Service providing apparatus that supports conducting surveys by selecting survey subjects based on user characteristics and the operating method thereof

The device and method enhance survey quality by selecting participants based on region and subject field relevance, ensuring accurate and reliable data collection.

KR103045461B1Active Publication Date: 2026-11-10TJHE HANCOMM INC
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
KR1020240181967
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Filing Date
2024-12-09
Publication Date
2026-11-10
Estimated Expiration
2044-12-09

AI Technical Summary

Technical Problem

Existing survey methods that randomly select individuals often result in unreliable survey results due to participants unrelated to the survey topic, compromising the quality of collected data.

Method used

A service providing device and method that utilize databases to store user device identification, address, and interest keywords, enabling targeted selection of survey subjects based on region and subject field relevance through similarity rankings, and sending push messages to eligible users.

Benefits of technology

Enables surveyors to obtain high-quality survey results by selecting participants aligned with the survey's focus, enhancing data accuracy and reliability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 112024136395973-PAT00001_ABST
    Figure 112024136395973-PAT00001_ABST
Patent Text Reader

Abstract

The present invention provides a service providing device and a method of operation that support conducting a survey by selecting survey subjects according to user characteristics, thereby enabling surveyors to more easily obtain high-quality survey results.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present invention relates to a service providing device and a method of operation that supports conducting a survey by selecting survey subjects according to user characteristics. Background Technology

[0002] Recently, it has become common for companies and organizations to conduct surveys targeting people. For instance, when companies need to consider consumer needs when making decisions, they frequently conduct surveys to collect the data necessary for those decisions.

[0003] As surveys are utilized as a means to collect people's opinions on specific topics and obtain information necessary for decision-making, it is of paramount importance for surveyors to ensure the accuracy and reliability of the results.

[0004] However, existing survey methods utilized a approach where surveys were conducted on a large number of randomly selected individuals. This method had a problem in that it could compromise the reliability of the survey results, as people completely unrelated to the topic of the survey sometimes responded.

[0005] If a surveyor is able to conduct a survey targeting people associated with the survey, the surveyor will be able to collect higher quality responses from the participants and obtain more meaningful survey results.

[0006] Therefore, research is needed on technology that supports surveyors in conducting surveys by selecting relevant individuals as subjects. The problem to be solved

[0007] The present invention aims to support surveyors in more easily obtaining high-quality survey results by presenting a service providing device and a method of operation that support conducting a survey by selecting survey subjects according to user characteristics. means of solving the problem

[0008] A service providing device according to an embodiment of the present invention comprises: a device information database in which device identification information of an electronic terminal possessed by each of a plurality of users is stored; an address database in which address information corresponding to each of the plurality of users is stored; an interest keyword database in which an interest keyword set corresponding to each of the plurality of users is stored—wherein the interest keyword set corresponding to each of the plurality of users is a set composed of two or more words designated by each user among a predetermined set of words; a word database in which a word set corresponding to each of the n types of subject fields (wherein is a natural number greater than or equal to 2) is stored—wherein the word set corresponding to each of the n types of subject fields is a set composed of two or more words predetermined as corresponding to each subject field among the plurality of words; and, when a filtering request command for selecting survey subjects is received from an investigator terminal possessed by a surveyor, a first guidance message is transmitted to the investigator terminal instructing it to reply by specifying a region name for selecting survey subjects, and at the same time, a second guidance message is transmitted instructing it to reply by specifying a ranking of relevance to the survey for each of the n types of subject fields. A transmission guidance message transmission unit, and a ranking calculation unit that calculates the similarity rankings for the n types of subject fields corresponding to each of the plurality of users by performing a process of calculating the similarity between a set of words corresponding to each of the n types of subject fields and a set of interest keywords corresponding to each user for each of the plurality of users, and then assigning rankings in order of high similarity for the n types of subject fields, when at least one first place name is replied from the investigator terminal in response to the first guidance message and a correlation ranking for the n types of subject fields designated by the surveyor is replied in response to the second guidance message.The apparatus includes: an extraction unit that extracts a user among the plurality of users having a similarity rank in which the degree of matching with a specified correlation rank for the n types of subject fields exceeds a preset threshold; a selection unit that, if first users are extracted as a result of the extraction unit and have a similarity rank exceeding the threshold among the plurality of users, checks address information corresponding to each of the first users by referring to the address database, and then selects a user among the first users who has address information including at least one first region name; and a push message transmission unit that, if second users are selected as a result of the selection unit and have address information including at least one first region name among the first users, checks device identification information for each of the second users' electronic terminals by referring to the device information database, identifies each of the second users' electronic terminals based on the confirmed device identification information, and then transmits a push message containing a request to participate in a survey to each of the second users' electronic terminals.

