KR20220091948A - Apparatus and Method For Providing Recommendation Service Through Social Media Activity Analysis - Google Patents

Abstract

The present invention relates to an apparatus for providing a recommendation service through social media activity analysis, comprising: at least one memory; at least one database unit; and at least one processor. According to the present invention, it is possible to recommend following actions, places, products, etc. suitable for social trends.

Description

Apparatus and Method For Providing Recommendation Service Through Social Media Activity Analysis

The present invention relates to an apparatus and method for providing a recommendation service through social media activity analysis, and more particularly, to a service for recommending a place, action, moving course, and product for a keyword of interest of a user through social media activity analysis. it's about

Due to the recent development of artificial intelligence and data analysis technology, it has a great influence on the online sales market and the social media market. For example, a personalized recommendation service that takes into account the preference of individuals who consume a specific product or news is rapidly spreading.

These services form cooperation based on individual consumption patterns, gender, age group, etc., and recommend products or news similar to products or news consumed by users within a specific group.

However, since matching of similar groups is based on demographic analysis such as gender or age and is limited to simple areas of interest, it is difficult to provide services based on changes in users' interest patterns or interest paths over time. There are limits.

In order to solve the above problems, in the present invention, an effective user is extracted through social media activity content analysis for each keyword of interest, and action route information or movement route information is extracted and stored by analyzing the social media activity content of the effective user. Analyze social media activity to recommend the next place, action, moving course, and product based on the action route information or movement route information for the keyword of interest by executing an internal recommendation algorithm to the user who has entered a specific interest keyword An object of the present invention is to provide an apparatus and method for providing a recommendation service through

The above object is to provide an apparatus for providing a recommendation service through social media activity analysis, comprising: at least one memory; at least one database unit; and at least one processor; The at least one processor may include: (a) receiving and registering a first interest keyword; (b) crawling relevant social media content on a plurality of social media platforms using the first interest keyword; (c) analyzing the crawled social media content to extract user account information of at least one valid content highly related to the first interest keyword, and corresponding the extracted account information of the user to the first interest keyword storing the effective user account information in the database unit; (d) crawling the social media activity content of the effective user based on the effective user account information stored in the database unit; (e) extracting the crawled content as a standardized text through morpheme analysis, classifying the text by category through machine learning, assigning a category value, and storing the text in the database unit; (f) Behavioral route information and movement of the effective user according to time based on at least one of the content creation time information, the context of the content creation, time information included in the content, and the positional relationship of the text generating at least one of the route information and storing it in the database unit; and (g) when a first interest keyword is inputted from the user, the database unit extracts the action route information and the movement route information related to the first interest keyword, and executes a recommendation algorithm to provide the user with a location and action It is achieved by the apparatus for providing a recommendation service through analysis of social media activity, characterized in that the process of recommending at least one of , a moving course, and a product is processed.

Here, the category may include at least one of location/place, food/product, brand, culture/leisure, nature/environment, person, issue, and emotion.

In addition, the processor may further process a process of extracting location information indicating a location where the user created the content or an artificially designated location from the crawled content in the process (e).

In addition, in the process (f), the processor determines a time-series order based on the creation time information of the content, the relationship before and after the creation of the content, the time information included in the content, and the positional relationship of the text, At least one of the action path information and the movement path information may be generated by assigning an order value and arranging the text to which the category value is assigned according to the time-series order value.

Furthermore, in the process (f), the processor is configured to: information on the creation time of the content, the relationship before and after creation of the content, time information included in the content, location information included in the content, positional relationship of the text, and extracting the user's action or place based on the category value of the text, grouping the text in a time-series order according to the action or place, and assigning an order value to at least one of the action route information and the movement route information can create

In addition, the processor may register the location information included in the content and the text to which the category value is given as keywords.

In addition, in the process (e), the processor may further process a process of extracting a text indicating satisfaction from the standardized text extracted from the crawled content and assigning a weight to the corresponding text.

Here, the recommendation algorithm assigns a score according to the number of users or content having the action, place, and route included in the extracted action route information and the movement route information, and a plurality of action routes based on the given score It may include calculating a score corresponding to the information and the moving route information, and recommending at least one of a place, an action, a moving course, and a product to the user according to the calculated score.

