KR20230049486A - Political tendency analysis device and service providing method using the same - Google Patents

Abstract

The present invention discloses a political propensity analysis device and a service providing method using the same. The political propensity analysis device includes a processor and a memory operatively connected to the processor, and the memory, when executed, causes the processor to collect information related to the user, and the information includes keywords related to politics. Based on the fact that the information contains keywords related to politics, identify the information as information to be evaluated, identify whether the information to be evaluated includes keywords related to political orientation, and determine whether the information to be evaluated contains keywords related to political orientation. Based on the inclusion of related keywords, instructions for assigning points to information to be evaluated may be stored. In addition to this, various embodiments grasped through the specification are possible.

Description

Political tendency analysis device and service providing method using the same

The present invention relates to a political propensity analysis device and a service providing method using the same. Specifically, it relates to a political tendency analysis apparatus for analyzing political orientation using data related to a user and a service providing method using the same.

The contents described in this part merely provide background information on the present embodiment and do not constitute prior art.

To analyze people's political tendencies, a method of analyzing political tendencies by randomly making phone calls from a public opinion survey system using a communication device or a polling organization may be used. On the other hand, analysis of people's political orientation can also be conducted on the Internet. Political propensity analysis using the Internet can be done in the form of a survey on individual sites.

However, the analysis of political propensity, which is conducted in the form of a survey on individual sites, may not be carried out accurately because an integrated search is not possible or it is only collected collectively. In addition, the accuracy of the user's political orientation analysis may be reduced due to limited and uniform questions.

An object of the present invention is to provide a political propensity analysis device capable of analyzing a user's political propensity using user-related data and a service providing method using the same.

In addition, the present invention is a political tendency analysis device that analyzes a user's political orientation using a deep learning module learned in advance based on learning data to which scores for each political orientation are added based on texts written by the user on the Internet, and a service using the same It is to provide a method of delivery.

An apparatus for analyzing political propensity according to an embodiment of the present invention includes a processor and a memory operatively connected to the processor, and the memory, when executed, causes the processor to collect information related to a user and store the information in the information. identify whether a keyword related to politics is included, identify the information as information to be evaluated based on whether the information includes a keyword related to politics, identify whether the information to be evaluated includes a keyword related to political orientation, Based on the fact that the evaluation target information includes a keyword related to a political orientation, instructions for assigning scores to the evaluation target information may be stored.

In one embodiment, the instructions may cause the processor to provide a question of the survey and answer data to the question as one piece of survey information, based on the fact that the information is information related to the survey, and to include the information in the survey information. You can have it identify if it contains keywords related to politics.

In one embodiment, the instructions cause the processor to provide a test question, receive answer data to the test question, and send the test question and the answer data to the test question to collect the information. It is provided as one test result information, and it is possible to identify whether a keyword related to the politics is included in the test result information.

In one embodiment, the instructions may cause the processor to identify whether a keyword related to the political orientation is included in the evaluation target information based on deep learning technology, and to determine whether the evaluation target information includes a keyword related to the political orientation. Scores may be assigned to the evaluation target information based on what is included.

In an embodiment, the instructions may cause the processor to receive the user's region information and to assign the score to the evaluation target information based on the user's region information.

In one embodiment, the deep learning model includes an input layer having, as an input node, a keyword related to the evaluation target information and a political orientation determined with respect to the evaluation target information, and a score assigned to the evaluation target information as an output node. and one or more hidden layers disposed between the input layer and the output layer, wherein weights of nodes and edges between the input node and the output node are updated by a training process of the deep learning model. can

In addition, the method according to an embodiment of the present invention includes an operation of collecting information related to a user, an operation of identifying whether a keyword related to politics is included in the information, and an operation of determining whether the information includes a keyword related to politics. Based on the operation of identifying the information as evaluation target information, identifying whether the evaluation target information includes a keyword related to political orientation, and based on the fact that the evaluation target information includes a keyword related to the political orientation , may include an operation of assigning a score to the evaluation target information.

In one embodiment, based on the fact that the information is information related to the survey, the operation of providing the question of the survey and the answer data to the question as one survey information, and the keyword related to the politics is included in the survey information It may further include an operation to identify whether there is.

In one embodiment, to collect the information, providing a test question, receiving answer data for the test question, combining the test question and the answer data for the test question into one piece of test result information. It may further include an operation of providing, and an operation of identifying whether a keyword related to the politics is included in the test result information.

In one embodiment, an operation of identifying whether the evaluation target information includes a keyword related to the political orientation, and scoring the evaluation target information based on whether the evaluation target information includes a keyword related to the political orientation. The granting operation may be based on deep learning technology.

