actively being developed and used as means of producing, consuming and sharing information, and the number of users using the service is rapidly increasing.

Social media services are characterized by the ability of users to directly produce, process, and share information, and the process is simple and convenient to spread quickly. Social data is increasing rapidly as a result of social media activities of many users, and it is possible to grasp the tendency of users to be used for personalized search.

FIG. 1 is a block diagram including a configuration of a search result providing apparatus according to an embodiment.

Referring to FIG. 1, the search result providing apparatus 100 may include a detecting unit 110, a determining unit 120, a classifying unit 130, and a first calculating unit 140. The detection unit 110 may collect data on the social activities of the first user and may detect words from the collected social data using, for example, a text mining technique. Alternatively, words can be detected using a morphological analyzer. The method of detecting a word is not limited to a specific method.

The determination unit 120 may extract at least one word of interest of the first user by applying a TF-IDF (Term Frequency-Inverse Document Frequency) technique to the detected word. TF stands for word frequency, which indicates how often a particular word appears in the document, and IDF is the frequency of reverse document, indicating how many documents have a specific word in the whole document. The higher the TF-IDF value, the more frequently the document appears in the document, and the higher the TF-IDF value, the more likely the first user will be in the field of interest. can do.

The classifying unit 130 may classify the extracted first user's interest words into categories. At least one or more words that can be tied to one field of interest can be categorized using a built-in database. With the category of interest of the first user, at least one or more words may be classified to be included in each category, and the ontology including the classified category and words may be constructed. The category may correspond to a topic or interest that the user is interested in.

The first arithmetic unit 140 may perform a calculation for assigning weights of the detected interest words to the time and the frequency. In order to determine the recent tendency of the first user, the tendency of the first user may be weighted over time in order to reduce the weight of the old tendency and increase the weight of the recent tendency.

Equation (1) shows a formula for giving a time and frequency weight to a word of interest detected from a social activity, summing values of a plurality of interest words included in one category, and deriving an interest degree for the category as TW. n is the number of words included in the category of interest, twi is the weight over time, and cnti is the frequency of the word.

Equation (2) represents a mathematical expression giving a weight according to time, in one embodiment. t denotes the time from the present time to the time of the social activity in which the word is detected. Referring to Equation (2), it can be seen that the lower the weight is, the more the detected word is in the past.

The higher the frequency of the words included in a certain field, the higher the TW value of the field. The more the recent words are included in the search field, the higher the TW value of the field. The higher the TW value, the more recent interest in the field may indicate.

2 shows a block diagram including a configuration of a search result providing apparatus according to another embodiment.

Referring to FIG. 2, the search result providing apparatus 200 may include a first calculation unit 210, a second calculation unit 220, a third calculation unit 230, and a fourth calculation unit 240. The first calculation unit 210 may perform a calculation of weighting time and frequency for at least one word of interest extracted from a social activity. A word extracted from a social activity of a past time point can be given a lower time weight, and a word extracted with a larger frequency number can be given a higher frequency weight. The average value of the time and frequency weighted values of at least one or more words included in one interest category may be used as the first user's tendency score for the area of interest.

The second calculation unit 220 may perform a calculation for determining a second user having a tendency similar to that of the first user.

Equation (3) is a calculation formula for determining a second user having a tendency similar to that of the first user by obtaining the difference between the propensity score of the first user and the propensity score of the user other than the first user.

Referring to Equation (3), n is the number of propensity categories of the user to be pre-classified in the ontology, and k is the number of matching prop- erties of the user and other users. S1 representing the difference in tendency between the users is obtained by dividing the total sum of the tendency scores TW of the users other than the first user and the first user for the k categories by the total inclination category number of the first user Can be obtained.

A user having the smallest S1 value among at least one user other than the first user may be extracted and the user may be determined as a second user having a similar tendency to the first user.

The similar user's tendency can be reflected in providing search results of the first user as follows. For example, assuming that the first user and the second user perform a lot of activities in the field of " sports ", the first user receives a search result reflecting the tendency of the second user together with a search result reflecting his or her tendency . In this way, information that may be missing can be provided together with the search results.