Meanwhile, the above object is to provide a method for providing a recommendation service through analysis of social media activity, the method comprising: (a) receiving and registering a first interest keyword; (b) crawling relevant social media content on a plurality of social media platforms using the first interest keyword; (c) analyzing the crawled social media content to extract user account information of at least one valid content highly related to the first interest keyword, and corresponding the extracted account information of the user to the first interest keyword storing the effective user account information in the database unit; (d) crawling the social media activity content of the effective user based on the effective user account information stored in the database unit; (e) extracting the creation time information and location information of the content from the crawled content, extracting the standardized text through morpheme analysis, classifying the text by category through machine learning, and assigning a category value to the database unit saving to; (f) Behavioral route information and movement of the effective user according to time based on at least one of the content creation time information, the context of the content creation, time information included in the content, and the positional relationship of the text generating at least one of the route information and storing it in the database unit; and (g) when a first interest keyword is inputted from the user, the database unit extracts the action route information and the movement route information related to the first interest keyword, and executes a recommendation algorithm to provide the user with a location and action It can also be achieved by a method of providing a recommendation service through social media activity analysis, characterized in that it includes the step of recommending at least one of , a moving course, and a product.

The step (f) determines a time-series order based on the content creation time information, the relationship before and after creation of the content, time information included in the content, and the positional relationship of the text, and assigns an order value; , by arranging the text to which the category value is given according to the time-series order value, at least one of the action route information and the movement route information may be generated.

As described above, the apparatus and method for providing a recommendation service through analysis of social media activity according to the present invention executes an internal recommendation algorithm based on the relevant action route information or movement route information to a user who has input a specific interest keyword to keep up with social trends and trends. We can recommend the following places, actions, travel routes, and products that fit your interests.

1 is a schematic diagram illustrating an environment for providing a recommendation service through social media activity analysis according to an embodiment of the present invention.
FIG. 2 is a flowchart illustrating an example of a process of providing a recommendation service through analysis of social media activity of FIG. 1 .
3 is a diagram illustrating detailed configurations of a data analysis unit and a database unit in order to explain a detailed operation of the service server according to an embodiment of the present invention.
4 illustrates data before and after classification of advertisement text by the user filter unit in the process of data analysis in the service server according to an embodiment of the present invention.
5 illustrates data before and after classification by an inference module in the process of data analysis in the service server according to an embodiment of the present invention.
6 illustrates an example of social media content to explain a method for extracting action route information and/or movement route information by a route information extraction unit according to an embodiment of the present invention.
7 is a schematic diagram for explaining a recommendation algorithm according to an embodiment of the present invention.
8 and 9 are diagrams illustrating an example of recommendation information provided to a user terminal by a recommendation unit according to an embodiment of the present invention.
10 is a flowchart illustrating a method of generating action route information and/or movement route information in a service server according to an embodiment of the present invention.
11 is a flowchart illustrating a method of generating and providing recommended route information by a service server according to an embodiment of the present invention.

Hereinafter, specific embodiments of the present invention will be described with reference to the drawings.

1 is a schematic diagram illustrating an environment for providing a recommendation service through social media activity analysis according to an embodiment of the present invention. The user terminal 10 , the SNS device 20 , and the service server 100 are connected by a network 30 .

The user terminal 10 may include a mobile terminal such as a smart phone or a tablet PC, and a user terminal 10 such as a PC, and is connected to the service server 100 through the network 30 according to the present invention. service can be provided. The user terminal 10 includes a communication module for transmitting and receiving data through a network such as the Internet.

The SNS device 20 refers to a server that provides social media platform services, and may include devices that provide various SNS platform services, such as Instagram, Facebook, Naver or Daum blog, cafe, and the like. Social media refers to online tools and platforms used to share opinions, thoughts, experiences, and perspectives with each other. Includes networks, instant message boards, podcasts, wikis, user-generated content (UCC), and more. In this specification, SNS (Social Network Service) and social media will be used with the same meaning.

The service server 100 is a device for providing a recommendation service through analysis of social media activity according to an embodiment of the present invention. The memory 110 for storing data and computer program code, and the processor 120 for executing the computer program code ), a database unit 130 in which the database is built, and a network interface 140 for network connection with the outside.

The database unit 130 includes a dictionary DB 131 used to analyze SNS activity, and a relational database management system (RDBMS) 133 in which user information and SNS activity analysis information are stored. Here, the dictionary DB 131 and the RDBMS 133 may be provided in a separate memory device or a separate DB server device, respectively. The network interface 130 includes an Internet communication module and the like.

When a keyword of interest is registered, the service server 100 accesses at least one SNS device 20, collects and analyzes relevant social media content, and extracts user account information of valid content highly related to the registered interest keyword. It is stored as the information of the effective user, and the SNS activity content of the effective user is collected and analyzed using the information of the effective user to extract the user's action route information or movement route information and stored in the RDBMS 133 . Here, the keyword of interest is continuously accumulated, and the user's action route information or movement route information related to the keyword of interest may also be periodically updated. Through this process, the RDBMS 133 builds a database of action route information and movement route information of a plurality of users for a plurality of keywords of interest.

