Information processing system, information processing method and program

Abstract

An information processing system and the like are provided that presents a test including questions that evaluate a user's characteristics using a pair of linguistic expressions. [Solution] According to one aspect of the present invention, an information processing system is provided that includes at least one processor, and the processor is configured to execute the following steps by reading a program: in the extraction step, a pair of linguistic expressions is extracted based on the similarity of emotional valence, and each of the pair of linguistic expressions corresponds to any one or more of a predetermined number of characteristics, thereby the pair of linguistic expressions corresponds to a set of characteristics, the emotional valence is a numerical representation of the emotional tendency of each of the pair of linguistic expressions, and in the presentation step, an information processing system is provided that presents a test including questions that evaluate the user's characteristics using the pair of linguistic expressions.

Description

[Technical Field]

[0001] The present invention relates to an information processing system, an information processing method, and a program. [Background technology]

[0002] Patent Document 1 is a document related to providing a counseling device that allows a respondent to freely express his or her own feelings, simplifies the aggregation and analysis of these feelings, and is capable of grasping the respondent's subjective feelings and stress. The counseling device is configured with a questionnaire creation unit 25 that creates a questionnaire by mixing a first question disclosed in Patent Document 1, a second question about images evoked by the answer to the first question, and a third question about subjective feelings about the answer to the first question, a communication control unit 29 that collects responses to the created questionnaire, a DB 21 that records answers to the third question among the collected questionnaire answers over multiple days, an answer analysis unit 26 that analyzes the answers to the third question based on the recorded answer results, a display data generation unit 27 that displays the obtained analysis results, and a display 28. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2008-293369 Summary of the Invention [Problem to be solved by the invention]

[0004] However, there is still room for technical improvement in information processing systems and the like that present tests including questions that evaluate user characteristics using pairs of linguistic expressions.

[0005] In view of the above circumstances, the present invention provides an information processing system and the like that presents a test including questions that evaluate a user's characteristics using a pair of linguistic expressions. [Means for solving the problem]

[0006] According to one aspect of the present invention, there is provided an information processing system comprising at least one processor, the processor being configured to execute the following steps by reading a program: in the extraction step, a pair of linguistic expressions is extracted based on the similarity of emotional valence, each of the pair of linguistic expressions corresponding to any one or more of a plurality of predetermined characteristics, thereby the pair of linguistic expressions corresponding to a set of characteristics, the emotional valence being a numerical representation of the emotional tendency of each of the pair of linguistic expressions, and in the presentation step, a test including questions for evaluating the user's characteristics using the pair of linguistic expressions is presented.

[0007] According to this aspect, it is possible to provide an information processing system or the like that presents a test including questions that evaluate a user's characteristics using a pair of linguistic expressions.

[0008] Hereinafter, embodiments of the present invention will be described. Note that various features shown in the following embodiments can be combined with each other. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a configuration diagram illustrating an information processing system 1. FIG. [Figure 2] FIG. 2 is a block diagram showing the hardware configuration of the server 2. [Figure 3] 2 is a block diagram showing the hardware configuration of a user terminal 3 and an administrator terminal 4. FIG. [Figure 4] FIG. 10 is a diagram showing question Q5, which is an example of a question presented by the information processing system 1. [Figure 5] 2 is a flowchart showing an outline of processing executed by the information processing system 1. [Figure 6] 2 is an activity diagram showing a specific example of processing executed by the information processing system 1. FIG. [Figure 7]10 is a diagram showing a list L6 that is an example of a list of candidates for linguistic expressions generated by the information processing system 1. FIG. [Figure 8] 10 is a diagram showing a list L7 that is an example of test questions 1 to 4 presented by the information processing system 1. FIG. DETAILED DESCRIPTION OF THE INVENTION

[0010] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to the accompanying drawings. Various features shown in the following embodiments can be combined with each other.

[0011] Incidentally, the program for realizing the software appearing in one embodiment may be provided as a non-transitory computer-readable medium, or may be provided so that it can be downloaded from an external server, or may be provided so that the program is started on an external computer and its functions are realized on a client terminal (so-called cloud computing).

[0012] Furthermore, various information processing according to an embodiment may realize input and output corresponding to the input. Here, the form of information referenced in such information processing (hereinafter referred to as reference information) is not limited as long as an output is obtained as a result of the input. The reference information may be, for example, rule-based information such as a database, a lookup table, or a predetermined function (including a decision formula such as a regression formula constructed using a statistical method), a trained model that has previously trained the correlation between input and output, or a large-scale language model that can output a desired result by inputting a prompt.

[0013] In one embodiment, a "unit" may include, for example, a combination of hardware resources implemented by a circuit in the broad sense and software information processing that can be specifically realized by these hardware resources. In one embodiment, various information is handled, and this information is represented, for example, by physical values ​​of signal values ​​representing voltage and current, high and low signal values ​​as a binary bit set consisting of 0 or 1, or quantum superposition (so-called quantum bits), and communication and calculations can be performed on a circuit in the broad sense.

[0014] Furthermore, a circuit in the broad sense is a circuit realized by at least an appropriate combination of a circuit, circuitry, processor, memory, etc. The processor may be a general-purpose processor or a dedicated circuit. That is, it includes an application specific integrated circuit (ASIC), a programmable logic device (e.g., a simple programmable logic device (SPLD), a complex programmable logic device (CPLD), and a field programmable gate array (FPGA)), etc.

[0015] 1. Hardware Configuration This section explains the hardware configuration.