In one embodiment, the operation of receiving the region information of the user further includes, and the operation of assigning a score to the evaluation target information, based on the region information of the user, to assign the score to the evaluation target information. action may be included.

The present invention analyzes the user's political orientation using text written by the user on the Internet, analyzes the sentence containing the political orientation, and adds a score to the political orientation of the analyzed sentence, thereby improving the accuracy of the user's political orientation analysis. can improve

In addition, the present invention can improve the accuracy of political orientation analysis by learning a deep learning module using learning data including determination of political orientation and analyzing the user's political orientation using the learned deep learning module. there is.

In addition, the present invention analyzes the user's political orientation based on the user's writing on the Internet and the user's local information, and generates and provides visualized data based on this, thereby improving the convenience and accuracy of political orientation analysis. there is.

In addition to the above description, specific effects of the present invention will be described together while explaining specific details for carrying out the present invention.

1 is a block diagram for explaining a political propensity analysis apparatus according to an embodiment of the present invention.
2 is a block diagram illustrating a processor of a political propensity analysis device according to an embodiment of the present invention.
3 is a diagram for explaining that a political propensity analysis apparatus according to an embodiment of the present invention uses a deep learning model.
4 is a diagram showing the configuration of the deep learning model of FIG. 3 by way of example.
5 is a flowchart illustrating a service providing method using a political propensity analysis device according to an embodiment of the present invention.
6 is a flowchart illustrating a method of providing a service using a political propensity analysis apparatus according to an embodiment of the present invention.
7 is a flowchart illustrating a method of providing a service using a political propensity analysis device according to an embodiment of the present invention.
8 is a diagram for explaining an example of using a service providing method using a political propensity analysis device according to an embodiment of the present invention.

Terms or words used in this specification and claims should not be construed as being limited to a general or dictionary meaning. According to the principle that an inventor may define a term or a concept of a word in order to best describe his/her invention, it should be interpreted as meaning and concept consistent with the technical spirit of the present invention. In addition, the embodiments described in this specification and the configurations shown in the drawings are only one embodiment in which the present invention is realized, and do not represent all of the technical spirit of the present invention, so they can be replaced at the time of the present application. It should be understood that there may be many equivalents and variations and applicable examples.

Terms such as first, second, A, and B used in this specification and claims may be used to describe various components, but the components should not be limited by the terms. These terms are only used for the purpose of distinguishing one component from another. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element, without departing from the scope of the present invention. The term 'and/or' includes a combination of a plurality of related recited items or any one of a plurality of related recited items.

Terms used in this specification and claims are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. It should be understood that terms such as "include" or "having" in this application do not exclude in advance the possibility of existence or addition of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification. .

Unless defined otherwise, all terms used herein, 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.

Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in the present application, they should not be interpreted in an ideal or excessively formal meaning. don't In addition, each configuration, process, process or method included in each embodiment of the present invention may be shared within a range that does not contradict each other technically.

Hereinafter, a political propensity analysis device and a service providing method thereof according to an embodiment of the present invention will be described with reference to FIGS. 1 to 8 .

1 is a block diagram for explaining a political propensity analysis apparatus according to an embodiment of the present invention. 2 is a block diagram illustrating a processor of a political propensity analysis device according to an embodiment of the present invention.

Referring to FIGS. 1 and 2 , the apparatus 101 for analyzing political tendencies according to an embodiment of the present invention may include a processor 110 , a memory 120 and a communication module 130 . The political orientation analysis device 101 may include at least one additional component in addition to the components shown in FIG. 1 . According to one embodiment, the components of the political tendency analysis device 101 may be the same entity or constitute a separate entity.

In one embodiment, one or more other components may be added to the political orientation analysis device 101 . Also, in other embodiments, some of these components may be implemented as a single integrated circuit.

The processor 110 may, for example, execute software to control at least one other component (eg, hardware or software component) of the political tendency analysis device 101 connected to the processor 110, and may control various components. Data processing or computation can be performed. According to one embodiment, as at least part of data processing or operation, processor 110 loads information, instructions, or data received from other components (eg, communication module 130) into memory 120, and Information, commands or data stored in 120 may be processed, and resultant data may be stored in memory 120 .

The memory 120 may store various data used by at least one component (eg, the processor 110) of the political propensity analysis device 101 . Data may include, for example, input data or output data for software and related instructions.

