JP2019164314A - Estimation device, estimation method, and estimation program - Google Patents

Abstract

To improve user attribute estimation accuracy.SOLUTION: An estimation device comprises a generation unit, an acquisition unit, and an estimation unit. The generation unit generates a question relating to user attributes in interaction with a user. The acquisition unit acquires a user's answer to the question generated by the generation unit. The estimation unit estimates user attributes on the basis of the answer acquired by the acquisition unit.SELECTED DRAWING: Figure 2

Description

  The present invention relates to an estimation device, an estimation method, and an estimation program.

  Conventionally, for example, for the purpose of providing appropriate information to a user, an estimation device that estimates a user attribute such as a user's hobby preference is known. The estimation device estimates an attribute based on, for example, a user's behavior on the network (see, for example, Patent Document 1).

JP 2006-162399 A

  However, the conventional technology has room for improvement in terms of improving the accuracy of user attribute estimation.

  The present application has been made in view of the above, and an object thereof is to provide an estimation device, an estimation method, and an estimation program capable of improving the accuracy of estimation of user attributes.

  The estimation apparatus according to the present application includes a generation unit, an acquisition unit, and an estimation unit. The said production | generation part produces | generates the question regarding the said user's attribute in a dialog with a user. The acquisition unit acquires the user's answer to the question generated by the generation unit. The estimation unit estimates the attribute of the user based on the answer acquired by the acquisition unit.

  According to one aspect of the embodiment, there is an effect that it is possible to improve the estimation accuracy of the user's attribute.

FIG. 1 is a diagram illustrating an outline of an estimation method according to the embodiment. FIG. 2 is a diagram illustrating a configuration of the estimation device according to the embodiment. FIG. 3 is an explanatory diagram of attribute information. FIG. 4 is a diagram illustrating a specific example of processing of the estimation apparatus according to the embodiment. FIG. 5 is a diagram illustrating a specific example of processing of the estimation apparatus according to the embodiment. FIG. 6 is a diagram illustrating a specific example of processing of the estimation apparatus according to the embodiment. FIG. 7 is a flowchart illustrating a procedure of processing executed by the estimation apparatus according to the embodiment. FIG. 8 is a hardware configuration diagram illustrating an example of a computer that realizes the function of the estimation device according to the embodiment.

  Hereinafter, modes for carrying out the estimation apparatus, the estimation method, and the estimation program according to the present application (hereinafter referred to as “embodiments”) will be described in detail with reference to the drawings. In addition, the estimation apparatus, the estimation method, and the estimation program according to the present application are not limited by this embodiment. In the following embodiments, the same portions are denoted by the same reference numerals, and redundant description is omitted.

  First, the estimation method executed by the estimation apparatus according to the embodiment will be described with reference to FIG. FIG. 1 is a diagram illustrating an outline of an estimation method according to the embodiment. FIG. 1 shows an estimation system S including an estimation device 1 according to the embodiment.

  As shown in FIG. 1, the estimation system S according to the embodiment includes an estimation device 1 and an information providing terminal 100. The estimation device 1 and the information providing terminal 100 are connected via a communication network such as the Internet, for example. Although FIG. 1 shows a case where one information providing terminal 100 is connected to the estimating apparatus 1, a plurality of information providing terminals 100 may be connected to the estimating apparatus 1.

  The estimation device 1 is an information processing device that estimates the attributes of the user U. The attribute here is information that characterizes the user U, for example, internal information such as the hobby, taste, personality, etc. of the user U. Note that the attributes may further include, for example, external information such as the appearance of the user U, additional information such as a career, and the like.

  The estimation device 1 is also a providing server that provides various contents to the user U based on the estimated attribute of the user U. The estimation apparatus 1 as a providing server includes various applications installed in the information providing terminal 100 (for example, portal application, news application, auction site, weather forecast application, shopping application, finance (stock price) application, route search application, map provision) An app, a travel app, a restaurant introduction app, a blog browsing app, etc.). Alternatively, the estimation device 1 as a providing server includes a portal site, news site, auction site, weather forecast site, shopping site, finance (stock price) site, route search site, map providing site, travel site, restaurant introduction site, web It is a device that distributes web pages in which various information related to blogs and the like are arranged in a tile shape. In addition, the estimation device 1 may be a server that distributes application data itself, or may be a server that distributes web pages such as a portal site.