[0016] <Information Processing System 1> FIG. 1 is a configuration diagram illustrating an information processing system 1. The information processing system 1 includes a server 2, a user terminal 3, and an administrator terminal 4. The server 2, the user terminal 3, and the administrator terminal 4 are configured to be able to communicate with each other via a telecommunications line. Here, the system exemplified as the information processing system 1 is composed of one or more devices or components. Therefore, it should be noted that the information processing system 1 includes either the server 2 alone or a combination of at least two of the server 2, the user terminal 3, and the administrator terminal 4. More specifically, the information processing system 1 may include elements selected from the group consisting of the server 2, the user terminal 3, and the administrator terminal 4. The elements not selected may not be included in the information processing system 1, but may be electrically connected to the selected elements as external elements. These components are described below.

[0017] <Server 2> 2 is a block diagram showing the hardware configuration of server 2. Server 2 includes a communication unit 21, a storage unit 22, and a control unit 23, and these components are electrically connected via a communication bus 20 inside server 2. Each component will be further described below.

[0018] The communication unit 21 is preferably a wired communication means such as USB, IEEE1394, Thunderbolt (registered trademark), wired LAN network communication, etc., but may also include wireless LAN network communication, mobile communication such as 3G / LTE / 5G, BLUETOOTH (registered trademark) communication, etc. as needed. In other words, it is more preferable to implement it as a collection of multiple communication means. In other words, the server 2 may communicate various information from the outside via the communication unit 21 and the network.

[0019] The memory unit 22 stores various pieces of information defined above. This can be implemented, for example, as a storage device such as a solid state drive (SSD) that stores various programs and the like related to the server 2 executed by the control unit 23, or as a memory such as a random access memory (RAM) that stores temporarily required information (arguments, arrays, etc.) related to the program operations. The memory unit 22 stores various programs, variables, etc. related to the server 2 executed by the control unit 23.

[0020] The control unit 23 processes and controls the overall operations related to the server 2. The control unit 23 is, for example, a central processing unit (CPU) not shown. The control unit 23 realizes various functions related to the server 2 by reading out predetermined programs stored in the storage unit 22. In other words, information processing by software stored in the storage unit 22 is specifically realized by the control unit 23, which is an example of hardware, and each step related to each function described below can be executed. This will be described in further detail in the next section. Note that the control unit 23 is not limited to being single, and multiple control units 23 may be provided for each function. Alternatively, a combination of these may be used.

[0021] <User terminal 3, administrator terminal 4> 3 is a block diagram showing the hardware configuration of the user terminal 3 and the administrator terminal 4. The user terminal 3 includes a communication unit 31, a storage unit 32, a control unit 33, a display unit 34, and an input unit 35, and these components are electrically connected within the user terminal 3 via a communication bus 30. Descriptions of the communication unit 31, the storage unit 32, and the control unit 33 are omitted here because they are similar to the descriptions of the respective units in the server 2. The administrator terminal 4 includes a communication unit 41, a storage unit 42, a control unit 43, a display unit 44, and an input unit 45, and these components are electrically connected within the administrator terminal 4 via a communication bus 40. Descriptions of the respective units in the administrator terminal 4 are omitted here because they are similar to the descriptions of the respective units in the server 2 and the user terminal 3.

[0022] The display unit 34 may be included in the housing of the user terminal 3 or may be externally attached. The display unit 34 displays a graphical user interface (GUI) screen that can be operated by the user. This is preferably implemented by selectively using display devices such as a CRT display, a liquid crystal display, an organic EL display, or a plasma display depending on the type of user terminal 3.

[0023] The input unit 35 may be included in the housing of the user terminal 3, or may be externally attached. For example, the input unit 35 may be implemented as a touch panel integrated with the display unit 34. A touch panel allows the user to input tapping, swiping, and the like. Of course, a switch button, a mouse, a QWERTY keyboard, or the like may be used instead of a touch panel. That is, the input unit 35 accepts an operation input made by the user. The input is transferred as a command signal to the control unit 33 via the communication bus 30, and the control unit 33 can execute predetermined control or calculation as necessary.

[0024] 2. Functional configuration of Server 2 The control unit 23 is configured to execute an extraction step, a presentation step, a generation step, an analysis step, a reception step, and a display control step.

[0025] The control unit 23 is configured to be able to extract various pieces of information by referring to reference information stored in the storage unit 22 of the server 2 or based on information generated by the server 2, as an extraction step. For example, the control unit 23 can extract a pair of linguistic expressions based on the similarity of valence, as an extraction step. The control unit 23 can extract multiple pairs of linguistic expressions based on the similarity of valence and predetermined conditions, as an extraction step.

[0026] The control unit 23 is configured to be able to present various information on the display unit 34 of the user terminal 3 and the display unit 44 of the administrator terminal 4 as a presentation step. The control unit 23 may present visual information such as a generated screen, an image including a still image or a video, an icon, a message, etc. on the display unit 34 of the user terminal 3 as a display control step.

[0027] The control unit 23 is configured to be able to generate various information by referring to the reference information stored in the storage unit 22 of the server 2 as a generation step. For example, the control unit 23 can generate a plurality of linguistic expression candidates corresponding to each characteristic as a generation step. Furthermore, the control unit 23 can generate a test design that satisfies predetermined conditions based on the reference information as a generation step.

[0028] The control unit 23 is configured to be able to refer to the reference information stored in the storage unit 22 of the server 2 and analyze the user's answer to the question as an analysis step.

[0029] The control unit 23 is configured to be able to acquire information from the user terminal 3, the administrator terminal 4, or other information processing terminals as an acquisition step. Also, the control unit 23 is configured to be able to acquire various pieces of information by reading out various pieces of information stored in a storage area that is at least a part of the memory unit 22 and writing the read information into a working area that is at least a part of the memory unit 22 as an acquisition step. The storage area is, for example, an area of ​​the memory unit 22 that is implemented as a storage device such as an SSD. The working area is, for example, an area that is implemented as a memory such as a RAM.