When a specific interest keyword is input from the user terminal 10, the service server 100 extracts action route information and movement route information of effective users related to the interest keyword from the RDBMS 133, and uses an internal recommendation algorithm to A recommendation of at least one of a place, an action, a moving course, and a product is provided to the user terminal 10 . The user can check the recommended information and check and decide the next action. The service server 100 may include a web server that stores component files of a website for receiving a service request from a user and providing recommended information.

In FIG. 1 , the data collection unit 121 , the data analysis unit 123 , and the recommendation unit 125 are implemented as a series of computer program codes that perform respective functions, are stored in the memory 110 and are stored in the processor 120 . is executed through In FIG. 1 , for convenience of explanation, it is expressed as a configuration included in the processor 120 .

The data collection unit 121 is for accessing the SNS device 20 to collect related social media content, and includes a crawler that crawls data.

The data analysis unit 123 analyzes the collected data to extract an effective user, and analyzes the content of the effective user collected by the data collection unit 121 to determine the user's action path information and/or movement path information for the keyword of interest. As for extracting , it may include a machine learning or deep learning inference engine for data extraction, classification and prediction.

When a keyword of interest is input and searched for by the user terminal 10 , the recommendation unit 125 executes an internal recommendation algorithm based on the action route information and/or movement route information of the effective users stored in the RDBMS 133 to execute the relevant keyword of interest. It is for generating at least one recommendation information of the user's next place, action, moving course, and product regarding It can include a GUI generator program to show through statistics and figures, various techniques, etc.

FIG. 2 is a flowchart illustrating an example of a process of providing a recommendation service through analysis of social media activity of FIG. 1 .

The user can use the recommendation service according to the present invention by accessing the website serviced by the service server 100 through the user terminal 10 (S10). The website of the service server 100 according to the present invention may provide a product sales service in addition to the recommendation service. For example, a service for recommending a product or a place according to the present invention may be provided to members who have subscribed to a website providing a product sales service.

When a keyword of interest is registered (S10), the service server 100 collects related social media content from a plurality of social media platforms by using the keyword of interest (S11). The registration of interest keywords may be performed collectively or as necessary by the server administrator, or search keywords input by members may be automatically registered as interest keywords.

The service server 100 analyzes the collected social media content and extracts the user's account information of the content highly related to the corresponding interest keyword as the effective user information (S12). The extracted effective user information is stored in the database unit 130 . By repeating this process, information on effective users corresponding to a plurality of keywords of interest is accumulated.

The service server 100 collects the social media content created by the corresponding user by using the user account information of the effective user (S13). In this case, the collection target may be limited to contents created during a certain period, in order to reflect the latest trends or trends.

The service server 100 analyzes the collected content, extracts action route information and/or movement route information of the corresponding user, and stores it in the database unit 130 (S14). While the conventional recommendation service simply extracts and analyzes only content highly related to the keyword of interest, the present invention collects even content before and after the content that is highly related to the keyword of interest based on the user's behavior regarding the keyword of interest. Route (pattern) information and/or even movement route information can be extracted. Processes S10 to S14, S11 to S14, or S13 to S14 described above may be performed periodically or repeatedly as needed.

When the user accesses the service server 100 using the user terminal 10 and inputs a first interest keyword, the service server 100 provides action path information and/or movement path for the first interest keyword input by the user. Information is extracted from the database unit 130 (S16), and an internal recommendation algorithm is executed to generate recommendation information for recommending at least one of a place, an action, a moving course, and a product to the corresponding user (S17). The service server 100 provides the generated recommendation information to the user terminal 10 (S18), and the user terminal 10 displays the recommendation information (S19). As described above, the recommendation information may be provided in various formats that the user can intuitively and easily understand. At this time, if the first interest keyword input by the user does not exist in the RDBMS 133, the recommendation results are provided by executing steps S10 to S18 in real time.

FIG. 3 shows detailed configurations of the data analysis unit 123 and the database unit 130 in order to explain a specific operation of the service server 100 according to an embodiment of the present invention.