  As shown in FIG. 1, the estimation device 1 stores attribute information 41 that is information related to the attributes of the user U. In the example illustrated in FIG. 1, the attribute information 41 includes “food” as the attribute type and “sushi” and “ramen” as the attribute items. In the attribute information 41 shown in FIG. 1, “O” is attached to the attribute item “sushi”, which indicates that the user U likes sushi (positive). “X” indicates that the food is disliked (negative), and a blank indicates that the food has not been estimated yet.

  The information providing terminal 100 acquires various contents from the estimation device 1 serving as a providing server and provides it to the user U. For example, the information providing terminal 100 has a voice recognition function and provides various contents based on the recognition result of the user U's utterance. Note that the information providing terminal 100 is not limited to voice recognition, and may provide various contents by accepting an operation of the user U to a touch panel display unit, for example.

  Moreover, the information provision method by the information provision terminal 100 is provided by audio | voice output, for example. That is, the information providing terminal 100 is a so-called smart speaker. Note that the information providing terminal 100 is not limited to audio output, and may provide information by video output from, for example, a projector or a display unit.

  Moreover, the information provision terminal 100 is not limited to a smart speaker, For example, a smart device (smart phone or tablet), a mobile phone, a personal computer, etc. may be sufficient.

  Here, a conventional estimation apparatus will be described. A conventional estimation device estimates such user attributes based on, for example, user behavior on a network. However, the conventional estimation device has room for improvement in terms of improving the estimation accuracy of user attributes.

  Therefore, the estimation apparatus 1 according to the embodiment estimates the attribute of the user U through a dialogue with the user U. For example, as illustrated in FIG. 1, it is assumed that the information providing terminal 100 according to the embodiment receives a voice input from a user U asking “I want to eat lunch” (step S1).

  In such a case, the information providing terminal 100 according to the embodiment requests content related to lunch from the estimation device 1 serving as a providing server (step S2). Subsequently, for example, the estimation device 1 according to the embodiment generates a question regarding the attribute of the user U using the request for content as a trigger, and transmits the question to the information providing terminal 100 (step S3). In the example illustrated in FIG. 1, the estimation device 1 does not estimate the preference of the user U for ramen in the attribute information 41, and therefore generates a question for estimating the ramen preference.

  And the information provision terminal 100 which concerns on embodiment outputs the question "How is ramen?" Generated by the estimation apparatus 1 (step S4). And the information provision terminal 100 which concerns on embodiment receives the reply "I don't like ramen" of the user U with respect to the output question (step S5).

  The information providing terminal 100 according to the embodiment transmits the received answer to the estimation device 1 (step S6). The estimation apparatus 1 according to the embodiment estimates the attribute of the user U based on the answer acquired from the information providing terminal 100 (step S7). In the example illustrated in FIG. 1, the estimation apparatus 1 performs a linguistic analysis on the user U's answer “I don't like ramen” and enters “×” in the attribute item of ramen.

  As described above, the estimation device 1 according to the embodiment actively asks for a missing attribute among the attributes of the user U, for example, by asking a question regarding the attribute of the user U in the dialogue with the user U. Since it can be heard, the estimation accuracy of the attribute of the user U can be improved.

  In addition, although the case where the estimation apparatus 1 and the information provision terminal 100 were comprised separately was demonstrated in FIG. 1, the estimation apparatus 1 and the information provision terminal 100 may be comprised by one apparatus.