[0030] As a display control step, the control unit 23 executes a process for displaying a system screen related to the information processing system 1 on each terminal. As a display control step, the control unit 23 performs processes such as generating and transmitting an HTML (Hyper Text Markup Language) file, and displays a web page showing the system screen on the display unit 34 of the user terminal 3 or the display unit 44 of the administrator terminal 4. Note that the control unit 23 may also perform processes such as generating and transmitting display data for an application for using the information processing system 1 as a display control step. Specifically, as a display control step, the control unit 23 controls the display unit 34 of the user terminal 3 or the display unit 44 of the administrator terminal 4 to display visual information such as screens, images including still images or moving images, icons, and messages. As a display control step, the control unit 23 may also generate only rendering information for displaying the visual information on the display unit 34 of the user terminal 3 or the display unit 44 of the administrator terminal 4.

[0031] 3. Information processing flow This section describes the flow of an information processing method executed by the information processing system 1. As shown below, the information processing method includes steps executed by the information processing system. The information processing program of this embodiment causes a computer to execute each step of the information processing system. The order of the processes can be changed as appropriate, multiple processes can be executed simultaneously, or some processes can be omitted.

[0032] 3.1 Overview The information processing system 1 presents a test including questions that evaluate a user's characteristics using a pair of linguistic expressions. Tests including questions that evaluate a user's characteristics include psychological tests that measure non-cognitive abilities, occupational aptitude, stress, personality, etc. In psychological tests, the Likert scale is often used to evaluate the strength of a specific characteristic relative to other characteristics. However, the Likert scale's response format is susceptible to various response biases. A particular problem is the bias that users may give false responses based on social desirability, making themselves appear more desirable than they actually are. Therefore, comparative measurements that control for social desirability are used. Comparative measurements enable psychological tests that are robust against response bias by analyzing the response data using a statistical model. However, creating a test including questions for comparative measurements that control for social desirability is not easy. Therefore, an information processing system that can easily present comparative tests that control for social desirability is desired.

[0033] FIG. 4 illustrates question Q5, an example of a question presented by the information processing system 1. Question Q5 is an example of a comparative measurement question. Question Q5 includes an area Q51, an item element Q52, an item element Q53, and objects Q54 to Q59. Area Q51 is an area for writing instructions to the user, and states, "Which of these applies to you, and to what extent?" Item element Q52 states, "I prefer quiet," and item element Q53 states, "Once I start worrying, I can't stop." Item element Q52 and item element Q53 are linguistic expressions corresponding to different predetermined characteristics. Item element Q52 and item element Q53 are linguistic expressions with similar emotional valence. Item element Q52 and item element Q53 are a pair of linguistic expressions with similar social desirability. Comparing a pair of linguistic expressions with similar social desirability reduces the bias of selecting the one considered socially desirable, enabling a correct evaluation of the user's characteristics.

[0034] Objects Q54 to Q59 are objects with text indicating degrees of difficulty and are presented to the user in a selectable manner. Object Q54 is written as "very," object Q55 is written as "very," object Q56 is written as "somewhat," object Q57 is written as "somewhat," object Q58 is written as "very," and object Q59 is written as "very." The user follows the instructions written in area Q51 to perform an operation on one of the objects. Objects Q54 to Q59 are symmetrical options and do not include a "neither" option. This configuration allows the user to select an option that fits either item element Q52 or item element Q53. By not providing the user with a neutral option, a clearer opinion can be elicited from the user.

[0035] Question Q5 is an example of a question presented by the information processing system 1, and the display format of the question is not limited to this. For example, the instructions to the user may be worded differently from those in area Q51, and the format of the selectable objects may be different from those of objects Q54 to Q59. Furthermore, the number of selectable objects is not limited to six. It is preferable that the number of selectable objects is two or more and less than ten. If the number of selectable objects is odd, an intermediate option such as "neither" may be provided.

[0036] 5 is a flowchart showing an outline of processing executed by the information processing system 1. In this processing, first, in an extraction step, the control unit 23 extracts a pair of linguistic expressions based on the similarity of valence (step S001). Next, in a presentation step, the control unit 23 presents a test including questions for evaluating the characteristics of the user using the pair of linguistic expressions (step S002). Note that each of the pair of linguistic expressions corresponds to one or more of a plurality of predetermined characteristics, and thus the pair of linguistic expressions corresponds to a set of characteristics. The valence is a numerical representation of the tendency related to the emotion of each of the pair of linguistic expressions.

[0037] In summary, an information processing system according to one embodiment includes at least one processor. The processor reads a program to perform the following steps: In the extraction step, the control unit 23 extracts a pair of linguistic expressions based on the similarity of valence. Each of the pair of linguistic expressions corresponds to one or more of a plurality of predetermined characteristics, thereby making the pair of linguistic expressions correspond to a set of characteristics. The valence is a numerical representation of the emotional tendency of each of the pair of linguistic expressions. In the presentation step, the control unit 23 presents a test including questions that evaluate the user's characteristics using the pair of linguistic expressions. In this manner, a test including questions that evaluate the user's characteristics using the pair of linguistic expressions can be presented to the user who answers the question.

[0038] 3.2 Specific examples FIG. 6 is an activity diagram showing a specific example of processing executed by the information processing system 1. The specific example may fall within the scope defined in the overview above. The following description will be given along with each activity in this activity diagram. Note that the information processing may include any exception handling not shown. Exception handling includes the interruption of the information processing or the omission of each process. The selection or input performed in the information processing may be based on a user operation or may be performed automatically without relying on a user operation.