The third arithmetic operation unit 230 may calculate the expertise score for at least one user who performs social activities. The expertise score may be determined by taking into account participation of other users' recommendations, comments, and sharing on the user's social activities. According to one embodiment, the higher the other user's participation in the user's social activity, the higher the expertise in the corresponding field corresponding to the user's social activity.

Equation (4) represents an equation for calculating a user's expertise score for a specific field. N is the number of activities of the user, N is the number of other users participating in the user's activity, NRi (i = 1,2,3 ... N) is the number of recommendations, NSi , 3 ... N) is the number of shares and NCi (i = 1, 2, 3 ... N) represents the number of comments. Elements representing participation activities of other users participating in other users' activities may also be considered.

A method of providing information to a search result that reflects a tendency of a first user by selecting an expert may be provided as follows. For example, when the first user is interested in the field of " sports ", the third computing unit 230 may perform a calculation to determine a third user with a higher expertise score S2 for the & have. The information provided by the determined third user or the information reflecting the tendency of the third user may be provided to the search result of the first user.

Or if the first user is an expert in the field of " sports ", it is necessary to provide the first user with professional knowledge in the field of " sports ". In this case, the third operation unit 230 determines a third user, which is an expert in the field of 'sports', and the search result reflecting the information provided by the third user or the tendency of the third user is included in the search result of the first user .

The fourth calculation unit 240 may calculate a networking score by weighting the users according to the networking relationship on the social media service.

Equation (5) shows a formula for giving a higher weight to the nearest distance on the social media by applying 'Kevin Bacon's six-step rule' according to an embodiment. Referring to Equation (5), x represents a social media service distance between a first user and a user other than the first user. For example, for a user who is directly friends with a first user, x may be one. In the case of a friend who is directly friend with a first user and a friend who is not a friend directly to the first user, the user can make a connection with each other through two times, and x may be 2. In the case of calculating the network score in this manner, the weight is not given if the network is connected to each other over the network more than seven times, and the user who is used to provide the information in consideration of the network in providing the search result of the first user Not classified.

The closer the user is to the first user, the higher the probability of having a tendency in common with the first user. The fourth computing unit 240 may determine a user having a high S3 value as a fourth user in relation to the first user. When there is a word included in a field having a high tendency score of the fourth user among the words extracted and searched by the first user, the result reflecting the tendency of the fourth user in the field can be provided as a search result of the first user have.

FIG. 3 illustrates a keyword ontology building method for determining a user's propensity according to an embodiment.

Referring to FIG. 3, at least one word (keyword) may be extracted from the social activity 300 of the first user. The keyword may be extracted using the TF-IDF (310) technique. That is, the frequency with which the word appears is high and words searched in many documents can be extracted from the social activity 300 of the first user.

The extracted keywords may be classified into at least one or more words corresponding to a specific category, and the weight of the time and the frequency may be calculated 330 for the classified keywords. An average of at least one weighted word belonging to the category may be calculated to calculate a first user's tendency score for the category and a tendency score of the first user may be compared for at least one category. Weighted words that are categorized by category and by category can be built on the ontology.

When the first user performs the search, the user's interest field corresponding to the keyword can be determined from the keyword extracted from the search from the ontology, and information on the corresponding field is provided as the search result to reflect the user's tendency Search results can be provided.

FIG. 4 illustrates a method for determining another user having a high degree of user similarity with a user according to an embodiment.

Referring to FIG. 4, a first user may calculate (410) S1, expertise score S2 and social network score S3 for each of at least one user 400 on a social media service having social activities, . The final user similarity with the first user can be calculated by assigning different weights to S1, S2, and S3, as the case may be.

Equation (6) represents a user similarity calculation formula. According to one embodiment, the sum of?,? And? May be equal to one. The values of alpha, beta and gamma may be determined according to the criterion determined and the user 430 most similar to the first user may be determined using the calculated user similarity (USIM).

According to one embodiment, the weight values of?,?, And? May be differently assigned to determine factors that most affect user similarity with the first user among S1, S2, and S3.

5 shows a ranking algorithm according to one embodiment.