The dictionary DB 131 includes a morpheme dictionary DB 131a, a similar word dictionary DB 131b, a filter dictionary DB 131c, and an inference dictionary DB 131d. The morpheme dictionary DB 131a is constructed with morpheme data of text as a DB, and the morpheme analysis module 123a utilizes it to perform morpheme analysis. The synonym dictionary DB 131b means a dictionary DB for similar words having the same meaning, and the filter dictionary DB 113c means a DB for advertisement morphemes. The reasoning dictionary DB 131d is for classifying the processed morphemes by category field, and the categories to which each morpheme belongs are matched. Here, the category fields may be defined differently according to service contents, and the number may be variously set according to the service. For example, the category defined in the inference dictionary DB 131d according to the present invention includes at least one of location/place, food/product, brand, culture/leisure, nature/environment, person, emotion/satisfaction, and issue can do.

The data analysis unit 123 includes a morpheme analysis module 123a, a synonym matching module 123b, a user filter unit 123c, an inference module 123d, and a path information extraction unit 123e. The modules included in the data analysis unit 123 divide software for data analysis by function, and may be provided as separate programs or integrated with each other. The data analysis unit may extract location information indicating a location where the user created the content or an artificially designated location from the crawled content and store it in the database unit 130 as location information of the corresponding content. The location information may be included in the content as text indicating a region or place, tags, latitude and longitude information, etc., and the data analysis unit may convert the location information included in the text or tag into latitude and longitude information and extract it. The data analysis unit may contain a program that converts a place name or area name into GPS information, that is, latitude and longitude information, or may be used as an external program. As such, when the content contains text or location information indicating a place or location, the data analysis unit converts it and stores it as location information of the corresponding content in the RDBMS, and this location information is used when the recommended content is displayed based on the map. can be utilized. In this case, the location information may be hierarchically configured according to the details of the location information included in the content, for example, in units of administrative districts such as city/gun/gu/dong/bunji.

The morpheme analysis module 123a is for extracting crawled content as a standardized text through morpheme analysis, and can be implemented as a pre-processing module that tokenizes the text in units of morphemes. The morpheme analysis module 123a separates text data constituting a sentence into morpheme units based on the morpheme dictionary DB 131a. The morpheme analysis module 123a may update the morpheme dictionary DB 131a for words that are not in the morpheme dictionary DB 131a.

The synonym matching module 123b is for processing different words having the same meaning into the same word, and processes morphemes having similar meanings based on the synonym dictionary DB 131b. For example, 'Hana C.C, Hana CC, Hana Golf Course' is matched with 'Hana Country Club'. Meanwhile, the synonym matching module 123b may include a machine learning model, and for morphemes that are not in the synonym dictionary DB 131b, similar words are classified through the machine learning model and registered in the synonym dictionary DB 131b.

The user filter unit 123c is for filtering effective users, and classifies advertisement morphemes into separate groups based on the filter dictionary DB 113c. The user filter unit 123c classifies the morpheme texts representing advertisement, spam, and illegality among the extracted morphemes into separate groups, and determines whether the content is valid by the effective user based on the frequency of the texts classified into the separate groups. By judging whether or not advertising content is valid, it is excluded from analysis. 4 illustrates data before and after classification of advertisement text by the user filter unit 123c in the process of data analysis in the service server 100 according to an embodiment of the present invention. Referring to FIG. 4 , it can be seen that sales, drives, test drives, estimates, used cars, dealers, etc. are classified into texts indicating advertisement, spam, and illegality, respectively.

The user filter unit 123c may extract user information of the content determined to be advertisement content, register it in the filter dictionary DB 113c, and exclude the corresponding user from the effective user from the content analysis thereafter. The user filter unit 123c and the filter dictionary DB 113c exclude content for commercial purposes, such as advertisement content, from valid content, thereby excluding from analysis, and exclude users of commercial content from valid users.

The reasoning module 123d is for classifying the morphemes of each word extracted from the content by category field, and classifies the morphemes by matching them for each corresponding category based on the inference dictionary DB 131d. The inference module 123d may include an inference model through machine learning, and morphemes not in the inference dictionary DB 131d are classified by category through machine learning. Here, the category field means a group name that represents the property of each morpheme word. The category, that is, the property of a word, can be set in various ways in advance according to a user's definition, and the number of the word can also be set variously. can be For example, if the attributes of a morpheme word are defined as location/place, food/product, brand, culture/leisure, nature/environment, person/family, issue, emotion/satisfaction, etc., the category is location/place, food/ It may include product, brand, culture/leisure, nature/environment, person/family, issue, emotion/satisfaction, and the like. 5 exemplifies data before and after classification by the inference module 123d in the process of data analysis in the service server 100 according to an embodiment of the present invention, (a) is a morpheme analysis module 123a ) and data extracted through the synonym matching module 123b, (b) illustrates the category matched by the inference module 123d.