[0009] In addition, a method of operation of a service providing device according to an embodiment of the present invention comprises the steps of: maintaining a device information database in which device identification information of an electronic terminal possessed by each of a plurality of users is stored; maintaining an address database in which address information corresponding to each of the plurality of users is stored; maintaining a database of interest keywords in which a set of interest keywords corresponding to each of the plurality of users is stored—the set of interest keywords corresponding to each of the plurality of users is a set composed of two or more words designated by each user among a plurality of pre-specified words; maintaining a word database in which a set of words corresponding to each of the pre-classified n types of subject fields (n is a natural number greater than or equal to 2) is stored—the set of words corresponding to each of the n types of subject fields is a set composed of two or more words pre-specified as corresponding to each subject field among the plurality of words; and when a filtering request command for selecting survey subjects is received from an investigator terminal possessed by a surveyor, a first guidance message is transmitted to the investigator terminal instructing it to reply by specifying a region name for selecting survey subjects, and at the same time, for each of the n types of subject fields, a ranking of the degree of association with the survey A step of transmitting a second guidance message instructing to specify and reply, wherein, from the investigator terminal, at least one first place name is replied in response to the first guidance message, and a correlation ranking for the n types of subject fields designated by the surveyor is replied in response to the second guidance message, and for each of the plurality of users, a process of calculating the similarity between a set of words corresponding to each of the n types of subject fields and a set of interest keywords corresponding to each user, and then assigning rankings for the n types of subject fields in order of high similarity,The method comprises the steps of: calculating a similarity ranking for n types of subject fields corresponding to each of the plurality of users; extracting, among the plurality of users, a user having a similarity ranking in which the degree of alignment with a designated association ranking for the n types of subject fields exceeds a preset threshold; if, as a result of the extraction step, first users are extracted among the plurality of users as users having a similarity ranking exceeding the threshold, confirming address information corresponding to each of the first users by referring to the address database, and then selecting a user among the first users who has address information containing at least one first region name; and, if, as a result of the selection step, second users are selected among the first users as users having address information containing at least one first region name, confirming device identification information for each of the second users' electronic terminals by referring to the device information database, identifying each of the second users' electronic terminals based on the confirmed device identification information, and then sending a push message containing a request to participate in a survey to each of the second users' electronic terminals. Effects of the invention

[0010] The present invention provides a service providing device and a method of operation that support conducting a survey by selecting survey subjects according to user characteristics, thereby enabling surveyors to more easily obtain high-quality survey results. Brief explanation of the drawing

[0011] FIG. 1 is a drawing illustrating the structure of a service providing device according to an embodiment of the present invention. FIG. 2 is a flowchart illustrating the operation method of a service providing device according to an embodiment of the present invention. Specific details for implementing the invention

[0012] Embodiments according to the present invention will be described in detail below with reference to the accompanying drawings. This description is not intended to limit the present invention to specific embodiments and should be understood to include all modifications, equivalents, and substitutions that fall within the spirit and scope of the present invention. Similar reference numerals have been used for similar components in describing each drawing, and unless otherwise defined, all terms used in this specification, including technical or scientific terms, have the same meaning as generally understood by a person skilled in the art to which the present invention pertains.

[0013] In this document, when a part is described as "including" a component, it means that, unless specifically stated otherwise, it does not exclude other components but may include additional components. Furthermore, in various embodiments of the present invention, each component, functional block, or means may be composed of one or more sub-components, and the electrical, electronic, or mechanical functions performed by each component may be implemented by various known devices or mechanical elements, such as electronic circuits, integrated circuits, and ASICs (Application Specific Integrated Circuits), and may be implemented separately or two or more may be integrated into one.

[0014] Meanwhile, the blocks in the attached block diagram or the steps in the flowchart may be interpreted as computer program instructions that perform designated functions by being loaded into the processor or memory of data-processing equipment, such as general-purpose computers, specialized computers, portable notebook computers, and network computers. Since these computer program instructions may be stored in memory provided in a computer device or in memory readable by a computer, the functions described in the blocks in the block diagram or the steps in the flowchart may be produced as manufactured products containing means of instruction to perform them. Furthermore, each block or each step may represent a module, segment, or part of code containing one or more executable instructions for executing a specific logical function(s). Also, it should be noted that in some alternative embodiments, the functions mentioned in the blocks or steps may be executed in a different order than the prescribed order. For example, two blocks or steps shown in succession may be performed substantially simultaneously or in reverse order, and in some cases, some blocks or steps may be omitted.

[0015] FIG. 1 is a drawing illustrating the structure of a service providing device according to an embodiment of the present invention.

[0016] Referring to FIG. 1, a service providing device (110) according to an embodiment of the present invention includes a device information database (111), an address database (112), an interest keyword database (113), a word database (114), a guidance message transmission unit (115), a ranking calculation unit (116), an extraction unit (117), a selection unit (118), and a push message transmission unit (119).

[0017] The device information database (111) stores device identification information of electronic terminals owned by each of the multiple users.

[0018] Here, the device identification information of an electronic terminal refers to unique device information for identifying each electronic terminal, and may include mobile phone numbers, MAC (Media Access Control) addresses, IMEI (International Mobile Equipment Identity) information, etc.

[0019] For example, the device information database (111) may have information stored as shown in Table 1 below.

[0021] Multiple users Device identification information of the electronic terminal owned by each user User 1 Device identification information 1 User 2 Device identification information 2 User 3 Device identification information 3 ... ...

[0023] The address database (112) stores address information corresponding to each of the aforementioned multiple users.

[0024] Here, the address information corresponding to each of the aforementioned multiple users may refer to the residential address of each user, or it may refer to the address of a specific place directly designated by each user.

[0025] For example, the address database (112) may have information stored as shown in Table 2 below.

[0027] Multiple users Address Information User 1 Yeouido-dong, Yeongdeungpo-gu, Seoul User 2 Goseok-dong, Gwonseon-gu, Suwon-si, Gyeonggi-do User 3 Jung-dong, Haeundae-gu, Busan Metropolitan City ... ...

[0029] The interest keyword database (113) stores a set of interest keywords corresponding to each of the plurality of users.

[0030] Here, the set of interest keywords corresponding to each of the aforementioned multiple users refers to a set composed of two or more words selected by each user from among a pre-specified set of multiple words. For example, the set of interest keywords corresponding to a specific user may be a set composed in the form of '{exercise, financial management, reading, ...}'.

[0031] In this regard, the interest keyword database (113) may have information stored as shown in Table 3 below.

[0033] Multiple users Set of interest keywords User 1 {Word 1, Word 2, Word 3, ...} User 2 {Word A, Word B, Word C, ...} User 3 {Word A, Word B, Word C, ...} ... ...

[0035] The word database (114) stores sets of words corresponding to each of the n (n is a natural number greater than or equal to 2) types of subject fields that are pre-classified.