  Next, a configuration example of the estimation apparatus 1 according to the embodiment will be described with reference to FIG. FIG. 2 is a diagram illustrating a configuration of the estimation apparatus 1 according to the embodiment. As illustrated in FIG. 2, the estimation device 1 according to the embodiment includes a communication unit 2, a control unit 3, and a storage unit 4. The estimation device 1 includes an input unit (for example, a keyboard and a mouse) that receives various operations from an administrator who uses the estimation device 1 and a display unit (for example, a liquid crystal display) for displaying various types of information. You may have.

  The communication part 2 is implement | achieved by NIC (Network Interface Card) etc., for example. The communication unit 2 is connected to a communication network by wire or wireless, and transmits / receives information to / from the information providing terminal 100 via the communication network.

  The storage unit 4 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. As shown in FIG. 2, the storage unit 4 stores attribute information 41.

  The attribute information 41 includes information regarding the attribute of the user U. Here, the attribute information 41 will be described with reference to FIG.

  FIG. 3 is an explanatory diagram of the attribute information 41. As shown in FIG. 3, the attribute information 41 includes contents such as “user ID”, “situation”, and “type”.

  “User ID” is identification information for identifying the user U. “Situation” is information indicating a situation around the user U. “Type” is information indicating the type of attribute. Each type includes a plurality of related attribute items. Specifically, the category “food” includes attribute items relating to food such as “sushi” and “ramen”.

  For example, when the user U identified by the user ID “U1” is alone, he likes to eat sushi, and when he is with his family, he likes not to eat sushi but to eat ramen. It is shown that. In the example illustrated in FIG. 3, when the user U identified by the user ID “U1” is alone, the attribute estimation as to whether or not to eat ramen is not completed.

  In the example illustrated in FIG. 3, “situation” indicates the presence or absence of surrounding people, but may be any situation such as weather, time, action history, and the like. As described above, by estimating the attribute for each situation, for example, when providing content to the user U, more appropriate content can be provided.

  Returning to FIG. 2, the control unit 3 will be described. The control unit 3 is a controller. For example, various programs stored in a storage device inside the estimation device 1 are stored in a RAM as a work area by a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like. This is realized by being executed. The control unit 3 is a controller, for example, and is realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

  The control unit 3 generates a question about the attribute of the user U, acquires the user U's answer to the question, and realizes a function of estimating the attribute of the user U based on the answer by the operation of each unit shown below .

  As shown in FIG. 2, the control unit 3 includes a reception unit 31, a detection unit 32, a generation unit 33, an acquisition unit 34, and an estimation unit 35.

  The accepting unit 31 accepts an inquiry from the user U to the information providing terminal 100. Specifically, the dialog with the user U starts when the receiving unit 31 receives the inquiry. Note that there are two types of questions for the user U1: a question specifying a specific attribute (also referred to as an active question) and a question not specifying a specific attribute (also referred to as a passive question). And will be described later with reference to FIG.

  In addition, the reception part 31 is not limited to the case where the user's U inquiry with respect to the information provision terminal 100 is received, For example, it is good also as a trigger of a dialog start by receiving the conversation with the user's U family. That is, the accepting unit 31 may accept as a question even when the user U is not talking directly to the information providing terminal 100.

  The inquiry received by the receiving unit 31 may be, for example, information converted from speech to language by the information providing terminal 100, or may be a sound source recorded by the information providing terminal 100. In addition, when acquiring the sound source which the information provision terminal 100 recorded, the reception part 31 analyzes this sound source and converts it into a language.

  The detection unit 32 detects the situation of the user U. For example, the detection unit 32 is configured with various sensors, and the environment information such as the space where the information providing terminal 100 is installed (that is, the same space as the user U), the temperature outside the space, humidity, weather, etc. Detect as. Moreover, the detection part 32 detects the information regarding the person around the user U as a condition, for example with a microphone, a vibration sensor, an infrared sensor, a camera.

  In addition to the situation around the user U, for example, the detection unit 32 is based on the utterance content of the user U collected by the microphone, for example, the state of the user U's emotions, fatigue, and the like Detects situations.