The RDBMS 133 maintains a database about data such as account information of an effective user corresponding to a plurality of keywords of interest, content of an effective user, content creation time, location information of content, and text to which a category is classified and a category value is assigned. it means. Information extracted and analyzed by the data collection unit 121 and the data analysis unit 123 is stored in the RDBMS 133 . The text to which location information and category values included in the content stored in the RDMS are assigned may also be registered as interest keywords. For example, the social media content analysis result of the effective user and the effective user regarding the first interest keyword is stored in the RDBMS, and it is not simply used for the first interest keyword, but other keywords of interest (text) extracted in the analysis process. can also be used for

The route information extracting unit 123e is the effective user's time based on at least one of the content creation time information of the effective user in the RDMS, the contextual relationship of content creation, time information included in the content, and the before-and-after positional relationship of the text in the RDMS. At least one of action route information and movement route information is generated according to the For example, the route information extraction unit 123e determines a time-series order based on the creation time information for each content, the relationship before and after creation of the content, time information included in the content, and the before-and-after positional relationship between the extracted text, At least one of the action route information and the movement route information may be generated by assigning an order value between the content and the text, and arranging the text to which the category value is assigned according to a time-series order value. As another example, the route information extraction unit 123e may include content creation time information, content creation time relationship, time information included in content, location information included in content, front-back position relationship of text in content, and text At least one of action route information and movement route information may be generated by extracting the user's action or place based on the category value, grouping texts in a time-series order according to the extracted action or location, and assigning an order value.

6 illustrates an example of social media content to explain a method for extracting action route information and/or movement route information of the route information extraction unit 123e according to an embodiment of the present invention. For example, if the keyword of interest is 'Gangneung travel', it is assumed that a user with an ID of Gil-dong is extracted as an effective user, and FIG. 6 is scrolled data of the effective user. In FIG. 6, for convenience of explanation, different colors are shown for each category so that categories for each text can be distinguished. The green square indicates the place/location category, the red square means emotion/satisfaction, the yellow square means food, and the black square means text belonging to other categories.

There are two postings in FIG. 6 , and the context of time between the two can be determined by the order of writing. In addition, if the texts to which the category values indicating the place/location are assigned are arranged in the order of the text position, it becomes 'Gangneung-Gapyeong Service Area-Gangneung-Gyeongpodae Beach-Junjin Hongge Unlimited Refill-Delicious Coffee Shop'. Each place/location is assigned an ordinal value. In this case, Gangneung is excluded because it is higher-level location information than other texts. Finally, the moving route is extracted to 'Gapyeong Rest Area - Gyeongpo Beach - Jumunjin Hongge Unlimited Refill - Delicious Coffee Shop'. In this case, texts matching each place may be grouped. For example, it can be grouped into 4 groups as 'Gapyeong Rest Area - Broiler Chicken, Gyeongpodae Beach - Baram, Jumunjin - All-You-Can-Eat Red Crab Refill, and Delicious Coffee Shop-Coffee-View'. In this case, the grouping may be determined in consideration of the order of texts in a sentence, punctuation marks such as commas and periods, and distances between texts. In this case, the information included in the action route information and the movement route information may be more detailed by grouping texts having categories other than the place together, as well as the time-series order of the places. In addition, when there are words representing emotions or evaluations, weights may be given to texts matching (grouping), for example, corresponding places or actions. For example, in FIG. 6 , there are words representing emotion or satisfaction, such as 'it was tasty', 'it was cool', 'it was really tasty', and 'it was very good'. In this case, weights may be assigned to these words and the grouped 'Gyeongpodae Beach', 'Jumunjin Hongge unlimited refill', and 'delicious coffee shop', respectively. Through the above process, the action route information (moving route information) extracted by the route information extraction unit 123e for the keyword of interest 'Gangneung Travel' is 'Gapyeong Rest Area-Gyeongpodae Beach-Jumunjin Honggeun Refill-Delicious Coffee Shop'. is stored in the RDBMS (133). On the other hand, the texts of 'Gapyeong Rest Area, Gyeongpo Beach, Jumunjin Hongge All-You-Can-Eat Refill, Delicious Coffee Shop' included in the travel route information above are not used only for keywords of interest in 'Gangneung Travel'. It can be registered and used for other keywords of interest.