[0036] Here, the aforementioned n types of subject fields refer to fields classified according to subject, such as 'law, literature, engineering, society, science, politics,' etc. Furthermore, the word set corresponding to each of the aforementioned n types of subject fields refers to a set composed of two or more words pre-designated as corresponding to each subject field among the aforementioned multiple words. For example, if 'literature' exists as one of the aforementioned n types of subject fields, the word set corresponding to the subject field 'literature' may be a set composed of two or more words corresponding to the subject field 'literature,' such as '{novel, poetry, e-book, ...}'.

[0037] In this regard, the word database (114) may have information stored as shown in Table 4 below.

[0039] Subject Areas word set Subject Area 1 {Word X, Word Y, Word Z, ...} Subject Area 2 {word a, word b, word c, ...} Subject Area 3 {Word G, Word N, Word D, ...} ... ...

[0041] When the guidance message transmission unit (115) receives a filtering request command for selecting survey subjects from the surveyor terminal (140) held by the surveyor (130), it transmits a first guidance message to the surveyor terminal (140) instructing it to reply by specifying a region name for selecting survey subjects, and at the same time transmits a second guidance message instructing it to reply by specifying a ranking of the degree of association with the survey for each of the n types of subject fields.

[0042] Then, the surveyor (130) can input a region name for selecting survey subjects and a correlation ranking for each of the n types of subject fields on the surveyor terminal (140) by looking at the first guidance message and the second guidance message received on the surveyor terminal (140). At this time, the surveyor terminal (140) is the service providing device (110) of the present invention, and can reply with the region name designated by the surveyor (130) in response to the first guidance message, and can reply with the correlation ranking for each of the n types of subject fields designated by the surveyor (130) in response to the second guidance message.

[0043] In this regard, as one embodiment, if a surveyor (130) inputs a regional name for selecting survey subjects on a surveyor terminal (140) such as ‘Seoul Metropolitan City, Busan Metropolitan City’ and inputs a correlation ranking for each of the n types of subject fields such as ‘Subject Field 1, Subject Field 2, Subject Field 3, ...’ as ‘1st, 3rd, 5th, ...’, the surveyor terminal (140) can be a service providing device (110) of the present invention and can reply with the regional name ‘Seoul Metropolitan City, Busan Metropolitan City’ in response to the first guidance message, and can reply with the correlation ranking for each of the n types of subject fields such as ‘1st, 3rd, 5th, ...’ in response to the second guidance message.

[0044] Thus, when at least one first regional name is replied from the investigator terminal (140) to the service providing device (110) of the present invention in response to the first guidance message, and when the correlation ranking for the n types of subject fields designated by the surveyor (130) is replied in response to the second guidance message, the ranking calculation unit (116) calculates the similarity ranking for the n types of subject fields corresponding to each of the plurality of users by performing a process of calculating the similarity between the word set corresponding to each of the n types of subject fields and the set of interest keywords corresponding to each user, and then assigning rankings to the n types of subject fields in order of high similarity.

[0045] In this case, according to one embodiment of the present invention, the ranking calculation unit (116) may include a sequence number assignment unit (120), a queue processing unit (121), and a calculation processing unit (122).

[0046] The sequence assignment unit (120) assigns a random sequence number to each of the plurality of users.

[0047] When a sequence number is assigned to each of the plurality of users, the queue processing unit (121) stores the sequence numbers of the plurality of users in ascending order in a queue.

[0048] The calculation processing unit (122) calculates the similarity ranking for the n types of subject fields corresponding to each of the plurality of users by repeating the following steps until no sequence number remains in the queue: a first step of extracting the fastest sequence number from the queue when the sequence number of the plurality of users is stored in the queue; a second step of designating the user according to the sequence number extracted from the queue as a reference user and calculating the similarity between the word set corresponding to each of the n types of subject fields and the set of interest keywords corresponding to the reference user; a third step of assigning a rank in order of high similarity for the n types of subject fields; and a fourth step of removing the sequence number extracted from the queue through the first step from the queue.

[0049] In this case, according to one embodiment of the present invention, the ranking unit (116) may further include a vector database (123).

[0050] The vector database (123) stores word vectors corresponding to each of the plurality of words.

[0051] Here, the word vector corresponding to each of the aforementioned multiple words is a pre-embedded vector based on the semantic similarity between words, and is specified so that a higher vector similarity is calculated as the semantic similarity between words increases.

[0052] For example, if 'word a' and 'word b' have similar meanings, word vectors corresponding to 'word a' and 'word b' may be assigned respectively so that a high vector similarity is calculated between the two words. These word vectors may be assigned to each word according to the Word2Vec embedding technique.

[0053] At this time, according to one embodiment of the present invention, when the calculation processing unit (122) calculates the similarity between the word set corresponding to each of the n types of subject fields and the interest keyword set corresponding to the reference user through the second process, if a situation arises where the similarity between the word set corresponding to the first subject field, which is one of the n types of subject fields, and the interest keyword set corresponding to the reference user needs to be calculated, the unit may refer to the vector database (123) to check the word vectors for the first words included in the word set corresponding to the first subject field and the second words included in the interest keyword set corresponding to the reference user, then calculate all the vector similarities between each of the first words and each of the second words, and when the calculated vector similarities are sorted in ascending order to calculate the quartiles, the overall average of the remaining vector similarities, excluding the vector similarities corresponding to the first quartile or lower, can be calculated as the similarity between the word set corresponding to the first subject field and the interest keyword set corresponding to the reference user.

[0054] Below, the operation of the ranking unit (116) will be explained in detail with examples.

[0055] First, let's assume that the above multiple users are a total of 10 users, such as 'User 1, User 2, User 3, ..., User 10'.

[0056] Then, the sequence assignment unit (120) can assign a random sequence number to each of the plurality of users as shown in Table 5 below.