The memory 120 may store commands, information, or data related to the operation of components included in the political tendency analysis device 101 . For example, memory 120 may store instructions that, when executed, enable processor 110 to perform various operations described herein.

The political orientation analysis device 101 may communicate with one or more users and/or one or more external devices through the communication module 130 . The communication module 130 establishes a direct (eg, wired) communication channel or a wireless communication channel between the political tendency analysis device 101 and an external electronic device (eg, another electronic device or server), and communicates through the established communication channel. can support implementation. The communication module 130 may include one or more communication processors that operate independently of the processor 110 (eg, an application processor) and support direct (eg, wired) communication or wireless communication. According to an embodiment, the communication module 130 is a wireless communication module (eg, a cellular communication module, a short-range wireless communication module, or a global navigation satellite system (GNSS) communication module) or a wired communication module (eg, a local area network (LAN)). ) communication module, or power line communication module).

The processor 110 may be operatively coupled to the memory 120 and the communication module 130 in order to perform overall functions of the political tendency analysis device 101 . Processor 110 may include, for example, one or more processors. The one or more processors may include, for example, an image signal processor (ISP), an application processor (AP), or a communication processor (CP).

In one embodiment, the processor 110 may include a first module 111 and a second module 112 .

The first module 111 of the processor 110 may collect user-related information. The collected information may include, for example, text written by the user on the Internet. For example, the first module 111 of the processor 110 may crawl information related to a user.

In an embodiment, based on information related to a survey, the first module 111 of the processor 110 may provide a question of the survey and answer data to the question as one piece of survey information. For example, when a user answers a questionnaire related to politics on the Internet, the information may be referred to as information related to the questionnaire (hereinafter, questionnaire information). At this time, the first module 111 of the processor 110 may identify the survey information as information related to the user.

In one embodiment, the first module 111 of the processor 110 may provide test questions to gather information. The test question may be, for example, a question provided to obtain information including a keyword related to politics when information that can be identified as evaluation target information is not obtained from the user. The first module 111 of the processor 110 may receive answer data to a test question from a user. The first module 111 of the processor 110 may provide test questions and answer data to the test questions as one test result information. The first module 111 of the processor 110 may identify test result information as user-related information.

The first module 111 of the processor 110 may identify whether a keyword related to politics is included in the information. The keyword related to politics may include, for example, a name of a specific political party, a word related to a specific political party, a word related to the current government, or a word related to a policy of the current government.

The first module 111 of the processor 110 may, for example, identify whether the information includes a keyword related to politics, and tag whether the information includes a keyword related to politics. The first module 111 of the processor 110 may identify information as evaluation target information based on the fact that the information includes a keyword related to politics.

In one embodiment, the second module 112 of the processor 110 may identify whether a keyword related to political orientation is included in the evaluation target information. Keywords related to political orientation may include, for example, words related to evaluation of a specific political party, words evaluating the current government, or words expressing opinions on policies of the current government.

In one embodiment, the second module 112 of the processor 110 may assign a score to the evaluation target information based on the fact that the evaluation target information includes a keyword related to a political orientation. For example, the second module 112 of the processor 110 determines a relatively high (or low) value for evaluation target information based on the fact that the evaluation target information includes a keyword related to a political orientation having a high progressive tendency. points can be awarded. The second module 112 of the processor 110 may assign a relatively low (or high) score based on the fact that the evaluation target information includes a keyword related to a political orientation having a high conservative orientation.

For example, assuming that a user writes 'I strongly agree with the policy promoted by the current government' on the Internet, the first module 111 of the processor 110 identifies the text written by the user as information and collects it. can do. The first module 111 of the processor 110 may identify information as evaluation target information based on whether the information includes a keyword related to politics (eg, 'government' or 'policy'). The second module 112 of the processor 110 may assign a score to the evaluation target information based on the fact that the evaluation target information includes a keyword (eg, 'strongly agree') related to a political orientation. At this time, the second module 112 of the processor 110 is based on the fact that the information to be evaluated includes keywords (eg, 'government' and 'strongly agree') related to a political orientation having a high progressive tendency. , a high (or low) score can be given to the information to be evaluated.

In one embodiment, the second module 112 of the processor 110 may receive the user's region information based on the fact that the evaluation target information does not include a keyword related to political orientation. For example, the second module 112 of the processor 110 may receive regional information related to the user, for information including keywords related to politics, but it is difficult to determine the political orientation. The second module 112 of the processor 110 may assign a score to evaluation target information based on the user's region information.