  Note that the detection unit 32 is not limited to the detection method described above, and may detect the status of the user U based on the user U's answer to the question generated by the generation unit 33, for example. For example, when the detection unit 32 detects that there are a plurality of other users around the user U by the above-described sensor or the like, the detection unit 32 generates a question for detecting details of the plurality of other users with respect to the generation unit 33. Instruct.

  For example, the detection unit 32, by generating unit 33 "Hello everyone, how are you?" To generate a question that contains the greeting such as, from the answer of the other users for such questions, gender and other user, identification information, such as age Or the relationship with the user U is detected. Thus, the accuracy of the detected situation can be improved by detecting the situation of the user U based on the answer to the question generated by the generation unit 33. Note that the detection unit 32 is not limited to a question including a greeting, and may make a question including a question regarding an attribute such as “Do you like ramen?”, For example. Thereby, the estimation accuracy of the attribute of the user U can be further improved.

  Alternatively, when the quality of the answer of the user U to the question about the attribute generated by the generation unit 33 (word strength, speech rate, volume, etc.) is different from normal, the detection unit 32 determines that the situation of the user U is normal. May detect that they are different. In such a case, the detection unit 32 temporarily suspends the question regarding the attribute of the user U and estimates the situation of the user U by asking the situation for estimating the situation of the user U.

  The generation unit 33 generates a question regarding the attribute of the user U based on the attribute information 41. The specific processing content of the generation unit 33 will be described later with reference to FIGS.

  The acquisition unit 34 acquires the user U's answer to the question generated by the generation unit 33. The acquisition unit 34 acquires which option has been selected when the question is a question that can be answered with one choice. Options include a method of selecting affirmative and negative such as “Yes” and “No”, a method of selecting from a plurality of candidates, and the like.

  Further, the acquisition unit 34 is not limited to the case of acquiring an alternative answer. For example, the acquisition unit 34 may acquire an answer indicating a preference for a specific attribute such as “I like XX”. Good.

  The estimation unit 35 estimates the attribute of the user U based on the answer acquired by the acquisition unit 34. For example, when the answer acquired by the acquisition unit 34 is an alternative answer indicating whether the answer is affirmative or negative, the estimation unit 35 displays “O” (positive) or “×” (negative) in the blank of the attribute item in the attribute information 41. )

  The estimation unit 35 estimates the attribute for each situation detected by the detection unit 32. That is, the estimation unit 35 associates an answer to the question of the user U with a situation around the user U. Thereby, since the attribute of the user U which changes according to a situation can be caught exactly, more appropriate content can be provided.

  Note that the estimation unit 35 is not limited to the case where the attribute items of the attribute information 41 are estimated by the XX method, and for example, a predetermined index value may be entered in each attribute item.

  For example, when there are both positive and negative preferences of the user U in a predetermined attribute item, the estimation unit 35 enters an index value corresponding to the ratio of positive and negative answers. That is, each attribute item is relatively evaluated without being classified into two types (positive or negative) as in the case of the XX system. Thereby, the attribute of the user U can be estimated in more detail.

  Next, the processing content of the estimation apparatus 1 according to the embodiment will be specifically described with reference to FIGS. 4-6 is a figure which shows the specific example of a process of the estimation apparatus 1 which concerns on embodiment.

  In FIG. 4, it is assumed that the user U asks the information providing terminal 100 “I want to eat lunch”. In other words, it can be said that such a question is a question for designating a specific attribute. That is, such a question is mainly made by the user U in anticipation of a specific response to the information providing terminal 100.

  When such a main question is generated, the generation unit 33 generates a question related to the question. That is, the generation unit 33 generates a question related to a specific attribute when the inquiry received by the reception unit 31 is a content specifying a specific attribute.

  In the example illustrated in FIG. 4, the generation unit 33 determines that the user U has specified an attribute related to food as a result of linguistic analysis of “I want to eat lunch”, and the type of the attribute information 41 is “food”. The attribute item is the target of the question.