[0039] First, the control unit 23 of the server 2 can receive an instruction via the administrator terminal 4 to determine multiple characteristics to be measured (activity A001). Here, the characteristics to be measured are preferably, for example, psychological dimensions used to understand an individual's psychological characteristics. Examples of psychological dimensions include personality dimensions, emotional dimensions, cognitive dimensions, motivational dimensions, social dimensions, values ​​and belief dimensions, behavioral dimensions, and mental health dimensions. Here, an example will be described using the five factors of the Big Five personality traits, which are one of the personality dimensions. The five factors of the Big Five personality traits are Agreeableness (A), Conscientiousness (C), Extraversion (E), Openness (O), and Neuroticism (N).

[0040] Next, the control unit 23 can acquire known linguistic expressions corresponding to each characteristic (activity A002). The control unit 23 may acquire known linguistic expressions corresponding to each characteristic by reading them out from the storage unit 22 in advance, or may acquire linguistic expressions that are already known to correspond to the characteristic based on literature information, etc. The control unit 23 may acquire known linguistic expressions corresponding to each characteristic via the administrator terminal 4.

[0041] Next, as a generation step, the control unit 23 can generate candidates for linguistic expressions corresponding to the characteristics (activity A003). The control unit 23 can generate candidates for linguistic expressions corresponding to the characteristics using a natural language processing model. The control unit 23 may generate multiple candidates for linguistic expressions corresponding to each characteristic. The control unit 23 may display the generated linguistic expressions on the display unit 44 of the administrator terminal 4 and allow the administrator user to confirm their semantic validity. In this case, the control unit 23 may proceed to the next step when it receives, via the administrator terminal 4, an operation instruction confirming that there is no problem with the semantic validity.

[0042] Here, in the generation step, the control unit 23 may generate candidate linguistic expressions that have a high semantic similarity to known linguistic expressions based on known linguistic expressions corresponding to each characteristic. For example, if "interesting" is a known linguistic expression that measures "extroversion," the control unit 23 can generate candidate linguistic expressions that have a high semantic similarity to "interesting." In this manner, candidate linguistic expressions corresponding to characteristics can be generated based on known linguistic expressions.

[0043] The control unit 23 can generate candidate linguistic expressions that have high semantic similarity to known linguistic expressions by, for example, representing words as vectors and calculating similarity based on the angle between the vectors. One method for calculating similarity based on the angle between vectors is to use cosine similarity. The control unit 23 can generate candidate linguistic expressions based on known linguistic expressions by using static embeddings such as Word2Vec, GloVe, and FastText, contextualized embeddings such as BERT (Bidirectional Encoder Representations from Transformers), or embedding matrices within neural models. For example, the control unit 23 can use pre-trained Word2Vec to generate candidate linguistic expressions with high semantic similarity, such as "cheerful" and "talkative," which have high similarity to "interesting." Note that the control unit 23 may perform preprocessing, such as affine transformation (centring, whitening, etc.) or preprocessing not limited to affine transformation, on the embedding of the linguistic expressions to improve performance in subsequent tasks.

[0044] The control unit 23 may generate candidates for linguistic expressions corresponding to the characteristics using AI (Artificial Intelligence) equipped with language models such as Transformers including GPT (Generative Pretrained Transformer, including GPT-1, GPT-2, GPT-3, GPT-3.5, GPT-4, etc.), BERT (Bidirectional Encoder Representations from Transformers), BART (Bidirectional and Auto-regressive Transformer), etc., and Recurrent Neural Networks (RNNs).

[0045] Here, in the generation step of activity A003, when generating candidates for linguistic expressions corresponding to a trait, the control unit 23 can also generate candidates for linguistic expressions of inverse items corresponding to the trait. Inverse items are linguistic expressions that have the opposite meaning to known linguistic expressions corresponding to the trait. For example, for the trait of extroversion, "shy" is an expression that is the opposite of extroversion, but can be used to evaluate the trait of "extroversion" in a test that includes questions that evaluate a user's traits using a pair of linguistic expressions. For example, the control unit 23 can generate "shy" as the inverse item of "extroversion" using pre-trained Word2Vec.

[0046] FIG. 7 is a diagram showing a list L6, which is an example of a list of candidate linguistic expressions generated by the information processing system 1. List L6 lists the corresponding traits, calculated valences, and the direction of the traits and linguistic expressions for each of the candidate linguistic expressions G1 to G12. The candidate linguistic expressions extracted include "simple," "strict," "aggressive," "unstable," "serious," "enthusiastic," "easily nervous," "gloomy," "easily shaken," "demure," "cooperative," and "loyal." Each linguistic expression corresponds to one of a plurality of predetermined traits. In the example of FIG. 7, the predetermined traits are conscientiousness (C), openness (O), extraversion (E), neuroticism (N), and agreeableness (A). "Simple" corresponds to conscientiousness, "strict" and "cooperative" correspond to openness, "aggressive," "gloomy," and "plain" correspond to extroversion, "unstable," "easily nervous," and "easily agitated" correspond to neuroticism, and "serious," "enthusiastic," and "loyal" correspond to agreeableness. In List L6, the valence column contains the calculated valence values ​​for each candidate linguistic expression. The direction column for "simple," "unstable," "gloomy," and "plain" also contains the word "reverse." A linguistic expression with "reverse" in the direction column indicates that it is a reversed item for the corresponding trait. For example, "gloomy" and "plain" are linguistic expressions with a direction opposite to extroversion, and such reversed items can also be extracted as candidates for linguistic expressions corresponding to traits.

[0047] Note that linguistic expressions generally refer to expressions related to language structures and linguistic techniques. While FIG. 7 shows a word as an example of a linguistic expression candidate, the linguistic expression candidate may be not only a word but also a phrase combining multiple words, a sentence combining words and phrases, or a paragraph combining multiple sentences. That is, the linguistic expression candidate may be a word, phrase, sentence, or paragraph. For example, FIG. 4 includes a pair of linguistic expressions, "It's better to be quiet" and "Once I start worrying, I can't stop." This type of configuration allows for the creation of a wide variety of questions.