For example, assuming that a user has written a text saying 'I don't like party A and I don't like party B' on the Internet, the first module 111 of the processor 110 identifies the text written by the user as information and can be collected The first module 111 of the processor 110 may identify information as evaluation target information based on the fact that the information includes a keyword related to politics (eg, 'per A' or 'per B). Meanwhile, the second module 112 of the processor 110 may identify that the evaluation target information does not include a keyword related to political orientation. In this case, the second module 112 of the processor 110 may assign a score to evaluation target information based on the user's region information.

However, the present invention is not limited thereto, and the second module 112 receives regional information related to the user even when a keyword related to political orientation is included in the evaluation target information, and performs evaluation using the received regional information. Scores can be assigned to target information.

In one embodiment, the second module 112 of the processor 110 may use Natural Language Processing (NLP) to assign scores to information to be evaluated.

In one embodiment, the second module 112 of the processor 110 may generate data associating the assigned score with the user.

The political orientation analysis device 101 according to an embodiment of the present invention assigns scores to articles (eg, information) containing keywords related to politics among texts written by users on the Internet, so as to determine the user's political orientation. can do.

In one embodiment, the processor 110 may generate visual data based on data relating a user to a score assigned to evaluation target information. The processor 110 may visualize a result of analyzing a user's political orientation in relation to a specific subject, for example. For example, if the processor 110 wants to generate visual data in relation to the subject of 'specialized high school', the processor 110 may include a keyword related to politics (eg, specialized high school) and political orientation in information related to the user. Scores can be assigned to the information to be evaluated based on the inclusion of relevant keywords (eg, maintenance or deprecation). In this case, the processor 110 may generate data indicating what kind of political inclination the users included in the area within a certain range have on the subject of 'specialized high school'. The processor 110 may generate and provide visualized visual data based on the generated data.

Hereinafter, a political propensity analysis device according to an embodiment of the present invention will be described with reference to FIGS. 3 and 4 . For clarity of description, overlapping items described above may be simplified or omitted.

3 is a diagram for explaining that a processor according to an embodiment of the present invention uses a deep learning model. 4 is a diagram showing the configuration of the deep learning model of FIG. 3 by way of example.

Referring to FIG. 3 , in one embodiment, the database module 140 and the learning module 150 may be configured separately from the political propensity analysis device (101 in FIG. 1 ). However, it is not limited thereto, and in another embodiment of the present invention, the database module 140 and the learning module 150 may be configured to be included in the political tendency analysis device (101 in FIG. 1).

In one embodiment, the processor 110 identifies, based on the deep learning model, whether a keyword related to political orientation is included in the evaluation target information, and evaluates based on whether the evaluation target information includes a keyword related to the political orientation. Scores can be assigned to target information. For example, the processor 110 may assign scores to evaluation target information based on a deep learning model using the database module 140 and the learning module 150 .

The database module 140 may include an information collection unit 141 and a deep learning memory unit 143.

The information collection unit 141 may store various pieces of information received by the processor 110 . For example, the information collection unit 141 may store learning data TD. At this time, the information collection unit 141 may store and manage information related to the user by separately separating the user's account. For example, the database module 140 may receive, from the processor 110, evaluation target information, a keyword related to a political orientation determined for the evaluation target information, and a score assigned to the evaluation target information. The database module 140 may generate learning data (TD) based on evaluation target information, keywords related to the political orientation determined with respect to the evaluation target information, and scores assigned to the evaluation target information. The generated learning data TD may be stored in the information collecting unit 141 .

At this time, the deep learning model that identifies whether the information to be evaluated includes keywords related to political orientation and the deep learning model that outputs the political orientation score for the information to be evaluated may be configured as separate models. .

The deep learning memory unit 143 may store the deep learning model DL used for machine learning and the deep learning model TL learned by the training data TD. In this case, the learned deep learning model (TL) may generate result data including a score assigned to information to be evaluated by using an artificial neural network learned based on big data. For example, the database module 140, based on the training data (TD) stored in the information collection unit 141, the deep learning model (TL) learned by the learning module 150, the deep learning memory unit 143 ) can be stored.

To explain in more detail, deep learning technology, a type of machine learning, learns by going down to a deep level in multiple stages based on data.

Deep learning represents a set of machine learning algorithms that extract core data from a plurality of data while stepping up.

The deep learning model may use various well-known deep learning structures. For example, the deep learning model may include a convolutional neural network (CNN), a recurrent neural network (RNN), a deep belief network (DBN), and a graph neural network (GNN). structures can be used.

Specifically, CNN (Convolutional Neural Network) extracts the basic features of an object when a person recognizes an object, and then performs complex calculations in the brain to recognize the object based on the result. It is a simulated model.