When a keyword of interest is input and searched for from the user terminal 10 , the recommendation unit 125 extracts a plurality of action route information and movement route information corresponding to the relevant keyword of interest stored in the RDBMS 133 , and executes an internal recommendation algorithm. to generate at least one recommendation information among a next place, an action, a moving course, and a product related to the corresponding interest keyword. As described above, since the RDBMS 133 stores effective user information for each keyword of interest and a plurality of action route information and movement route information corresponding thereto, the recommendation unit 125 utilizes the data stored in the RDBMS 133 . .

The recommendation algorithm gives, for example, a score according to the number and distance of users or contents, to the action path information of the effective user extracted from the RDBMS 133 and the action, place, and path included in the moving path information, Based on the obtained score, a score corresponding to the number of cases of a plurality of action route information and movement route information is calculated, and according to the calculated score, at least one of a place, an action, a moving course, and a product is recommended to the user.

7 is a schematic diagram for explaining a recommendation algorithm according to an embodiment of the present invention. Assume that an interest keyword of 'Gangneung travel' is input from the user terminal 10, and there are three types of relevant action route information and movement route information within the reference area during the reference period extracted by the recommendation unit 125 as follows.

ㆍA user, route information 'a -> b -> c -> d -> ㅁ -> f',

ㆍB user, route information 'ㄴ -> ㄹ -> ㅅ -> ㅁ -> ㅁ'

ㆍC user, route information 'a -> c -> b -> ㅇ'

Here, a, b, c, d, ㅁ, ㅅ, ㅅ, ㅇ, etc. indicate a place/position.

In this case, the starting location (reference location) may be determined in consideration of the current location of the user, a place weight according to satisfaction, a place with the most number of visits, a geographical location, and the like. In FIG. 7 , it is assumed that the reference position (the starting position) is determined to be 'C' in consideration of the current user's position.

In the recommendation algorithm, the value of each path starts from 0.9 according to the visit order of the users who visited the place, decreases by 0.1 as the distance increases, and the path from the moment it becomes 0 is not considered. In this case, the higher value of the two is taken as that value.

The value of each place (location) is determined according to the number of users who have visited the place and has a value of 1.1 to 2.0. For example, if all 10 users visited place 'a', the value of place 'a' would be 2.0, and if only 1 out of 10 users visited place 'a', the value of place a would be 1.1. When a score is given to each place (location) and route according to these criteria, a score is given as shown in FIG. 7 .

The recommendation algorithm calculates the recommended route score by multiplying the location value by the route value according to each location/course by reflecting the score given to the place and route to the recommended location/course, and the value of the route/place based on ‘c’ It is calculated as follows.

ㆍc→a, 0.9 * 1.1 = 0.99

ㆍc→b→a, 0.9 * 1.2 * 0.8 * 1.1 = 0.9504

ㆍc→b, 0.9 * 1.2 = 1.08

ㆍc→a→b, 0.9 * 1.1 * 0.8 * 1.2 = 0.9504

ㆍc → d, 0.9 * 1.1 = 0.99

ㆍc → d → ㅁ, 0.9 * 1.1 * 0.8 * 1.1 = 0.8712

ㆍC → ㄹ → ㅁ → f, 0.9 * 1.1 * 0.8 * 1.1 * 0.7 * 1.1 = 0.670824

ㆍc → b → o, 0.9 * 1.2 * 0.8 * 1.0 = 0.8640

ㆍC→a→b→ㅇ, 0.9 * 1.1 * 0.8 * 1.2 * 0.8 * 1.0 = 0.760320

Based on the above calculation result, the following recommendations are possible according to the depth (step) of the path.

ㆍLocation recommendation: b

ㆍRecommended route information (2depth): c→b→a

ㆍRecommended route information (3depth): c→a→b→ㅇ

8 and 9 show an example of recommendation information provided to the user terminal 10 by the recommendation unit 125 according to an embodiment of the present invention. Referring to (a) of FIG. 8 , when the user inputs 'Busan' as a keyword of interest, a window for inputting a keyword of interest (search keyword) is provided in area A, and a sub-interest of the input interest keyword in area B is provided. As a region in which recommended keywords are displayed, it can be seen that 4 'coffee, pork, bakery, beef' are displayed as sub-interested keywords in Busan in FIG. 8(a). Area C displays the keywords of interest on the map, and by displaying the locations where each keyword is mentioned in different colors for each keyword on the map, the distribution of the keywords of interest according to regions can be intuitively known. Area B includes an icon for selecting and excluding recommended sub-keywords, and when the user clicks on it, it is selected or canceled. In this case, the distribution of only the selected keyword is displayed in area C below.