[0058] Multiple users User 1 User 2 User 3 ... User 10 turn 1 2 3 ... 10

[0060] When a sequence number is assigned to each of the above multiple users in this way, the queue processing unit (121) can store the sequence numbers of the above multiple users in ascending order in the queue. In this regard, the queue processing unit (121) can store the sequence numbers in the queue in the form of Table 6 below.

[0062] 1 2 3 ... 10

[0064] When the sequence numbers of the plurality of users, 'User 1, User 2, User 3, ..., User 10', are stored in the queue in this way, the calculation processing unit (122) can extract the earliest sequence number, '1', from the queue.

[0065] Then, the calculation processing unit (122) can designate 'User 1', a user with a sequence number '1' extracted from the queue, as the reference user, and then calculate the similarity between the set of words corresponding to each of the n types of subject fields, 'subject field 1, subject field 2, subject field 3, ...' and the set of interest keywords corresponding to the reference user, 'User 1'.

[0066] That is, the calculation processing unit (122) can calculate the similarity between the word set corresponding to each of the n types of subject fields and the interest keyword set corresponding to 'User 1' by calculating the similarity between the word set corresponding to 'subject field 1' and the interest keyword set corresponding to 'User 1', the similarity between the word set corresponding to 'subject field 2' and the interest keyword set corresponding to 'User 1', and the similarity between the word set corresponding to 'subject field 3' and the interest keyword set corresponding to 'User 1'. At this time, assuming that the calculation processing unit (122) has a situation where it needs to calculate the similarity between the word set corresponding to 'subject field 1' and the interest keyword set corresponding to 'User 1', the details of how the calculation processing unit (122) calculates the similarity between the word set corresponding to 'subject field 1' and the interest keyword set corresponding to 'User 1' are explained in detail as follows.

[0067] First, the calculation processing unit (122) can check word vectors for the first words, 'word X, word Y, word Z, ...' included in the word set '{word X, word Y, word Z, ...}' corresponding to 'subject field 1', and the second words, 'word 1, word 2, word 3, ...' included in the interest keyword set '{word 1, word 2, word 3, ...}' corresponding to 'user 1' by referring to the vector database (123).

[0068] Then, the calculation processing unit (122) can calculate all vector similarities between each of the first words and each of the second words, such as the vector similarity between 'word X' and 'word 1', the vector similarity between 'word X' and 'word 2', the vector similarity between 'word X' and 'word 3', the vector similarity between 'word Y' and 'word 1', the vector similarity between 'word Y' and 'word 2', the vector similarity between 'word Y' and 'word 3', the vector similarity between 'word Z' and 'word 1', the vector similarity between 'word Z' and 'word 2', the vector similarity between 'word Z' and 'word 3', ...

[0069] After that, the calculation processing unit (122) can calculate the quartiles by sorting the calculated vector similarities in ascending order.

[0070] Here, quartiles refer to data sorted in ascending order divided into four equal parts, where the first quartile is the variable value corresponding to the 1 / 4 place and the third quartile is the variable value corresponding to the 3 / 4 place.

[0071] At this time, if the vector similarities are referred to as 'V1, V2, V3, V4, V5, V6, V7, V8, V9, ...' and the vector similarities corresponding to the first quartile or lower are referred to as 'V4, V7', the calculation processing unit (122) can calculate the overall average of the remaining vector similarities, 'V1, V2, V3, V5, V6, V8, V9, ...', excluding the vector similarities corresponding to the first quartile or lower among the vector similarities, as the similarity between the word set corresponding to 'subject field 1' and the interest keyword set corresponding to 'user 1'.

[0072] In this way, the calculation processing unit (122) can calculate the similarity between the set of words corresponding to each of the n types of subject fields and the set of interest keywords corresponding to the reference user, 'User 1'.

[0073] When the similarity calculation between the set of words corresponding to each of the n types of subject fields and the set of interest keywords corresponding to the reference user 'User 1' is completed, the calculation processing unit (122) can assign ranks to the n types of subject fields in order of high similarity. In this regard, if the n types of subject fields, 'subject field 1, subject field 2, subject field 3, ...', are arranged in order of high similarity, that is, from the largest similarity value to the smallest similarity value, it is said to be 'subject field 1 → subject field 3 → subject field 2, ...', the calculation processing unit (122) can assign ranks to 'subject field 1, subject field 2, subject field 3, ...' as '1st place, 3rd place, 2nd place, ...'.

[0074] After that, the calculation processing unit (122) can remove the sequence number '1' extracted from the queue from the queue.

[0075] In this way, the above queue can be changed into the form shown in Table 7 below.

[0077] 2 3 ... 10

[0079] After that, the calculation processing unit (122) can extract the fastest sequence number '2' from the above queue.

[0080] Then, the calculation processing unit (122) can designate 'User 2', a user with a sequence number '2' extracted from the queue, as the reference user, calculate the similarity between the set of words corresponding to each of the n types of subject fields and the set of interest keywords corresponding to the reference user 'User 2', and then assign ranks to the n types of subject fields in order of high similarity.

[0081] Then, the calculation processing unit (122) can remove the sequence number '2' extracted from the queue from the queue.

[0082] In this way, the above queue can be changed into the form shown in Table 8 below.

[0084] 3 ... 10

[0086] According to the method described so far, the calculation processing unit (122) can calculate the similarity ranking for the n types of subject fields corresponding to each of the plurality of users by repeating the first process, the second process, the third process, and the fourth process until no sequence number remains in the queue.

[0087] When the similarity ranking for the n types of subject fields corresponding to each of the plurality of users is calculated through the ranking calculation unit (116), the extraction unit (117) extracts among the plurality of users a user whose similarity ranking has a matching degree with the specified association ranking for the n types of subject fields that exceeds a preset threshold.