RNN (Recurrent Neural Network) is widely used for natural language processing, etc., and is an effective structure for processing time-series data that changes over time.

DBN (Deep Belief Network) is a deep learning structure composed of multiple layers of RBM (Restricted Boltzman Machine), a deep learning technique. When a certain number of layers is obtained by repeating RBM (Restricted Boltzman Machine) learning, a DBN (Deep Belief Network) having a corresponding number of layers may be configured.

GNN (Graphic Neural Network, hereinafter, GNN) represents an artificial neural network structure implemented in a way to derive similarities and feature points between modeling data using modeling data modeled on the basis of data mapped between specific parameters. .

On the other hand, learning of the artificial neural network of the deep learning model can be performed by adjusting the weight of the connection line between nodes (and adjusting the bias value if necessary) so that a desired output is produced for a given input. In addition, the artificial neural network can continuously update weight values through learning.

In this case, both unsupervised learning and supervised learning may be used as machine learning methods of the artificial neural network. In addition, the deep learning model may be controlled to automatically update an artificial neural network structure for outputting coefficient data or coordinate data after learning according to settings.

The learning module 150 may perform a learning operation of an artificial neural network based on machine learning using the learning data TD as input data. That is, the learning module 150 may train the deep learning model DL stored in the deep learning memory unit 143 based on the learning data TD stored in the information collection unit 141 . The learned deep learning model (TL) trained in the learning module 150 is transferred back to the deep learning memory unit 143 to be stored separately or to update an existing model.

For example, the learning module 150 applies evaluation target information and a keyword related to the political orientation determined for the evaluation target information to an input node of the deep learning model (DL), and to an output node based on the keyword related to the political orientation. Then, the deep learning model (DL) can be trained by inputting the score assigned to the evaluation target information.

Referring to FIG. 4, the deep learning model (DL) outputs an input layer (Input) having as an input node evaluation target information and a keyword related to a political orientation determined for the evaluation target information, and a score assigned to the evaluation target information. It may include an output layer (Output) as a node, and M hidden layers disposed between the input layer (Input) and the output layer (Output).

A weight may be set on an edge connecting nodes of each layer. The presence or absence of weights or edges may be added, removed, or updated in the learning process. Through the learning process, weights of nodes and edges disposed between k input nodes and i output nodes may be updated.

Before the deep learning model (DL) performs learning, all nodes and edges may be set to initial values. However, when result data including cumulative evaluation target information, keywords related to political inclinations determined for evaluation target information, and scores assigned to evaluation target information are input, the weights of nodes and edges are changed, and in this process Matching is made between the parameters input as learning factors (e.g. keywords related to the evaluation target information and the political orientation determined for the evaluation target information) and the values assigned to the output node (e.g. scores given to the evaluation target information). can

In one embodiment, when using a cloud server (not shown), the deep learning model DL may receive and process a large number of parameters. Therefore, the deep learning model (DL) can perform learning based on a large amount of data.

Weights of nodes and edges between an input node and an output node performing the deep learning model (DL) may be updated by a learning process of the deep learning model (DL). In addition, it goes without saying that parameters output from the deep learning model (DL) can be additionally extended with various data.

Hereinafter, a service providing method using a political propensity analysis apparatus according to an embodiment of the present invention will be described with reference to FIG. 5 . For clarity of description, overlapping items described above may be simplified or omitted.

5 is a flowchart illustrating a service providing method using a political propensity analysis device according to an embodiment of the present invention.

Hereinafter, it is assumed that the political tendency analysis device (eg, the political tendency analysis device 101 of FIG. 1 ) performs the process of FIG. 5 . The operation described as being performed by the political tendency analysis device will be implemented as instructions (commands) that can be performed (or executed) by a processor (eg, the processor 110 of FIG. 1) of the political tendency analysis device. can The instructions may be stored in, for example, a computer recording medium or a memory (eg, memory 120 of FIG. 1 ) of a political tendency analysis device.

Referring to FIG. 5 , in operation S110, the apparatus for analyzing political propensity may collect information related to the user.

In operation S120, the political propensity analysis device may identify whether the information is evaluation target information. For example, the political orientation analysis device may identify whether a keyword related to politics is included in the information. The apparatus for analyzing political propensity may identify information as evaluation target information based on the fact that the information includes keywords related to politics.

In operation S130, the apparatus for analyzing political propensity may assign scores to the evaluation target information. For example, the political orientation analysis device may identify whether a keyword related to political orientation is included in evaluation target information. The political orientation analysis device may assign scores to the evaluation target information based on the fact that the evaluation target information includes a keyword related to the political orientation.