  In the example illustrated in FIG. 4, the generation unit 33 generates a question “how is ramen?” Regarding the attribute item “ramen” that has not been estimated by the estimation unit 35. In this way, by making a question that matches the question in which the user U has specified a specific attribute, it is possible to eliminate the user's uncomfortable feeling with respect to the question and to reliably obtain an answer to the question.

  Moreover, since the potential attribute of the user U can be searched by generating a question regarding the attribute item for which the estimation of the attribute information 41 has not been completed, the attribute of the user U can be widely estimated. As a result, more appropriate content can be provided.

  Note that the generation unit 33 generates a question regarding “food” again when the answer of the user U to “How about ramen?” Is negative. At this time, for example, the generation unit 33 generates a question regarding the attribute item “sushi” that has already been estimated.

  Thereby, since the response suitable for the user U can be performed, the situation where the user U is questioned many times can be avoided. That is, the satisfaction level of the user U can be improved.

  Note that the generation unit 33 may generate again a question regarding attribute items for which the estimation by the estimation unit 35 has not been completed. The attributes of the user U can be estimated more widely.

  Moreover, although the production | generation part 33 produced | generated the question of the attribute relevant to the user's U inquiry, it is not limited to this, What is necessary is just the question of the attribute relevant to the content of the user's U interaction. That is, the generation unit 33 is an attribute question regarding the content of the conversation with another user or the content generated in a series of conversations with the information providing terminal 100 regardless of the user U's inquiry to the information providing terminal 100. May be.

  In this way, the generation unit 33 can perform a question that does not give the user U a sense of incongruity by analyzing the content of the dialogue and asking a question according to the content.

  Furthermore, the generation unit 33 can generate a question that can be answered with an alternative such as “How about ramen?” So that the user U can easily answer the question. That is, an answer for estimating the attribute of the user U can be obtained more easily.

  Next, in FIG. 5, it is assumed that the user U asks the information providing terminal 100 that he / she is “free”. In other words, it can be said that such a question is a question not specifying a specific attribute. That is, such a question is not intended for a specific response to the information providing terminal 100, but is passively performed by the user U in the hope of some new information.

  In such a case, the generation unit 33 makes a question regarding an arbitrary attribute. Specifically, the generation unit 33 generates a question related to an arbitrary attribute when the inquiry received by the reception unit 31 is content that does not specify a specific attribute. An arbitrary attribute is, for example, a blank attribute item in the attribute information 41 that has not been estimated for an arbitrary type.

  In the example illustrated in FIG. 5, the generation unit 33 makes a question regarding an arbitrary attribute such as “Would you like to watch the Olympics?”. The attribute item corresponding to “Would you like to watch the Olympics?” May be “sport” or “television”. Alternatively, it may be a type corresponding to “Olympic”.

  The generation unit 33 may preferentially select a type having a large number of attribute items for which estimation has not been completed, for example, as an arbitrary attribute selection method. Or the production | generation part 33 may presume and select the attribute which the user U answers actively from the user's U situation detected by the detection part 32, and the action on the past network of the user U.

  That is, the generation unit 33 actively asks the passive user U a question. Thereby, the preference property of the user U's potential attribute can be easily searched.

  In the example shown in FIG. 5, when the user U gives an affirmative answer to “Would you like to watch the Olympics?”, You may ask a question regarding more detailed attribute items such as the type of Olympics. .

  When the user U gives an affirmative answer to “Do you want to watch the Olympics?”, The generation unit 33 may generate a question regarding different types of attributes.

  Next, in FIG. 6, it is assumed that the user U answered “Like! Thank you” to the question “How about ramen?” Output from the information providing terminal 100 and the series of dialogues has been completed.

  When the dialog is completed, the generation unit 33 generates a question regarding an attribute different from the attribute related to the dialog. In the example illustrated in FIG. 6, the dialogue regarding the attribute “food” is completed, and a question regarding an attribute (for example, sports) related to “food” is made.