[0048] When generating linguistic expressions in units larger than words as candidates for linguistic expressions, the control unit 23 may use various methods, such as methods based on sentence embeddings, to calculate the semantic similarity with known linguistic expressions, but as a simple method, for example, the average of the vectors of the words that make up the unit can be used.

[0049] Next, the control unit 23 can calculate the emotional valence of each candidate linguistic expression (activity A004).

[0050] Here, affective valence is a numerical representation of the emotional tendency of each of a pair of linguistic expressions. The emotional tendency of each of a pair of linguistic expressions includes the positive / negative direction of the emotion, the degree of positivity / negativity, the degree of social desirability, etc.

[0051] The control unit 23 can calculate the valence of each candidate linguistic expression by latent semantic scaling. For example, the control unit 23 can calculate the valence of each linguistic expression by sentiment analysis based on positive and negative sentiments based on pre-trained Word2Vec. Pre-trained Word2Vec is a model that acquires distributed representations of linguistic expressions by learning text data and can quantify the meaning of the linguistic expressions. By using such pre-trained Word2Vec, the control unit 23 can calculate the valence of each linguistic expression based on the similarity between each linguistic expression and reference positive and negative linguistic expressions.

[0052] Furthermore, the control unit 23 may calculate the emotional valence of each candidate linguistic expression using a language model, etc. For example, the control unit 23 may use a language model to calculate the positive / negative direction of the emotion, the degree of positivity / negativeness of the emotion, the degree of social desirability, etc., of each candidate linguistic expression.

[0053] Next, the control unit 23 generates a test design that satisfies predetermined conditions (activity A005). The predetermined conditions include, for example, that the number of linguistic expressions corresponding to each characteristic is (approximately) the same, that the number of the set of characteristics in all patterns is (approximately) the same, that the number of questions using the same directional key and the mixed directional key is (approximately) the same, and that each characteristic is measured at least twice using the mixed directional key items. By satisfying these conditions, highly accurate evaluation can be performed. Note that this test design corresponds to scale design in psychological measurement.

[0054] FIG. 8 is a diagram showing a list L7, which is an example of questions 1 to 4 of a test presented by the information processing system 1. List L7 is a list of corresponding traits, linguistic expressions, and valences of pairs of linguistic expressions in questions 1 to 4. In question 1, "simple" and "generous" are a pair of linguistic expressions. "Simple" corresponds to the trait of conscientiousness, and "generous" corresponds to the trait of agreeableness. The valence of "simple" is 0.56, and the valence of "generous" is 0.57. In question 2, "funny" and "reasonable" are a pair of linguistic expressions. "Funny" corresponds to the trait of extroversion, and "reasonable" corresponds to the trait of conscientiousness. The valence of "funny" is 1.44, and the valence of "reasonable" is 1.46. In question 3, "easily upset" and "temperate" are a pair of linguistic expressions. "Easily shaken" corresponds to the trait of neuroticism, and "hot temper" corresponds to the trait of "openness." The valence of "easily shaken" is -1.74, and the valence of "hot temper" is -1.71. In question 4, "unscrupulous" and "shy" are a pair of linguistic expressions. "Unscrupulous" corresponds to the trait of conscientiousness, and "shy" corresponds to the trait of "extroversion." The valence of "unscrupulous" is 0.47, and the valence of "shy" is 0.43.

[0055] In the extraction step, the control unit 23 may extract multiple pairs of linguistic expressions so that the number of linguistic expressions corresponding to each trait is the same in the test. For example, in questions 1 to 4 of list L7, there is one linguistic expression for agreeableness, three linguistic expressions for conscientiousness, two linguistic expressions for extraversion, one linguistic expression for openness, and one linguistic expression for neuroticism. The control unit 23 may extract multiple pairs of linguistic expressions so that the number of linguistic expressions corresponding to each of these traits is approximately the same. In this manner, a test can be created that includes linguistic expressions corresponding to all traits in a balanced manner. Furthermore, the more the number of linguistic expressions corresponding to each trait is the same or close to the same, the more accurate the measurement.

[0056] In the extraction step, the control unit 23 may extract multiple pairs of linguistic expressions so that the number of pairs of traits in all patterns is the same. For example, in questions 1 to 4 of list L7, the pairs of traits are one pair of conscientiousness and agreeableness, one pair of extraversion and conscientiousness, one pair of neuroticism and openness, and one pair of conscientiousness and extraversion. If there are five traits to be measured, there are 10 pairs of traits. The control unit 23 may extract multiple pairs of linguistic expressions so that the number of questions corresponding to each pair of traits is (approximately) the same. This allows for the creation of a test that includes a balanced combination of traits in all patterns. Furthermore, the more the number of questions corresponding to each pair of traits is the same or close to the same, the more accurate the measurement.

[0057] In the extraction step, the control unit 23 may extract multiple pairs of linguistic expressions so that the number of questions with the same directional key is equal to the number of questions with mixed directional key. The same directional key designates a design in which all linguistic expressions included in the questions are scored in the same direction. That is, in the case of the same directional key design, a pair of linguistic expressions is scored in the same direction. On the other hand, the mixed directional key designates a design in which the linguistic expressions included in the questions are scored in different directions. That is, in the case of the mixed directional key design, one of the pair of linguistic expressions is scored in a positive direction and the other is scored in a negative direction. For example, in questions 1 to 4 of list L7, question 1 is a mixed directional key design because only "simple" is a reversed item. Question 2 is a same directional key design in which both "interesting" and "reasonable" are scored in a positive direction. Question 3 is a mixed directional key design because only "temperate" is a reversed item. Question 4 is a same directional key design because both "unprincipled" and "shy" are reversed items. That is, in questions 1 to 4 of list L7, the number of questions using the same directional key and the number of questions using mixed directional key are the same. The control unit 23 may extract a plurality of pairs of linguistic expressions so that the number of questions using the same directional key and the number of questions using mixed directional key are approximately the same. This configuration makes it possible to create a test in which some options are provided in a positive direction and some in a negative direction. This has the effect of encouraging the user to answer carefully and reducing bias in the user, resulting in more accurate evaluation. Furthermore, the measurement accuracy improves as the number of questions using the same directional key and the number of questions using mixed directional key are equal or close to equal.