Hereinafter, with reference to FIG. 6, a method of providing a service using a political propensity analysis device according to an embodiment of the present invention will be described. For clarity of description, overlapping items described above may be simplified or omitted.

6 is a flowchart illustrating a method of providing a service using a political propensity analysis apparatus according to an embodiment of the present invention.

Hereinafter, it is assumed that the political tendency analysis device (eg, the political tendency analysis device 101 of FIG. 1 ) performs the process of FIG. 6 . The operation described as being performed by the political tendency analysis device will be implemented as instructions (commands) that can be performed (or executed) by a processor (eg, the processor 110 of FIG. 1) of the political tendency analysis device. can The instructions may be stored in, for example, a computer recording medium or a memory (eg, memory 120 of FIG. 1 ) of a political tendency analysis device.

Referring to FIG. 6 , in operation S210, the apparatus for analyzing political propensity may identify whether information related to a user can be collected.

Based on the fact that information related to the user can be collected (eg, YES in operation S210), the apparatus for analyzing political propensity may identify whether the information is survey information (operation S230).

Based on the fact that the information is not survey information (eg, NO in operation S230), the apparatus for analyzing political propensity may identify whether the information includes a keyword related to politics (operation S250).

Based on the fact that the information includes a keyword related to politics (eg, YES in operation S250), the apparatus for analyzing political propensity may identify the information as evaluation target information (operation S120). The political orientation analysis device may identify whether the evaluation target information includes a keyword related to political orientation, and assign a score to the evaluation target information based on whether the evaluation target information includes a keyword related to the political orientation (Fig. Operation of 5 (S130)).

Based on the impossibility of collecting information related to the user in operation S210 (eg, NO in operation S210), the apparatus for analyzing political tendencies may provide a test question (operation S211). It goes without saying that the test questions may be set in advance by an administrator or crawled and used by other organizations.

In operation S213, the apparatus for analyzing political tendencies may receive answer data including the user's answer to the test question.

In operation S215, the political tendency analysis device may provide test result information. For example, the apparatus for analyzing political tendencies may generate test questions and answer data to the test questions as one piece of test result information. The test result information is treated as user-related information, and the political orientation analysis device may identify whether a keyword related to politics is included in the information (operation S250).

On the other hand, based on the fact that the information is survey information in operation S230 (eg, YES in operation S230), the apparatus for analyzing political propensity may provide the questions of the survey and the answer data to the questions as one piece of survey information. (Operation (S231)). The survey information is treated as user-related information, and the apparatus for analyzing political propensity may identify whether a keyword related to politics is included in the information (operation S250).

Based on the fact that the information includes a keyword related to politics (eg, YES in operation S250), the apparatus for analyzing political propensity may identify the information as evaluation target information (operation S120).

A service providing method and apparatus using a political orientation analysis device according to an embodiment of the present invention analyzes a user's political orientation using information directly created by the user rather than a standardized question, thereby improving the accuracy of the political orientation analysis result can make it

A service providing method and apparatus using a political orientation analysis device according to an embodiment of the present invention can facilitate information collection by analyzing a user's political orientation using information directly created by the user rather than a standardized question. .

Hereinafter, with reference to FIG. 7 , a method of providing a service using a political propensity analysis apparatus according to an embodiment of the present invention will be described. For clarity of description, overlapping items described above may be simplified or omitted.

7 is a flowchart illustrating a method of providing a service using a political propensity analysis device according to an embodiment of the present invention.

Hereinafter, it is assumed that the political tendency analysis device (eg, the political tendency analysis device 101 of FIG. 1 ) performs the process of FIG. 7 . The operation described as being performed by the political tendency analysis device will be implemented as instructions (commands) that can be performed (or executed) by a processor (eg, the processor 110 of FIG. 1) of the political tendency analysis device. can The instructions may be stored in, for example, a computer recording medium or a memory (eg, memory 120 of FIG. 1 ) of a political tendency analysis device.

Referring to FIG. 7 , based on the information being identified as evaluation target information in operation S120, in operation S310, the political orientation analysis apparatus may identify whether the evaluation target information includes a keyword related to political orientation. there is.

Based on the fact that the evaluation target information includes a keyword related to political orientation (eg, YES in operation S310 ), the political orientation analysis apparatus may assign a score to the evaluation target information (operation S130 ). The user's political orientation may be identified by the political orientation analysis device assigning points to evaluation target information.