[0088] In this case, according to one embodiment of the present invention, the extraction unit (117) may include a feature vector generation unit (124), an association vector generation unit (125), and an extraction processing unit (126).

[0089] The feature vector generation unit (124) generates a feature vector corresponding to each of the plurality of users by performing a process of constructing an n-dimensional vector having a similarity ranking for each of the n types of subject fields corresponding to each user as a component for each of the plurality of users.

[0090] In this regard, as in the example described above, if the similarity ranking for the n types of subject fields corresponding to 'User 1' is calculated as '1st, 3rd, 2nd, ...' through the ranking unit (116), the feature vector generation unit (124) can generate '[1 3 2 ...]' as a feature vector corresponding to 'User 1' by configuring an n-dimensional vector having the corresponding similarity rankings '1, 3, 2, ...' as components as '[1 3 2 ...]'.

[0091] In this way, the feature vector generation unit (124) can generate a feature vector corresponding to each of the plurality of users as shown in Table 9 below.

[0093] Multiple users Feature vector User 1 Feature Vector 1 User 2 Feature Vector 2 User 3 Feature Vector 3 ... ...

[0095] The association vector generation unit (125) generates an n-dimensional association vector having an association rank specified for the n types of subject fields as a component.

[0096] In this regard, as in the example described above, the n types of subject fields are referred to as ‘subject field 1, subject field 2, subject field 3, ...’ and the assigned association ranking for ‘subject field 1, subject field 2, subject field 3, ...’ is ‘1st, 3rd, 5th, ...’, and the association vector generation unit (125) can generate an association vector ‘[1 3 5 ...]’ having the corresponding association rankings ‘1, 3, 5, ...’ as components.

[0097] The extraction processing unit (126) calculates the vector similarity between the feature vector corresponding to each of the plurality of users and the association vector, and the degree of alignment between the similarity rank for the n types of subject fields corresponding to each of the plurality of users and the degree of alignment specified for the n types of subject fields, and then extracts the user among the plurality of users whose similarity rank has a degree of alignment that exceeds the reference value.

[0098] Here, vector similarity can be calculated using methods such as Euclidean distance or cosine similarity.

[0099] In this regard, when the feature vector corresponding to each of the above multiple users is 'feature vector 1, feature vector 2, feature vector 3, ...' as shown in Table 9, the extraction processing unit (126) can calculate the vector similarity between each of 'feature vector 1, feature vector 2, feature vector 3, ...' and the association vector as the degree of alignment between the similarity ranking for the n types of subject fields corresponding to each of the above multiple users and the association ranking specified for the n types of subject fields.

[0100] Then, the extraction processing unit (126) can extract users among the plurality of users whose calculated match value has a similarity rank exceeding the threshold value. In this case, as one embodiment, if the users whose calculated match value has a similarity rank exceeding the threshold value are 'User 1, User 3, User 5, User 7, User 9', the extraction processing unit (126) can extract 'User 1, User 3, User 5, User 7, User 9' from among the plurality of users 'User 1, User 2, User 3, ...'.

[0101] That is, the extraction processing unit (126) can extract users among the plurality of users who have a similarity ranking similar to the association ranking for the n types of subject fields specified by the surveyor (130).

[0102] The selection unit (118), when first users are extracted as a result of extraction by the extraction unit (117) and have a similarity ranking exceeding the threshold among the plurality of users, refers to the address database (112) to verify address information corresponding to each of the first users, and then selects users among the first users who have address information including at least one first region name.

[0103] In relation to this, as in the example above, let us assume that the first users, 'User 1, User 3, User 5, User 7, User 9', were extracted through the extraction unit (117).

[0104] Then, the selection unit (118) can check the address information corresponding to each of the first users, 'User 1, User 3, User 5, User 7, User 9', by referring to the address database (112) as shown in Table 2 above.

[0105] Then, the selection unit (118) can select a user among the first users, 'User 1, User 3, User 5, User 7, User 9', who has address information including at least one first regional name. If the at least one first regional name is 'Seoul Metropolitan City, Busan Metropolitan City', and the address information corresponding to each of 'User 1, User 3, User 5, User 7, User 9' is 'Yeouido-dong, Yeongdeungpo-gu, Seoul Metropolitan City, Jung-dong, Haeundae-gu, Busan Metropolitan City, Yeonmu-dong, Jangan-gu, Suwon-si, Gyeonggi-do, Gap-dong, Yuseong-gu, Daejeon Metropolitan City, Gahyeon-dong, Wonju-si, Gangwon-do', the selection unit (118) can select 'User 1', who has address information including the regional name 'Seoul Metropolitan City', and 'User 3', who has address information including the regional name 'Busan Metropolitan City', among the first users, 'User 1, User 3, User 5, User 7, User 9'.

[0106] When second users are selected as users among the first users who have address information including at least one first region name as a result of selection by the selection unit (118), the push message transmission unit (119) refers to the device information database (111) to verify device identification information for each of the second users' electronic terminals, identifies each of the second users' electronic terminals based on the verified device identification information, and then transmits a push message containing a request to participate in a survey to each of the second users' electronic terminals.

[0107] Let us assume that the second users, 'User 1, User 3', were selected through the selection unit (118) as in the example above.

[0108] Then, the push message transmission unit (119) can check the device identification information for each of the electronic terminals of the second users, 'User 1, User 3', as 'Device identification information 1, Device identification information 3' by referring to the device information database (111) as shown in Table 1 above.

[0109] Then, the push message transmission unit (119) can identify each of the electronic terminals of the second users, 'User 1, User 3', based on 'device identification information 1, device identification information 3', and then transmit a push message containing a request to participate in a survey to each of the electronic terminals of the second users, 'User 1, User 3'.