[0058] In the extraction step, the control unit 23 may extract a plurality of pairs of linguistic expressions so that each trait is included at least once in the mixed-direction key questions. For example, among questions 1 to 4 in list L7, questions 1 and 3 are mixed-direction key questions. Therefore, it can be said that openness is included twice, agreeableness once, and neuroticism once in the mixed-direction key questions. It is more preferable that the control unit 23 extracts a plurality of pairs of linguistic expressions so that each trait is measured at least twice in the mixed-direction key items. This embodiment makes it possible to create a test in which some options are provided in a positive direction and some in a negative direction. This has the effect of encouraging users to respond carefully and reducing bias in users responding, thereby enabling more accurate evaluation.

[0059] It is preferable that the control unit 23 extracts a plurality of pairs of linguistic expressions so as to satisfy all predetermined conditions. For example, in the extraction step, it is preferable that the control unit 23 extracts a plurality of pairs of linguistic expressions so that the number of linguistic expressions corresponding to each characteristic is the same, the number of sets of characteristics in all patterns is the same, the number of problems using the same directional key and the number of problems using mixed directional keys are the same, and each characteristic is included in problems using mixed directional keys at least twice.

[0060] Next, in the extraction step, the control unit 23 extracts a pair of linguistic expressions based on the similarity in valence or the like (activity A006). For example, in the extraction step, the control unit 23 can extract a pair of linguistic expressions from a plurality of candidates based on the similarity in valence calculated for each of the plurality of candidates. The control unit 23 can extract a pair of linguistic expressions based on the test design generated in activity A005 using an optimization algorithm that aims to minimize the difference in valence, with the difference in valence as the objective function. The optimization algorithm here includes mathematical optimization algorithms such as rule-based combinatorial optimization algorithms and simulated annealing.

[0061] The control unit 23 repeats the processing of activity A006 until it determines in activity A005 that extraction of multiple pairs of linguistic expressions corresponding to the generated test design has been completed. When the control unit 23 determines that extraction of multiple pairs of linguistic expressions corresponding to the generated test design has been completed, the control unit 23 proceeds to the next step. Here, the control unit 23 may display the extracted pairs of linguistic expressions on the display unit 44 of the administrator terminal 4 to allow the administrator user to confirm their semantic validity. In this case, the control unit 23 may proceed to the next step when it receives an operation instruction via the administrator terminal 4 confirming that there is no problem with the semantic validity.

[0062] Next, the control unit 23 presents a question on the display unit 34 of the user terminal 3 (activity A007). For example, when question Q5 shown in Fig. 4 is presented, the user compares item element Q52 "It's better to be quiet" with item element Q53 "Once I start worrying, I can't stop", considers which of the two applies to him and to what extent, and gives an operation instruction to one of objects Q54 to Q59.

[0063] Next, when the user terminal 3 receives an operation instruction from the user, the control unit 23 acquires the user's answer via the network (activity A008).

[0064] Next, the control unit 23 analyzes the answers obtained from the user based on the reference information. Here, the control unit 23 can numerically evaluate the user's answers, such as "very," "very," "somewhat," "somewhat," "very," and "very." The control unit 23 can estimate respondent parameters using methods such as maximum likelihood estimation, Bayesian estimation, and the EM algorithm. For example, the control unit 23 preferably uses a Thirstonian Item Response Theory (TIRT) model or the like in analyzing the numerical values ​​obtained as the user's answers, treating the respondent's characteristic values ​​as parameters and statistically estimating them. Such respondent parameters can be estimated using maximum likelihood estimation or Bayesian estimation, more specifically, using a method such as the Expectation-Maximization (EM) algorithm. It is also preferable to estimate characteristic values ​​using a TIRT model extended to handle reaction time information.

[0065] The respondent parameters are numerical representations of the latent characteristics (e.g., personality traits, abilities, and attitudes) of the respondent user. The larger the value of the respondent parameter, the stronger the characteristic of the respondent user. However, it is difficult to tell from the respondent parameter alone whether the characteristic is stronger than that of other users. Therefore, when presenting the respondent parameter to the user, instead of the numerical value of the respondent parameter itself, the respondent parameter may be converted into a characteristic score of the respondent user, which indicates the respondent user's relative position within a group. The control unit 23 can calculate the characteristic score of the user based on the respondent parameters. The control unit 23 may store the respondent parameters of each user in the storage unit 22 and use them as reference information when calculating the characteristic score of the next respondent user.

[0066] Next, the control unit 23 presents the calculated characteristic scores to the user. In this manner, the characteristics of the user can be evaluated.

[0067] 5. Variations Furthermore, the following aspects may be adopted.

[0068] In the above embodiment, in activity A003, the control unit 23 can generate candidates for linguistic expressions corresponding to characteristics as a generation step. However, the control unit 23 may receive candidates for linguistic expressions from an administrator user via the administrator terminal 4. Furthermore, the control unit 23 may infer characteristics corresponding to candidates for linguistic expressions received via the administrator terminal 4 based on known linguistic expressions. The control unit 23 can infer characteristics corresponding to candidates for linguistic expressions based on semantic similarity with known linguistic expressions.