In the method of assigning points to information to be evaluated, the apparatus for analyzing political orientation according to an embodiment may preset and use different political orientation scores for each keyword. In addition, of course, the score for each set keyword can be adjusted later by inverse calculation based on the frequency of use of the keyword by people with specific political inclinations.

As another example, the apparatus for analyzing political propensity may derive and use a score according to information to be evaluated using a pre-learned deep learning model.

Based on the fact that a keyword related to political orientation is not included in the evaluation target information in operation S310 (eg, NO in operation S310), the apparatus for analyzing political orientation may receive regional information related to the user (operation ( S311)).

In operation S313, the apparatus for analyzing political tendencies may assign scores to evaluation target information based on regional information. For example, with respect to information including keywords related to politics but difficult to determine political orientation, the apparatus for analyzing political orientation may receive regional information related to the user. Then, based on the user's regional information, a score for the regional political orientation assigned to the corresponding region may be additionally assigned to the evaluation target information. However, this is only one example and the present invention is not limited thereto.

Accordingly, the present invention analyzes the political orientation of the user by using the text written by the user on the Internet, analyzes the sentence containing the political orientation, and adds a score to the political orientation of the analyzed sentence, thereby analyzing the political orientation of the user. accuracy can be improved.

In addition, the present invention can improve the accuracy of political orientation analysis by learning a deep learning module using learning data including determination of political orientation and analyzing the user's political orientation using the learned deep learning module. there is.

Hereinafter, a service providing method using a political propensity analysis apparatus according to an embodiment of the present invention will be described with reference to FIG. 8 . For clarity of description, overlapping items described above may be simplified or omitted.

8 is a diagram for explaining an example of using a service providing method using a political propensity analysis device according to an embodiment of the present invention.

Referring to FIG. 8 , the political orientation analysis device (101 in FIG. 1 ) generates data visualizing keywords and/or political orientations related to political orientations of a plurality of users based on evaluation target information of each of a plurality of users. can do.

For example, the political orientation analysis device may classify a plurality of users into a specific category (eg, region, age, gender, etc.). The apparatus for analyzing political orientation may generate keywords related to the political orientation of at least one or more users within a specific category and/or data visualizing the political orientation based on evaluation target information of at least one or more users corresponding to a specific category. .

For example, as shown in FIG. 8 , the apparatus for analyzing political tendencies may identify a plurality of users based on regional categories. The apparatus for analyzing political orientation may generate visualized data of keywords related to the political orientation of at least one user, based on the information to be evaluated of the identified at least one or more users.

On the other hand, keywords related to the political orientation of a plurality of users generated by the political orientation analysis device and/or data visualizing the political orientation are not limited to FIG. 8 . For example, the political orientation analysis device may provide keywords and/or political orientations related to political orientations of a plurality of users in the form of sentences. For example, the political orientation analysis device may identify at least one user associated with a keyword related to a specific political orientation. The apparatus for analyzing political orientation may generate sentence data representing the political orientation of the identified at least one or more users based on scores assigned to evaluation target information of the identified at least one or more users. For example, the apparatus for analyzing political orientation may identify at least one user related to information including a keyword related to the political orientation, based on a keyword related to the political orientation, such as a specialized high school. The apparatus for analyzing political orientation may identify, for example, a moderately conservative user among at least one or more users based on a score assigned to evaluation target information of at least one or more classified users. The apparatus for analyzing political orientation may generate sentence data representing the political orientation of the identified moderate-conservative user (eg, 'I think a specialized high school should be maintained.').

Through this, the present invention analyzes the user's political orientation based on the user's writing on the Internet and the user's local information, and generates and provides visualized data based on this, thereby improving the convenience and accuracy of political orientation analysis. can

The term "module" used in this document may include a unit implemented in hardware, software, or firmware, and may be used interchangeably with terms such as logic, logic block, component, or circuit, for example. A module may be an integrally constructed component or a minimal unit of components or a portion thereof that performs one or more functions. For example, according to one embodiment, the module may be implemented in the form of an application-specific integrated circuit (ASIC).

Various embodiments of this document are software that includes one or more instructions stored in a storage medium (eg, memory 120) readable by a machine (eg, political tendency analysis device 101). can be implemented For example, a processor (eg, the processor 110) of a device (eg, the political tendency analysis device 101) may call at least one command among one or more instructions stored from a storage medium and execute it. This enables the device to be operated to perform at least one function according to the at least one command invoked. The one or more instructions may include code generated by a compiler or code executable by an interpreter. The device-readable storage medium may be provided in the form of a non-transitory storage medium. Here, 'non-temporary' only means that the storage medium is a tangible device and does not contain a signal (e.g. electromagnetic wave), and this term refers to the case where data is stored semi-permanently in the storage medium. It does not discriminate when it is temporarily stored.