  In the example illustrated in FIG. 6, after the dialogue regarding “food” is completed, the generation unit 33 makes a question regarding an attribute related to “food” such as “How about sports after lunch?”. Then, when the estimation unit 35 obtains a positive answer to a question such as “just like having a stomach!”, The estimation unit 35 estimates an attribute that associates “food” with “sports”.

  That is, since the generation unit 33 has been in conversation until immediately before, the user U asks a new attribute using the fact that the user U can easily answer. Thereby, when making a question of a new attribute to the user U, it is possible to reduce the occurrence of a sudden feeling and confusion with respect to the user U.

  In this case, the generation unit 33 can further reduce the sense of suddenness for the user U by making a question regarding the type related to the attribute type in the previous conversation.

  In addition, the generation unit 33 generates the question by using the reception unit 31 as a trigger of the user U's inquiry. For example, the generation unit 33 may generate the question without the user U's inquiry like a push notification.

  Further, the generation unit 33 may calculate the degree of association between attribute items or various types based on the attribute information 41 of a plurality of users U, and may generate a question based on the degree of association. Specifically, the generation unit 33 calculates a high degree of relevance when preferences among attribute items are similar among a plurality of users U having similar living behaviors.

  For example, after eating ramen among a plurality of users U, if the number of users U exercising is a predetermined number or more, after asking questions about ramen, questions relating to exercise attribute items are generated. That is, the generation unit 33 generates a question by predicting the preferences of the users U in advance from the tendency of the preferences of the plurality of users U. Thereby, the satisfaction degree of the user U can be improved while searching for the potential attribute of the user U.

  In the above, the estimation unit 35 estimates that the user U prefers the attribute of the question when the user U makes a positive answer to the question, but also when the user U makes a negative answer The attribute of the user U is estimated.

  Specifically, for example, the estimation unit 35 supposes that the user U1 has answered “No” to the question “Do you hear entertainment-related news?” Generated by the generation unit 33. In such a case, the estimation unit 35 estimates that the entertainment-related news is not liked, and also estimates that the news related to the entertainment-related news is not liked among the news of other genres (types or attribute items). To do.

  More specifically, for example, the estimation unit 35 estimates that the same or similar news (for example, sports news) that is the same as or similar to the type of abstracting entertainment related news (for example, sports news) is not liked. Thus, the attribute of the user U can be widely estimated by estimating the attribute also from the negative answer of the user U.

  In addition, although the estimation part 35 estimated not to like the news similar to entertainment-related news, on the other hand, you may estimate that it likes the news (economic etc.) opposite to entertainment-related news. .

  Next, a procedure of processing executed by the estimation apparatus 1 according to the embodiment will be described with reference to FIG. FIG. 7 is a flowchart illustrating a procedure of processing executed by the estimation apparatus 1 according to the embodiment.

  As illustrated in FIG. 7, first, the reception unit 31 receives an inquiry from the user U1 (step S101). The generation unit 33 determines whether the inquiry received by the reception unit 31 is an inquiry for designating a specific attribute (step S102).

  If the question received by the receiving unit 31 is a question for specifying a specific attribute (Yes in step S102), the generating unit 33 generates a question regarding the specified attribute (step S103).

  Subsequently, the acquisition unit 34 acquires the user U's answer to the question generated by the generation unit 33 (step S104). Subsequently, the estimation unit 35 estimates the attribute of the user U based on the answer acquired by the acquisition unit 34 (step S105), and ends the process.

  On the other hand, in step S102, when the question received by the receiving unit 31 is a question not specifying a specific attribute (No in step S102), the generating unit 33 generates a question regarding an arbitrary attribute (step S106). The process proceeds to step S104.

  Moreover, the estimation apparatus 1 according to the embodiment described above is realized by a computer 1000 having a configuration as shown in FIG. 8, for example. FIG. 8 is a hardware configuration diagram illustrating an example of a computer 1000 that implements the functions of the estimation apparatus 1 according to the embodiment. The computer 1000 includes a CPU 1100, RAM 1200, ROM 1300, HDD 1400, communication interface (I / F) 1500, input / output interface (I / F) 1600, and media interface (I / F) 1700.