[0069] Although the linguistic expressions shown in the above embodiment are all in Japanese, the linguistic expressions corresponding to the characteristics may be in a language other than Japanese. That is, in activity A003, the control unit 23 can generate candidates for linguistic expressions corresponding to the characteristics in a language other than Japanese as a generation step. In activity A004, the control unit 23 can calculate the emotional valence of each candidate linguistic expression in a language other than Japanese.

[0070] In list L6, which is an example of a list of linguistic expression candidates generated by information processing system 1 shown in FIG. 7, one linguistic expression corresponds to any one of a plurality of predetermined characteristics. However, one linguistic expression may correspond to any two or more of a plurality of predetermined characteristics. For example, control unit 23 may extract one linguistic expression that simultaneously corresponds to two characteristics, extroversion and agreeableness, or one linguistic expression that simultaneously corresponds to two or more characteristics. Furthermore, control unit 23 may generate linguistic expression candidates corresponding to two or more characteristics in activity A003 as a generation step.

[0071] Furthermore, when one linguistic expression corresponds to two or more characteristics among a plurality of predetermined characteristics, the total number of characteristics corresponding to the pair of linguistic expressions is three or more. That is, one set of characteristics may be composed of three or more characteristics. In this case, it is preferable to extract multiple pairs of linguistic expressions so that the number of sets of characteristics for all patterns is approximately the same. In the extraction step, the control unit 23 preferably extracts multiple pairs of linguistic expressions so that each combination of characteristics corresponding to the pair of linguistic expressions is measured the same number of times in the test.

[0072] In the above embodiment, the control unit 23 acquires and analyzes the user's answers via the network, but the control unit 33 of the user terminal 3 may analyze the answers. For example, the analysis can be performed by downloading a program to the user terminal 3.

[0073] In the above embodiment, the control unit 23 performs an analysis based on the user's answers in activity A009, and then presents a characteristic score for the user who answered in activity A010. However, the control unit 23 may obtain answers to some of the questions (e.g., the first 5-10 questions), perform an analysis based on the answers, and then return to activity A005 to re-generate a test design based on the analysis results, thereby presenting questions tailored to the user. In other words, the control unit 23 performs an analysis of the user based on the user's answers to the questions in the analysis step. In the extraction step, the control unit 23 extracts multiple pairs of linguistic expressions tailored to the user based on the analysis results. In this manner, a test including questions tailored to the user who answers can be created.

[0074] The above-described information processing mode is merely an example, and the present invention is not limited to this, and can be modified as appropriate within the scope of the technical concept of the invention.

[0075] <Configuration variations> The configuration shown in FIG. 1 and other figures is merely an example, and other configurations are possible as long as they are not inconvenient for implementation. For example, one device may be distributed among two or more devices, or may be replaced by a cloud computing system. Furthermore, the functions of one device may be distributed among two or more devices, or the functions of two or more devices may be centralized in one device. Furthermore, the operation of one function may be distributed among two or more functions, or two or more functions may be integrated into one function. In short, as long as each function required by the entire information processing system 1 is realized, the devices that realize those functions may have any configuration. In particular, the artificial intelligence module may be external to the server 2. In this case, the external artificial intelligence module may be provided by, for example, an artificial intelligence service server, and configured to receive inputs from each functional unit of the server 2, receive requests to execute artificial intelligence services, and return the specified output as a processing result to the server 2. The artificial intelligence service server may provide services using a language model as a learning model, or may perform language processing tasks using a language model. The artificial intelligence service server may be constructed using a large language model. The artificial intelligence service server accepts prompt inputs such as text, images, and voice, and generates and responds with answers to the prompts.

[0076] <Other variations> The output destination of information or data (hereinafter referred to as "information, etc.") may be another device, a display, a memory unit (including an internal memory unit and an external memory unit), etc. Acquisition of information, etc. includes acquiring information, etc. transmitted from another device, as well as acquiring information, etc. generated by the device itself. The table associating parameters is not limited to the table shown in the figure, and the number of parameters may be reduced or increased. Furthermore, information, etc. corresponding to parameters may be obtained using a mathematical formula, a conditional formula, etc., without using a table.

[0077] <Additional Notes> Furthermore, it may be provided in the following aspects.

[0078] (1) An information processing system comprising at least one processor, the processor being configured to execute the following steps by reading a program: in the extraction step, a pair of linguistic expressions is extracted based on the similarity of emotional valence, wherein each of the pair of linguistic expressions corresponds to any one or more of a plurality of predetermined characteristics, thereby the pair of linguistic expressions corresponds to a set of characteristics; the emotional valence is a numerical representation of the emotional tendency of each of the pair of linguistic expressions; and in the presentation step, a test including questions that evaluate the user's characteristics using the pair of linguistic expressions is presented.

[0079] In this manner, a user responding to a test can be presented with questions that assess the user's characteristics using a pair of linguistic expressions that have similar emotional valence.

[0080] (2) In the information processing system described in (1) above, the generation step further generates a plurality of linguistic expression candidates corresponding to each of the characteristics, and the extraction step extracts the pair of linguistic expressions from the plurality of candidates based on the similarity of the emotional valence calculated from each of the plurality of candidates.

[0081] In this manner, a plurality of new linguistic expressions can be generated and a pair of linguistic expressions can be extracted.

[0082] (3) In the information processing system described in (2) above, in the generation step, linguistic expressions that have a high semantic similarity to known linguistic expressions corresponding to each of the characteristics are generated as candidates based on the known linguistic expressions.

[0083] In this manner, it is possible to generate candidates for linguistic expressions based on known linguistic expressions.

[0084] (4) In the information processing system described in (2) or (3) above, the candidate is a word, a phrase, a sentence, or a paragraph.

[0085] In this manner, linguistic expressions consisting of not only words but also phrases, sentences, and paragraphs are extracted, making it possible to create a wide variety of questions.