8(b) is a case in which the user inputs 'coffee' into the keyword input window of area A. Since there is no sub-interested keyword of coffee, recommendation information is not displayed in area B, and the location related to 'coffee' in the country is displayed on the map in area C. By looking at this, it is possible to know at a glance which regions are most interested in the corresponding keywords.

FIG. 9 shows a screen when the user clicks on Gangneung on the map of area C of FIG. 8(b) after the user inputs 'coffee' into the keyword input window of area A. Referring to FIG. In this case, recommended route information related to coffee in Gangneung is displayed on the map in area C, and recommended route information is displayed in text in area D. Referring to FIG. 9, each place name is connected with an arrow to indicate a recommended course as a whole, and a selection button to view recommended post information corresponding to each place is provided. By linking to the content of the post, you can check the details of the user's post.

10 is a flowchart illustrating a method of generating action route information and/or movement route information in the service server 100 according to an embodiment of the present invention. Referring to FIG. 10 , relevant social media content is collected from a plurality of social media platforms using registered interest keywords and effective users are extracted ( S20 ). Specifically, the user's account information of at least one valid content highly related to the relevant interest keyword by crawling the relevant social media content on a plurality of social media platforms using the registered interest keyword, and analyzing the crawled social media content is extracted, and the extracted account information of the user is stored in the RDBMS 133 as effective user account information corresponding to the corresponding interest keyword. In this case, a plurality of valid user account information is registered in one interest keyword.

When the account information about the effective user is extracted, the account information of the effective user is used to access the SNS platform, the social media activity content of the corresponding user is crawled (S21), and the crawled content is extracted as a standardized text through morphological analysis. do (S22). At this time, in addition to text classification, location information and writing time information included in the content are extracted and stored in the RDBMS 133 . The service server 100 performs synonym matching on the standardized text extracted through the synonym matching module (S23). The service server 100 classifies the texts representing advertisements, spam, and illegality among the standardized texts into separate groups through the user filter unit 123c, and when the texts corresponding to the advertisements have a certain frequency or more, the corresponding users are selected as valid users. and excluded from valid content (S25, s26). The service server 100 classifies the texts matched with similar words by predefined categories through the inference module 123d, assigns a category value to the classified text, and stores it in the RDBMS 133 (S27). Here, similar word matching and category classification can be performed through a machine learning model.

11 is a flowchart illustrating a method of generating and providing recommended route information by the service server 100 according to an embodiment of the present invention. Referring to FIG. 11 , when a keyword of interest is input from the user terminal 10 and received by the service server 100 ( S30 ), the service server 100 provides action path information corresponding to the keyword of interest stored in the RDBMS 133 . and extracts movement path information (S31). Then, by executing the recommendation algorithm, a score is given to each place, action, and path between each place or action included in the route information, the number of cases of various recommended candidate routes is extracted, A score is calculated for each recommended candidate path by multiplying the score by the path score (S32). Then, a place, an action, a moving course, and a product having the highest score are selected as the final recommended route, and the result is provided to the user terminal 10 and displayed on the map (S33).

In the above, even though it has been described that all components constituting the embodiment of the present invention operate by being combined or combined into one, the present invention is not necessarily limited to this embodiment. That is, within the scope of the object of the present invention, all the components may operate by selectively combining one or more. In addition, although all the components may be implemented as one independent hardware, some or all of the components are selectively combined to perform some or all of the functions of the combined hardware in one or a plurality of hardware program modules It may be implemented as a computer program having Codes and code segments constituting the computer program can be easily inferred by those skilled in the art of the present invention. Such a computer program is stored in a computer readable storage medium (Computer Readable Media), read and executed by the computer, thereby implementing the embodiment of the present invention. The storage medium of the computer program may include a magnetic recording medium, an optical recording medium, and the like.

In addition, terms such as "comprises", "comprises" or "have" described above mean that the corresponding component may be inherent, unless otherwise specified, excluding other components. Rather, it should be construed as being able to further include other components. All terms, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs, unless otherwise defined. Terms commonly used, such as those defined in the dictionary, should be interpreted as being consistent with the contextual meaning of the related art, and are not interpreted in an ideal or excessively formal meaning unless explicitly defined in the present invention.

The above description is merely illustrative of the technical spirit of the present invention, and various modifications and variations will be possible without departing from the essential characteristics of the present invention by those skilled in the art to which the present invention pertains. Accordingly, the embodiments disclosed in the present invention are not intended to limit the technical spirit of the present invention, but to explain, and the scope of the technical spirit of the present invention is not limited by these embodiments. The protection scope of the present invention should be construed by the following claims, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of the present invention.