  The CPU 1100 operates based on a program stored in the ROM 1300 or the HDD 1400 and controls each unit. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 is started up, a program depending on the hardware of the computer 1000, and the like.

  The HDD 1400 stores programs executed by the CPU 1100, data used by the programs, and the like. The communication interface 1500 receives data from other devices via the network N and sends the data to the CPU 1100, and transmits data generated by the CPU 1100 to other devices via the network N.

  The CPU 1100 controls an output device such as a display and a printer and an input device such as a keyboard and a mouse via the input / output interface 1600. The CPU 1100 acquires data from the input device via the input / output interface 1600. Further, the CPU 1100 outputs the generated data to the output device via the input / output interface 1600.

  The media interface 1700 reads a program or data stored in the recording medium 1800 and provides it to the CPU 1100 via the RAM 1200. The CPU 1100 loads such a program from the recording medium 1800 onto the RAM 1200 via the media interface 1700, and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or PD (Phase change rewritable disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. Etc.

  For example, when the computer 1000 functions as the estimation device 1 according to the embodiment, the CPU 1100 of the computer 1000 implements the function of the control unit 3 by executing a program loaded on the RAM 1200. The HDD 1400 stores data in the storage unit 4. The CPU 1100 of the computer 1000 reads these programs from the recording medium 1800 and executes them, but as another example, these programs may be acquired from other devices via the network N.

  As described above, the estimation device 1 according to the embodiment includes the generation unit 33, the acquisition unit 34, and the estimation unit 35. The generation unit 33 generates a question regarding the attribute of the user U in the dialog with the user U. The acquisition unit 34 acquires the user U's answer to the question generated by the generation unit 33. The estimation unit 35 estimates the attribute of the user U based on the answer acquired by the acquisition unit 34.

  Thereby, the estimation accuracy of the attribute of the user U can be improved.

  In the estimation device 1 according to the above-described embodiment, the generation unit 33 generates a question that can be answered by the user U.

  Thereby, since the user U can easily answer a question, an answer for estimating the attribute of the user U can be obtained more easily.

  Moreover, in the estimation apparatus 1 according to the above-described embodiment, the generation unit 33 generates an attribute question related to the content of the dialogue.

  Thereby, the sudden feeling with respect to the user's U question can be reduced.

  In the estimation device 1 according to the above-described embodiment, the attribute of the user U includes a plurality of attribute items. The estimation unit 35 estimates each of a plurality of attribute items. The generation unit 33 generates a question regarding attribute items that have not been estimated by the estimation unit 35.

  Thereby, the attribute of the user U can be estimated widely.

  Moreover, in the estimation apparatus 1 according to the above-described embodiment, the generation unit 33 generates a question regarding an attribute different from the attribute related to the dialog when the dialog is completed.

  Thereby, when making a question of a new attribute to the user U, it is possible to reduce the occurrence of a sudden feeling and confusion with respect to the user U.

  In addition, the estimation device 1 according to the above-described embodiment further includes a reception unit 31. The accepting unit 31 accepts an inquiry from the user U. The generation unit 33 generates a question related to a specific attribute when the inquiry received by the reception unit 31 is a content specifying a specific attribute.

  Thereby, while being able to eliminate the uncomfortable feeling with respect to the question of the user U, an answer to the question can be reliably obtained.

  Moreover, in the estimation apparatus 1 which concerns on embodiment mentioned above, the production | generation part 33 produces | generates the question regarding arbitrary attributes, when the inquiry which the reception part 31 received is the content which does not designate a specific attribute.

  Thereby, the preference property of the user U's potential attribute can be easily searched.

  In addition, the estimation device 1 according to the above-described embodiment further includes a detection unit 32. The detection unit 32 detects the situation of the user U. The generation unit 33 generates a question corresponding to the situation detected by the detection unit 32. The estimation unit 35 estimates an attribute for each situation in the user U.

  Thereby, since the attribute of the user U which changes according to a situation can be caught exactly, more appropriate content can be provided.