[0110] That is, the service providing device (110) of the present invention can support the surveyor in collecting high-quality survey results more easily by selecting users among the plurality of users who match the region name and subject field related to the survey designated by the surveyor as survey subjects and sending a message requesting participation in the survey to the electronic terminals of the selected users.

[0111] FIG. 2 is a flowchart illustrating the operation method of a service providing device according to an embodiment of the present invention.

[0112] In step (S210), a device information database is maintained in which device identification information of electronic terminals possessed by each of the multiple users is stored.

[0113] In step (S220), an address database is maintained in which address information corresponding to each of the plurality of users is stored.

[0114] In step (S230), an interest keyword database is maintained in which an interest keyword set corresponding to each of the plurality of users (the interest keyword set corresponding to each of the plurality of users is a set consisting of two or more words designated by each user among a pre-specified plurality of words) is stored.

[0115] In step (S240), a word database is maintained in which word sets corresponding to each of n (n is a natural number greater than or equal to 2) of pre-classified subject fields are stored (the word sets corresponding to each of the n types of subject fields are sets composed of two or more words pre-designated as corresponding to each subject field among the plurality of words).

[0116] In step (S250), when a filtering request command for selecting survey subjects is received from an investigator terminal held by the surveyor, a first guidance message is transmitted to the investigator terminal instructing it to reply by specifying a region name for selecting survey subjects, and at the same time, a second guidance message is transmitted instructing it to reply by specifying a ranking of the degree of association with the survey for each of the n types of subject fields.

[0117] In step (S260), when at least one first regional name is replied from the investigator terminal in response to the first guidance message and a correlation ranking for the n types of subject fields designated by the surveyor is replied in response to the second guidance message, for each of the plurality of users, the similarity between the word set corresponding to each of the n types of subject fields and the set of interest keywords corresponding to each user is calculated, and then a process of assigning rankings in order of high similarity for the n types of subject fields is performed, thereby calculating the similarity ranking for the n types of subject fields corresponding to each of the plurality of users.

[0118] In step (S270), among the plurality of users, users are extracted who have a similarity rank in which the degree of match with the specified association rank for the n types of subject fields exceeds a preset threshold.

[0119] In step (S280), if first users are extracted as a result of the extraction in step (S270) among the plurality of users having a similarity ranking exceeding the threshold, the address database is referenced to confirm the address information corresponding to each of the first users, and then a user among the first users who has address information including at least one first region name is selected.

[0120] In step (S290), if second users are selected as users among the first users who have address information including at least one first region name as a result of the selection in step (S280), the device identification information for each of the second users' electronic terminals is verified by referring to the device information database, and after identifying each of the second users' electronic terminals based on the verified device identification information, a push message containing a request to participate in a survey is transmitted to each of the second users' electronic terminals.

[0121] At this time, according to one embodiment of the present invention, step (S260) may include a step of assigning a random number to each of the plurality of users; a step of storing the numbers of the plurality of users in ascending order in a queue when a number is assigned to each of the plurality of users; a first process of extracting the fastest number from the queue when the number of users is stored in the queue; a second process of designating the user according to the number extracted from the queue as a reference user and calculating the similarity between a set of words corresponding to each of the n types of subject fields and a set of interest keywords corresponding to the reference user; a third process of assigning ranks in order of high similarity to the n types of subject fields; and a fourth process of removing the number extracted from the queue through the first process from the queue, by repeating these steps until no number remains in the queue, thereby calculating the similarity ranking for the n types of subject fields corresponding to each of the plurality of users.

[0122] At this time, according to one embodiment of the present invention, step (S260) may further include the step of maintaining a vector database in which word vectors corresponding to each of the plurality of words are stored (the word vectors corresponding to each of the plurality of words are pre-embedded vectors based on semantic similarity between words, and are specified so that a higher vector similarity is calculated as the semantic similarity between words is higher).

[0123] At this time, the step of calculating processing above, when calculating the similarity between the word set corresponding to each of the n types of subject fields and the set of interest keywords corresponding to the reference user through the second process, if a situation arises where the similarity between the word set corresponding to the first subject field, which is one of the n types of subject fields, and the set of interest keywords corresponding to the reference user needs to be calculated, the word vectors for the first words included in the word set corresponding to the first subject field and the second words included in the set of interest keywords corresponding to the reference user are identified by referring to the vector database, then all vector similarities between each of the first words and each of the second words are calculated, and when the calculated vector similarities are sorted in ascending order to calculate the quartiles, the overall average of the remaining vector similarities, excluding the vector similarities corresponding to the first quartile or lower, can be calculated as the similarity between the word set corresponding to the first subject field and the set of interest keywords corresponding to the reference user.

[0124] In addition, according to one embodiment of the present invention, step (S270) may include the step of generating a feature vector corresponding to each of the plurality of users by performing a process of constructing an n-dimensional vector having a similarity ranking for each of the n types of subject fields corresponding to each user as a component, the step of generating an n-dimensional association vector having a specified association ranking for the n types of subject fields as a component, and the step of calculating the vector similarity between the feature vector corresponding to each of the plurality of users and the association vector as the degree of alignment between the similarity ranking for each of the n types of subject fields corresponding to each of the plurality of users and the degree of alignment specified for the n types of subject fields, and then extracting a user among the plurality of users whose similarity ranking has a calculated degree of alignment exceeding a reference value.

[0125] Hereinafter, a method of operation of a service providing device according to an embodiment of the present invention has been described with reference to FIG. 2. Since the method of operation of a service providing device according to an embodiment of the present invention may correspond to the configuration of the operation of the service providing device (110) described using FIG. 1, a more detailed description thereof will be omitted.

[0126] The method of operation of a service providing device according to one embodiment of the present invention may be implemented as a computer program stored in a storage medium for execution through combination with a computer.