10: user terminal 20: SNS device
100: service server 110: memory
120: processor 121: data collection unit
123: data analysis unit 125: recommendation unit
130: database unit 131: dictionary DB
133: RDBMS 140: network interface unit

Claims (10)

In the apparatus for providing a recommendation service through social media activity analysis,
at least one memory;
at least one database unit; and
at least one processor;
the at least one processor,
(a) a process of receiving and registering a first interest keyword;
(b) crawling relevant social media content on a plurality of social media platforms by using the first interest keyword;
(c) analyzing the crawled social media content to extract user account information of at least one valid content highly related to the first interest keyword, and corresponding the extracted account information of the user to the first interest keyword storing the effective user account information in the database unit;
(d) crawling the social media activity content of the effective user based on the effective user account information stored in the database unit;
(e) extracting the crawled content as a standardized text through morphological analysis, classifying the text by category through machine learning, assigning a category value, and storing the text in the database unit;
(f) Behavioral route information and movement of the effective user according to time based on at least one of creation time information of the content, the relationship before and after creation of the content, time information included in the content, and the positional relationship of the text generating at least one of the route information and storing it in the database unit; and
(g) when a first interest keyword is input from the user, the database unit extracts the action route information and the movement route information related to the first interest keyword, and executes a recommendation algorithm to provide the user with a location, action, A device for providing a recommendation service through analysis of social media activity, characterized in that the process of recommending at least one of a moving course and a product is processed.
According to claim 1,
The category includes at least one of location/place, food/product, brand, culture/leisure, nature/environment, person/family, issue, and emotion/satisfaction. .
3. The method of claim 2,
The processor in the process (e),
Recommendation service providing apparatus through social media activity analysis, characterized in that the process of extracting location information indicating the location where the user created the content or the location artificially designated from the crawled content is further processed.
4. The method of claim 3,
The processor in the process (f),
Based on the creation time information of the content, the relationship before and after creation of the content, the time information included in the content, and the positional relationship of the text, a chronological order is determined and an order value is assigned, and the category value is assigned The apparatus for providing a recommendation service through analysis of social media activity, characterized in that the text is arranged according to the time-series order value to generate at least one of the action route information and the movement route information.
4. The method of claim 3,
The processor in the process (f),
User action or Through social media activity analysis, characterized in that extracting a place, grouping the text in a time-series order according to the action or place, and assigning an order value to generate at least one of the action route information and the movement route information Recommendation service provider.
4. The method of claim 3,
The processor is
The apparatus for providing a recommendation service through social media activity analysis, characterized in that the text to which the location information and the category value included in the content are assigned as keywords is registered.
According to claim 1,
The processor, in the process (e),
The apparatus for providing a recommendation service through analysis of social media activity, characterized in that the process of extracting a text indicating satisfaction from the standardized text extracted from the crawled content and assigning a weight to the corresponding text is further processed.
7. The method of claim 6,
The recommendation algorithm gives a score according to the number of users or contents having the action, place, and route included in the extracted action route information and the movement route information, and a plurality of action route information and Recommendation service through social media activity analysis, comprising calculating a score corresponding to the moving route information and recommending at least one of a place, an action, a moving course, and a product to the user according to the calculated score Device.
In the method of providing a recommendation service through social media activity analysis,
(a) receiving and registering a first interest keyword;
(b) crawling relevant social media content on a plurality of social media platforms using the first interest keyword;
(c) analyzing the crawled social media content to extract user account information of at least one valid content highly related to the first interest keyword, and corresponding the extracted account information of the user to the first interest keyword storing the effective user account information in the database unit;
(d) crawling the social media activity content of the effective user based on the effective user account information stored in the database unit;
(e) extracting the creation time information and location information of the content from the crawled content, extracting the standardized text through morphological analysis, classifying the text by category through machine learning, and assigning a category value to the database unit saving to;
(f) Behavioral route information and movement of the effective user according to time based on at least one of the content creation time information, the context of the content creation, time information included in the content, and the positional relationship of the text generating at least one of the route information and storing it in the database unit; and
(g) when a first interest keyword is input from the user, extracts the action route information and the movement route information related to the first interest keyword from the database unit, and executes a recommendation algorithm to provide the user with location, action, A method of providing a recommendation service through analysis of social media activity, comprising the step of recommending at least one of a moving course and a product.
10. The method of claim 9,
The step (f) determines a time-series order based on the content creation time information, the relationship before and after creation of the content, time information included in the content, and the positional relationship of the text, and assigns an order value; , wherein at least one of the action route information and the movement route information is generated by arranging the texts to which the category values are assigned according to the time-series sequence value.

Images