Referring to FIG. 5, a pseudo code of a ranking algorithm is shown. First, a word is extracted from a user's search query to determine a field corresponding to the search query, and a search result for the user query can be received. Thereafter, the words related to the user's tendency are matched with the fractional-cascading method shown in FIG. 5, and the search results for words having a large number of matches with the user's tendency can be increased to the upper rank.

Next, the words related to the similar user's tendency are matched with the fractional-cascading method of FIG. 5 to determine the ranking for the words related to the similar user's tendency, and reflected on the search result considering the user's tendency, can do.

According to an embodiment of the present invention, when a search result is provided in consideration of a tendency of a user having a tendency similar to a user, the search result reflecting the tendency of the user is provided higher than the search result reflecting the tendency of the user having the similar tendency can do.

6 is a flowchart illustrating an overall method of providing a search result according to an embodiment.

Collecting (610) the social activities of the first user (600), extracting words from the collected social activities, and determining the words of interest taking into account the document appearing and the frequency of word occurrences in one document. The interest field including at least one word of interest may be classified as a category, and at least one word of interest may be classified for each category. A time and frequency weight is assigned to at least one word of interest belonging to a specific category, and the first and second user's tendencies to a specific category are summed 620.

The expertise score of at least one user who performs social activities can be calculated, and similarly, the first user can determine (620) a field with a high professionalism score. Then, a network score for a first user of a user who has a personal relationship with the first user on the social media service may be calculated to determine 620 a user who is close to the first user.

It is possible to determine (630) a second user having a similar tendency among the users other than the first user and the first user. A database may be constructed (640) for the propensity of the first user and the propensity of similar users that have similar propensity to the first user.

When the first user 600 performs the query / search 650, the first user's query is analyzed (660) using the database that has been constructed, and the field corresponding to the query is determined by the user Information related to the user who is closely related to the first user, information related to the user who is closely related to the first user, It is possible to determine the order in which the results are provided through the ranking algorithm 670 and to output the search results accordingly (680).

The apparatus described above may be implemented as a hardware component, a software component, and / or a combination of hardware components and software components. For example, the apparatus and components described in the embodiments may be implemented within a computer system, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA) A programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to execution of the software. For ease of understanding, the processing apparatus may be described as being used singly, but those skilled in the art will recognize that the processing apparatus may have a plurality of processing elements and / As shown in FIG. For example, the processing unit may comprise a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of the foregoing, and may be configured to configure the processing device to operate as desired or to process it collectively or collectively Device can be commanded. The software and / or data may be in the form of any type of machine, component, physical device, virtual equipment, computer storage media, or device , Or may be permanently or temporarily embodied in a transmitted signal wave. The software may be distributed over a networked computer system and stored or executed in a distributed manner. The software and data may be stored on one or more computer readable recording media.

The method according to an embodiment 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, and the like, alone or in combination. The program instructions to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

Although the embodiments have been described with reference to the drawings, various modifications and variations may be made by those skilled in the art. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (20)