[0127] In addition, the method of operation of a service providing device according to one embodiment of the present invention may be implemented in the form of computer program instructions for execution through combination with a computer and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., either individually or in combination. The program instructions recorded on the medium may be those specifically designed and configured for the present invention, or they may be those known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory. Examples of program instructions include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc.

[0128] As described above, the present invention has been explained by specific details such as specific components, limited embodiments, and drawings; however, this is provided merely to aid in a more comprehensive understanding of the invention, and the invention is not limited to the above embodiments. A person skilled in the art can make various modifications and variations from this description.

[0129] Therefore, the scope of the present invention is not limited to the described embodiments, and all things equivalent to or having equivalent variations to the claims set forth below, as well as the claims set forth below, shall be considered to fall within the scope of the concept of the present invention. Explanation of the symbols

[0130] 110: Service providing device 111: Device information database 112: Address database 113: Interest Keyword Database 114: Word Database 115: Guidance message transmission unit 116: Ranking calculation unit 117: Extraction unit 118: Sorting unit 119: Push message transmission unit 120: Sequence assignment unit 121: Queue Processing Unit 122: Calculation Processing Unit 123: Vector Database 124: Feature Vector Generation Unit 125: Association Vector Generation Unit 126: Extraction Processing Unit 130: Surveyor 140: Investigator Terminal