[0086] (5) In an information processing system described in any one of (1) to (4) above, in the extraction step, multiple pairs of linguistic expressions are extracted so that the number of linguistic expressions corresponding to each of the characteristics in the test is the same, and in the presentation step, the test including multiple questions is presented.

[0087] In this manner, a test can be created that includes a balanced range of linguistic expressions corresponding to all characteristics.

[0088] (6) In the information processing system described in any one of (1) to (5) above, in the extraction step, multiple pairs of linguistic expressions are extracted so that the number of pairs of characteristics in all patterns is the same, and in the presentation step, the test including multiple questions is presented.

[0089] In this manner, it is possible to create a test that includes a balanced combination of characteristics of all patterns.

[0090] (7) In the information processing system described in any one of (1) to (6) above, in the extraction step, a plurality of pairs of linguistic expressions are extracted so that the number of questions using the same directional key is the same as the number of questions using mixed directional keys, and in the presentation step, the test including a plurality of the questions is presented.

[0091] This allows for the creation of tests with options that are partly positive and partly negative, which has the effect of encouraging users to answer carefully and reducing bias in users, resulting in more accurate evaluations.

[0092] (8) In the information processing system described in any one of (1) to (7) above, in the extraction step, a plurality of pairs of linguistic expressions are extracted so that each of the characteristics is included at least once in the question of the mixed directional key, and in the presentation step, the test including a plurality of the questions is presented.

[0093] This allows for the creation of tests with options that are partly positive and partly negative, which has the effect of encouraging users to answer carefully and reducing bias in users, resulting in more accurate evaluations.

[0094] (9) An information processing system according to any one of (1) to (8) above, further comprising, in the analysis step, an analysis of the user based on the user's response to the question, and in the extraction step, extracting a plurality of pairs of linguistic expressions corresponding to the user based on the results of the analysis.

[0095] In this manner, a test can be created that includes questions that are tailored to the user answering them.

[0096] (10) An information processing method, comprising steps executed by the information processing system according to any one of (1) to (9) above.

[0097] (11) A program that causes a computer to execute each step of the information processing system described in any one of (1) to (9) above. Of course, this is not the case. Furthermore, the above-described embodiments and modifications may be combined in any desired manner.

[0098] Finally, while various embodiments of the present invention have been described, they are presented by way of example only and are not intended to limit the scope of the invention. The novel embodiments may be embodied in various other forms, and various omissions, substitutions, and modifications may be made without departing from the spirit of the invention. The embodiments and their modifications are intended to be included within the scope and spirit of the invention, as well as within the scope of the inventions and their equivalents as defined in the accompanying claims. [Explanation of symbols]

[0099] 1: Information processing system 2: Server 20: Communication bus 21: Communications Department 22: Storage section 23: Control section 3: User terminal 30: Communication bus 31: Communications Department 32: Storage section 33: Control section 34:Display section 35: Input section 4: Administrator terminal 40: Communication bus 41: Communications Department 42: Storage section 43: Control section 44: Display section 45: Input section L6: List L7: List Q5 :Problem Q51 :Area Q52:Item element Q53:Item element Q54: Object Q55: Object Q56: Object Q57: Object Q58: Object Q59: Object

Claims

1. An information processing system, at least one processor; The processor is configured to execute the following steps by reading the program: In the extraction step, a pair of linguistic expressions is extracted based on the similarity of emotional valence, where: each of the pair of linguistic expressions corresponds to one or more of a plurality of predetermined characteristics, whereby the pair of linguistic expressions corresponds to a set of characteristics; The emotional valence is a numerical representation of the tendency of each of the pair of linguistic expressions with respect to an emotion, In the presenting step, the information processing system presents a test including questions for evaluating the characteristics of the user using the pair of linguistic expressions.

2. 2. The information processing system according to claim 1, Furthermore, in the generating step, a plurality of linguistic expression candidates corresponding to each of the characteristics are generated; In the extraction step, the pair of linguistic expressions is extracted from the plurality of candidates based on the similarity of the emotional valence calculated from each of the plurality of candidates.

3. 3. The information processing system according to claim 2, In the generating step, an information processing system generates, as the candidates, linguistic expressions that have a high degree of semantic similarity to known linguistic expressions corresponding to each of the characteristics, based on the known linguistic expressions.

4. 3. The information processing system according to claim 2, The candidate is a word, a phrase, a sentence, or a paragraph.

5. 2. The information processing system according to claim 1, In the extraction step, a plurality of pairs of linguistic expressions are extracted so that the number of linguistic expressions corresponding to each of the characteristics is the same in the test; In the presenting step, the information processing system presents the test including a plurality of the questions.

6. 2. The information processing system according to claim 1, In the extraction step, a plurality of pairs of linguistic expressions are extracted so that the number of pairs of characteristics of all patterns is the same; In the presenting step, the information processing system presents the test including a plurality of the questions.

7. 2. The information processing system according to claim 1, In the extraction step, a plurality of pairs of linguistic expressions are extracted so that the number of questions using the same directional key is equal to the number of questions using the mixed directional key; In the presenting step, the information processing system presents the test including a plurality of the questions.

8. 2. The information processing system according to claim 1, In the extracting step, a plurality of pairs of linguistic expressions are extracted so that each of the characteristics is included in the question of a mixed directional key at least once; In the presenting step, the information processing system presents the test including a plurality of the questions.

9. 2. The information processing system according to claim 1, Furthermore, in the analysis step, an analysis is performed on the user based on the user's answer to the question, In the extraction step, a plurality of pairs of linguistic expressions corresponding to the user are extracted based on the results of the analysis.

10. An information processing method, comprising: An information processing method comprising the steps executed by the information processing system according to any one of claims 1 to 9.

11. A program, A program causing a computer to execute each step of the information processing system according to any one of claims 1 to 9.

Images