  Further, in the estimation device 1 according to the above-described embodiment, the generation unit 33 further generates a situation question regarding the situation of the user U. The acquisition unit 34 acquires the user U's situation answer to the situation question generated by the generation unit 33. The detection unit 32 detects the situation of the user U based on the situation answer acquired by the acquisition unit 34.

  Thereby, the estimation precision in the attribute for every situation of the user U can be improved more.

  As described above, some of the embodiments of the present application have been described in detail with reference to the drawings. However, these are merely examples, and various modifications, including the aspects described in the disclosure section of the invention, based on the knowledge of those skilled in the art, It is possible to implement the present invention in other forms with improvements.

  In addition, among the processes described in the above embodiment, all or part of the processes described as being automatically performed can be performed manually, or the processes described as being performed manually can be performed. All or a part can be automatically performed by a known method. In addition, the processing procedures, specific names, and information including various data and parameters shown in the document and drawings can be arbitrarily changed unless otherwise specified. For example, the various types of information illustrated in each drawing is not limited to the illustrated information.

  Further, each component of each illustrated apparatus is functionally conceptual, and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured.

  In addition, the processes described in the above-described embodiments can be appropriately combined within a range that does not contradict the processing contents.

  Moreover, the above-mentioned “section (module, unit)” can be read as “means”, “circuit”, and the like. For example, the detection unit 32 can be read as detection means or a detection circuit.

DESCRIPTION OF SYMBOLS 1 Estimation apparatus 2 Communication part 3 Control part 4 Memory | storage part 31 Reception part 32 Detection part 33 Generation part 34 Acquisition part 35 Estimation part 41 Attribute information 100 Information provision terminal U user

Claims (11)

A generation unit for generating a question regarding the attribute of the user in the dialog with the user;
An acquisition unit for acquiring the user's answer to the question generated by the generation unit;
An estimation device comprising: an estimation unit that estimates an attribute of the user based on the answer acquired by the acquisition unit. The generator is
The estimation apparatus according to claim 1, wherein the question is generated so that the user can answer with an alternative. The generator is
The estimation apparatus according to claim 1, wherein the question of the attribute related to the content of the dialogue is generated. The attribute includes a plurality of attribute items,
The estimation unit includes
Estimating each of the plurality of attribute items;
The generator is
The estimation device according to claim 1, wherein the question regarding the attribute item that has not been estimated by the estimation unit is generated. The generator is
The estimation apparatus according to claim 1, wherein when the dialogue is completed, the question related to the attribute different from the attribute related to the dialogue is generated. A reception unit for receiving the user's inquiry;
The generator is
The said question relevant to the said specific attribute is produced | generated when the said inquiry which the said reception part received is the content which designates the specific said attribute, The Claim 1 characterized by the above-mentioned. Estimating device. The generator is
The estimation device according to claim 6, wherein, when the inquiry received by the reception unit is content that does not designate a specific attribute, the question regarding any attribute is generated. A detection unit for detecting the situation of the user;
The generator is
Generating the question according to the situation detected by the detection unit;
The estimation unit includes
The estimation apparatus according to any one of claims 1 to 7, wherein the attribute for each situation of the user is estimated. The generator is
Further generating a status question regarding the user's status;
The acquisition unit
Obtaining the user's situation answer to the situation question generated by the generator;
The detector is
The estimation apparatus according to claim 8, wherein the situation of the user is detected based on the situation answer obtained by the acquisition unit. An estimation method performed by a computer,
Generating a question for the user in the dialog with the user;
An acquisition step of acquiring the user's answer to the question generated by the generation step;
An estimation method comprising: an estimation step of estimating the attribute of the user based on the answer acquired by the acquisition step. A generation unit for generating a question to the user in the dialog with the user;
An acquisition unit for acquiring the user's answer to the question generated by the generation unit;
An estimation program that causes a computer to execute an estimation unit that estimates an attribute of the user based on the answer acquired by the acquisition unit.

Images