According to one embodiment, the method according to various embodiments disclosed in this document may be provided by being included in a computer program product. Computer program products may be traded between sellers and buyers as commodities. A computer program product is distributed in the form of a device-readable storage medium (eg compact disc read only memory (CD-ROM)), or through an application store (eg Play Store ^™ ) or on two user devices ( It can be distributed (eg downloaded or uploaded) online, directly between smartphones. In the case of online distribution, at least part of the computer program product may be temporarily stored or temporarily created in a device-readable storage medium such as a manufacturer's server, an application store server, or a relay server's memory.

According to various embodiments, each component (eg, module or program) of the components described above may include a singular entity or a plurality of entities. According to various embodiments, one or more components or operations among the aforementioned corresponding components may be omitted, or one or more other components or operations may be added. Alternatively or additionally, a plurality of components (eg modules or programs) may be integrated into a single component. In this case, the integrated component may perform one or more functions of each of the plurality of components identically or similarly to those performed by a corresponding component of the plurality of components prior to the integration. . According to various embodiments, the actions performed by a module, program, or other component are executed sequentially, in parallel, iteratively, or heuristically, or one or more of the actions are executed in a different order, or omitted. or one or more other actions may be added.

Claims (11)

In the political tendency analysis device,
processor; and
a memory operatively coupled to the processor;
The memory, when executed, causes the processor to:
collect information related to users;
identify whether the information contains keywords related to politics;
Based on the fact that the information includes a keyword related to the politics, identifying the information as evaluation target information,
Identify whether the evaluation target information includes keywords related to political orientation,
Storing instructions for assigning scores to the evaluation target information based on the fact that the evaluation target information includes a keyword related to the political orientation
Political orientation analysis device.
According to claim 1,
The instructions, the processor,
Based on the fact that the information is information related to the survey, the question of the survey and the answer data to the question are provided as one survey information;
A political propensity analysis device for identifying whether the survey information includes a keyword related to the politics.
According to claim 1,
The instructions, the processor,
To collect said information, provide test questions;
Receive answer data to the test question;
Providing the test question and the answer data for the test question as one test result information;
A political propensity analysis device for identifying whether the test result information includes a keyword related to the politics.
According to claim 1,
The instructions, the processor,
Using a pre-learned deep learning model, it is identified whether the evaluation target information includes a keyword related to the political orientation, and the evaluation target information is based on the fact that the evaluation target information includes a keyword related to the political orientation. A political tendency analysis device that gives scores to.
According to claim 4,
The deep learning model,
An input layer having, as an input node, keywords related to the evaluation target information and a political orientation determined with respect to the evaluation target information;
An output layer having a score assigned to the evaluation target information as an output node;
Including one or more hidden layers disposed between the input layer and the output layer;
A political tendency analysis device in which weights of nodes and edges between the input node and the output node are updated by a training process of the deep learning model.
According to claim 1,
The instructions, the processor,
Receive local information of the user;
A political tendency analysis device for assigning an additional score to the evaluation target information based on the user's regional information.
In the service providing method using the political tendency analysis device,
collecting information related to the user;
identifying whether the information includes a keyword related to politics;
identifying the information as evaluation target information based on the fact that the information includes a keyword related to the politics;
identifying whether a keyword related to political orientation is included in the evaluation target information; and
Based on the fact that the evaluation target information includes a keyword related to the political orientation, an operation of assigning a score to the evaluation target information
How We Provide Our Services.
According to claim 7,
providing a question of the survey and answer data to the question as one survey information, based on the fact that the information is information related to the survey; and
The service providing method further comprising an operation of identifying whether the survey information includes a keyword related to the politics.
According to claim 7,
providing test questions to collect the information;
receiving answer data for the test question;
providing the test question and the answer data for the test question as one piece of test result information; and
The service providing method further comprising an operation of identifying whether the test result information includes a keyword related to the politics.
According to claim 7,
The operation of identifying whether the evaluation target information includes a keyword related to the political orientation, and the operation of assigning a score to the evaluation target information based on whether the evaluation target information includes a keyword related to the political orientation, A method of providing services based on deep learning technology.
According to claim 7,
Further comprising receiving the user's area information;
The operation of assigning a score to the evaluation target information,
and assigning the score to the evaluation target information based on the user's region information.

Images