A search result providing apparatus for providing a search result that reflects a tendency of a user,
A detector for collecting a social activity of a first user and analyzing the social activity to detect at least one or more words;
A determination unit for calculating a frequency of the word and extracting at least one word of interest detected with a frequency greater than or equal to a predetermined threshold value;
A classification unit configured to classify the interest word by a related subject and construct an ontology;
By applying a time weight and a frequency weight to at least one or more of the interest words related to any one of the topics to calculate a tendency score of the first user for any one of the topics, A first operation unit for providing content related to a first topic higher than a predetermined threshold value among the at least one topic corresponding to a word included in the preference score as an upper search result; And
Calculating a difference between the propensity score of the first user and the propensity score of the user other than the first user to determine a second user having a similar tendency to the first user with a predetermined difference or less , Providing content related to a second topic higher than a predetermined threshold of the second user's at least one topic among at least one topic corresponding to a word included in the query retrieved by the first user as an upper search result The second calculating section
And the search result providing device. The method according to claim 1,
Wherein the first calculation unit assigns the time weight less as the current time point and the past time point interval in which the interest word is input to the social activity are longer. The method according to claim 1,
Wherein the first calculation unit gives the higher frequency weight value as the frequency of interest calculated by the determination unit is higher. delete The method according to claim 1,
Considering at least one user who performs the social activity on any one of the topics, considering participation of other users in the social activities of each user, such as recommendation, comment and sharing of the other user, And a third calculation unit for calculating the expertise score of each user. 6. The method of claim 5,
Wherein the third calculation unit determines a third user whose expertise score is equal to or greater than the threshold value as an expert for the third topic for a third subject whose expertise score of the first user is equal to or higher than a predetermined threshold,
Wherein when the third topic is included in at least one topic corresponding to a word included in a query retrieved by the first user, the at least one topic corresponding to a word included in the query retrieved by the first user Wherein the search result providing unit provides content related to a topic whose tendency score of the third user is higher than a predetermined threshold as an upper search result. The method according to claim 1,
Further comprising a fourth calculation unit for calculating a network score by weighting according to the social activity relationship between the first user and the user other than the first user. 8. The method of claim 7,
The fourth calculation unit determines a user other than the first user having the networking score equal to or higher than a predetermined threshold as a fourth user,
A search result providing a content related to a fourth topic having a tendency score of the fourth user higher than a predetermined threshold among at least one topic corresponding to a word included in a query retrieved by the first user as an upper search result . The method according to claim 1,
And provides contents related to the first topic as search results higher than contents related to the second topic. The method according to claim 6,
Wherein the content related to the first topic is provided as a search result higher in content than a content related to a topic higher than a predetermined threshold of the propensity score of the third user. A search result providing method for providing a search result that reflects a tendency of a user,
Collecting a social activity of a first user and analyzing the social activity to detect at least one or more words;
Calculating a frequency of the words and extracting at least one word of interest that is detected as a frequency greater than or equal to a predetermined threshold value;
Constructing an ontology by classifying the words of interest into related topics;
Calculating a tendency score of a first user for any one of the topics by assigning a time weight and a frequency weight to at least one of the interest words related to any one of the topics;
Providing content related to a first topic higher than a predetermined threshold value of at least one topic among at least one topic corresponding to a word included in a query retrieved by the first user as an upper search result;
Calculating a difference between the propensity score of the first user and the propensity score of the user other than the first user to determine a second user having a difference of less than a predetermined threshold value as a user having a similar propensity to the first user step; And
Providing a content related to a second topic higher than a predetermined threshold value of at least one topic among at least one topic corresponding to a word included in a query retrieved by the first user as an upper search result
And a search result providing step. 12. The method of claim 11,
Wherein the calculating step gives the time weight less as the current time and the longer time interval of the past when the interest word is entered into the social activity. 12. The method of claim 11,
Wherein the calculating step gives the higher frequency weight as the frequency of the extracted interest words is higher. delete 12. The method of claim 11,
Considering a plurality of users performing the social activity for any one of the topics, taking into consideration participation of other users in recommending, commenting, and sharing the social activities of the respective users with respect to each other, And calculating a user's professionalism score. 16. The method of claim 15,
Determining a third user whose expertise score is greater than or equal to the threshold value as an expert for the third topic for a third subject whose expertise score of the first user is equal to or greater than a predetermined threshold; And
When the third topic is included in at least one topic corresponding to a word included in a query retrieved by the first user, content related to a topic higher than a predetermined threshold value of the tendency score of the third user is referred to as an upper And providing the result as a search result. 12. The method of claim 11,
Further comprising the step of calculating a network score by weighting according to the social activity relationship between the first user and the user other than the first user. 18. The method of claim 17,
Determining a user other than the first user having the networking score equal to or greater than a predetermined threshold as a fourth user; And
Providing, as an upper search result, content related to a fourth topic, the tendency score of the fourth user among at least one topic corresponding to a word included in the query retrieved by the first user is higher than a predetermined threshold; Further comprising: 12. The method of claim 11,
Wherein the content related to the first topic is provided as a search result higher than the content related to the second topic. 17. The method of claim 16,
Wherein the content related to the first subject is provided as a search result higher in content than a content related to a topic whose tendency score of the third user is higher than a predetermined threshold.

Images