Claims

Claim 1 A device information database storing device identification information of electronic terminals possessed by each of multiple users; an address database storing address information corresponding to each of the multiple users; an interest keyword database storing a set of interest keywords corresponding to each of the multiple users—wherein the set of interest keywords corresponding to each of the multiple users is a set composed of two or more words designated by each user among a pre-specified set of words; a word database storing a set of words corresponding to each of the pre-classified n (n is a natural number greater than or equal to 2) types of subject fields—wherein the set of words corresponding to each of the n types of subject fields is a set composed of two or more words pre-specified as corresponding to each subject field among the multiple words; and a guidance message transmission unit that, upon receiving a filtering request command for selecting survey subjects from a surveyor terminal possessed by a surveyor, transmits a first guidance message instructing the surveyor terminal to reply by specifying a region name for selecting survey subjects, and simultaneously transmits a second guidance message instructing the surveyor to reply by specifying a ranking of the degree of relevance to the survey for each of the n types of subject fields. A ranking calculation unit that calculates similarity rankings for the n types of subject fields corresponding to each of the plurality of users by performing a process of calculating the similarity between a set of words corresponding to each of the n types of subject fields and a set of interest keywords corresponding to each user for each of the plurality of users, and then assigning rankings to the n types of subject fields in order of high similarity, when at least one first region name is replied from the investigator terminal in response to the first guidance message and a correlation ranking for the n types of subject fields designated by the surveyor is replied in response to the second guidance message;An extraction unit for extracting a user among the plurality of users having a similarity rank in which the degree of matching with the specified association rank for the n types of subject fields exceeds a preset threshold; and a selection unit for, when first users are extracted as the extraction result of the extraction unit as users among the plurality of users having a similarity rank exceeding the threshold, confirming address information corresponding to each of the first users by referring to the address database, and then selecting a user among the first users having address information including at least one first region name. The system includes a push message transmission unit that, when second users are selected as users among the first users who have address information including at least one first region name based on the selection result of the selection unit, checks device identification information for each of the second users' electronic terminals by referring to the device information database, identifies each of the second users' electronic terminals based on the confirmed device identification information, and then transmits a push message containing a request to participate in a survey to each of the second users' electronic terminals; and the ranking unit includes a vector database in which word vectors corresponding to each of the plurality of words are stored—the word vectors corresponding to each of the plurality of words are pre-embedded vectors based on semantic similarity between words, and are designated so that a higher vector similarity is calculated as the semantic similarity between words is higher—; a sequence assignment unit that assigns a random sequence number to each of the plurality of users; and a queue processing unit that stores the sequence numbers of the plurality of users in ascending order in a queue when a sequence number is assigned to each of the plurality of users.The method comprises a calculation processing unit that calculates and processes similarity rankings for the n types of subject fields corresponding to each of the plurality of users by repeating the following steps until no sequence numbers remain in the queue: a first step of extracting the fastest sequence number from the queue when the sequence number of the plurality of users is stored in the queue; a second step of designating the user corresponding to the sequence number extracted from the queue as a reference user and calculating the similarity between a word set corresponding to each of the n types of subject fields and a set of interest keywords corresponding to the reference user; a third step of assigning ranks to the n types of subject fields in order of high similarity; and a fourth step of removing the sequence number extracted from the queue through the first step from the queue, thereby calculating similarity rankings for the n types of subject fields corresponding to each of the plurality of users. When the calculation processing unit calculates the similarity between a word set corresponding to each of the n types of subject fields and a set of interest keywords corresponding to the reference user through the second step, if a situation arises where the similarity between a word set corresponding to a first subject field, which is one of the n types of subject fields, and a set of interest keywords corresponding to the reference user needs to be calculated, the calculation processing unit refers to the vector database and calculates the similarity for the first subject field A service providing device characterized by verifying word vectors for first words included in a corresponding word set and second words included in a set of interest keywords corresponding to the reference user, calculating all vector similarities between each of the first words and each of the second words, and when the calculated vector similarities are sorted in ascending order to calculate quartiles, calculating the overall average of the remaining vector similarities, excluding vector similarities corresponding to the first quartile or lower, as the similarity between the word set corresponding to the first subject field and the set of interest keywords corresponding to the reference user. Claim 2 delete Claim 3 delete Claim 4 A service providing device comprising: a feature vector generating unit that generates a feature vector corresponding to each of the plurality of users by performing a process of constructing an n-dimensional vector having a similarity ranking for each of the n types of subject fields corresponding to each user as a component for each of the plurality of users; an association vector generating unit that generates an n-dimensional association vector having a specified association ranking for the n types of subject fields as a component; and an extraction processing unit that calculates the vector similarity between the feature vector corresponding to each of the plurality of users and the association vector as the degree of alignment between the similarity ranking for each of the n types of subject fields corresponding to each of the plurality of users and the degree of alignment specified for the n types of subject fields, and then extracts a user among the plurality of users whose similarity ranking has a calculated degree of alignment exceeding a reference value. Claim 5 A step of maintaining a device information database in which device identification information of electronic terminals possessed by each of multiple users is stored; a step of maintaining an address database in which address information corresponding to each of the multiple users is stored; a step of maintaining an interest keyword database in which a set of interest keywords corresponding to each of the multiple users is stored—the set of interest keywords corresponding to each of the multiple users is a set composed of two or more words designated by each user among a pre-specified set of words; a step of maintaining a word database in which a set of words corresponding to each of the pre-classified n types of subject fields (n is a natural number greater than or equal to 2) is stored—the set of words corresponding to each of the n types of subject fields is a set composed of two or more words pre-specified as corresponding to each subject field among the multiple words; and a step of, when a filtering request command for selecting survey subjects is received from an investigator terminal possessed by a surveyor, transmitting a first guidance message to the investigator terminal instructing it to reply by specifying a region name for selecting survey subjects, and simultaneously, a second guidance message instructing it to reply by specifying a ranking of relevance to the survey for each of the n types of subject fields. A step of transmitting; and, when at least one first place name is replied from the investigator terminal in response to the first guidance message and a correlation ranking for the n types of subject fields designated by the surveyor is replied in response to the second guidance message, for each of the plurality of users, a process of calculating the similarity between a word set corresponding to each of the n types of subject fields and a set of interest keywords corresponding to each user, and then assigning rankings to the n types of subject fields in order of high similarity, thereby calculating the similarity ranking for the n types of subject fields corresponding to each of the plurality of users.A step of extracting a user among the plurality of users having a similarity rank in which the degree of matching with the specified association rank for the n types of subject fields exceeds a preset threshold; and, if first users are extracted as a result of the extraction step, among the plurality of users having a similarity rank exceeding the threshold, a step of confirming address information corresponding to each of the first users by referring to the address database, and then selecting a user among the first users having address information including at least one first region name. The method includes the step of, when the second users are selected as the result of the selection step described above, as users having address information including at least one first region name among the first users, confirming device identification information for each of the second users' electronic terminals by referring to the device information database, identifying each of the second users' electronic terminals based on the confirmed device identification information, and then sending a push message containing a request to participate in a survey to each of the second users' electronic terminals; and the step of calculating the similarity ranking includes the step of maintaining a vector database in which word vectors corresponding to each of the plurality of words are stored—the word vectors corresponding to each of the plurality of words are pre-embedded vectors based on semantic similarity between words, and are designated so that a higher vector similarity is calculated as the semantic similarity between words is higher—; the step of assigning a random sequence number to each of the plurality of users; and the step of storing the sequence numbers of the plurality of users in ascending order in a queue when a sequence number is assigned to each of the plurality of users.The method comprises a first step of extracting the fastest sequence number from the queue when the sequence numbers of the plurality of users are stored in the queue; a second step of designating the user corresponding to the sequence number extracted from the queue as a reference user and calculating the similarity between a word set corresponding to each of the n types of subject fields and a set of interest keywords corresponding to the reference user; a third step of assigning ranks to the n types of subject fields in order of high similarity; and a fourth step of removing the sequence number extracted from the queue through the first step from the queue, by repeating these steps until no sequence numbers remain in the queue, thereby calculating the similarity rankings for the n types of subject fields corresponding to each of the plurality of users. In the calculation step, when calculating the similarity between the word set corresponding to each of the n types of subject fields and the set of interest keywords corresponding to the reference user through the second step, if a situation arises where the similarity between a word set corresponding to a first subject field, which is one of the n types of subject fields, and a set of interest keywords corresponding to the reference user needs to be calculated, the word corresponding to the first subject field is referenced from the vector database. A method of operation of a service providing device characterized by verifying word vectors for first words included in a set and second words included in a set of interest keywords corresponding to the reference user, calculating all vector similarities between each of the first words and each of the second words, and when the calculated vector similarities are sorted in ascending order to calculate quartiles, calculating the overall average of the remaining vector similarities, excluding vector similarities corresponding to the first quartile or lower, as the similarity between the word set corresponding to the first subject field and the set of interest keywords corresponding to the reference user. Claim 6 delete Claim 7 delete Claim 8 In claim 5, the extraction step comprises: a step of generating a feature vector corresponding to each of the plurality of users by performing a process of constructing an n-dimensional vector having a similarity rank for each of the n types of subject fields corresponding to each user as a component for each of the plurality of users; a step of generating an n-dimensional association vector having a specified association rank for the n types of subject fields as a component; and a step of calculating the vector similarity between the feature vector corresponding to each of the plurality of users and the association vector as the degree of alignment between the similarity rank for each of the n types of subject fields corresponding to each of the plurality of users and the degree of alignment specified for the n types of subject fields, and then extracting a user among the plurality of users whose similarity rank has a calculated degree of alignment exceeding a reference value. Claim 9 A computer-readable recording medium having a computer program for executing the method of either paragraph 5 or paragraph 8 through combination with a computer. Claim 10 A computer program stored on a storage medium for executing the method of either paragraph 5 or paragraph 8 through combination with a computer.

Citation Information

Patent Citations

  • Method of providing mapped keyword-item information service based on real-time news article analysis

    KR1020220061408A

  • Research message providing method and system

    KR1020230127888A

  • Mobile targeting advertising system for local residents

    KR102026900B1

  • Apparatus and method for recommending news of user interest using user-defined topic modeling

    KR102368043B1