Menstrual cycle-related information output device, learning device, learning information generation method, and storage medium

Abstract

[Technical Problem] Previously, it was not possible to predict menstrual-related information using abdominal sounds. [Technical Means] A menstrual-related information output device comprising a learning information storage unit, a sound information acquisition unit, a prediction unit, and an output unit can predict menstrual-related information using abdominal sounds from the abdomen. The learning information storage unit stores learning information consisting of two or more teaching data points, including sound information acquired from the user's abdominal sounds and menstrual-related information related to menstruation. The sound information acquisition unit acquires sound information from the user's abdominal sounds. The prediction unit applies the learning information to the sound information acquired by the sound information acquisition unit and acquires menstrual-related information. The output unit outputs the menstrual-related information acquired by the prediction unit.

Description

Technical Field

[0001] This invention relates to a menstrual-related information output device that acquires and outputs menstrual-related information associated with menstruation. Background Technology

[0002] Previously, there have been technologies that can predict menstrual dates even when there are large differences in menstrual cycles, suppressing the decrease in prediction accuracy (see Patent Document 1).

[0003] In addition, there is a technique that obtains measurement information about the state of leukorrhea and uses this measurement information to predict the ovulation date (see Patent Document 2).

[0004] Patent documents

[0005] Patent Document 1: Japanese Patent Application No. 2015-523319

[0006] Patent Document 2: Japanese Patent Application Publication No. 2014-64706 Summary of the Invention

[0007] The technical problem that the invention aims to solve

[0008] However, previous techniques, which utilized abdominal sounds from the abdomen or surrounding area, failed to obtain menstrual-related information.

[0009] Technical means to solve technical problems

[0010] The menstrual-related information output device of the present invention comprises: a learning information storage unit that stores learning information consisting of two or more teaching data having sound information obtained from the user's abdominal voice and menstrual-related information associated with menstruation; a sound information acquisition unit that acquires sound information from the user's abdominal voice; a prediction unit that applies the learning information to the sound information acquired by the sound information acquisition unit and acquires menstrual-related information; and an output unit that outputs the menstrual-related information acquired by the prediction unit.

[0011] By utilizing this configuration, menstrual-related information can be obtained by taking abdominal sounds from the abdomen or surrounding area.

[0012] Furthermore, the menstrual-related information output device of the second invention is as follows: Regarding the first invention, the menstrual-related information is menstrual date relationship information concerning the relationship between dates related to menstruation.

[0013] Through this configuration, abdominal sounds can be used to obtain menstrual correlation information about the relationship between menstrual dates.

[0014] Furthermore, the menstrual-related information output device of this third invention is as follows: Regarding the first invention, the menstrual-related information is pain information related to menstrual pain.

[0015] This structure allows for the acquisition of pain information related to menstrual cramps by utilizing abdominal sounds.

[0016] Furthermore, the menstrual-related information output device of the fourth invention is as follows: with respect to any one of the first to third inventions, two or more teaching data are composed of teaching data having sound information and menstrual-related information obtained from abdominal sounds obtained from the user's abdomen daily between menstrual cycles.

[0017] Through this structure, information related to menstruation can be obtained by using abdominal sounds.

[0018] Furthermore, the menstrual-related information output device of the fifth invention is as follows: with respect to any one of the first to fourth inventions, it also includes a learning unit that learns and processes two or more teaching data through a machine learning algorithm to obtain learning information; and a prediction unit that uses the sound information already obtained by the sound information acquisition unit and the learning information to perform prediction processing through a machine learning algorithm to obtain menstrual-related information.

[0019] With this structure, menstrual-related information can be obtained using abdominal sounds through machine learning algorithms.

[0020] Furthermore, the menstrual-related information output device of this sixth invention is as follows: with respect to any one of the first to fifth inventions, the sound information is two or more characteristic quantities of the user's abdominal sound.

[0021] With this configuration, sound characteristics are obtained from abdominal sounds originating from or around the abdomen, and menstrual-related information can be obtained using these characteristics.

[0022] Furthermore, the learning device of this seventh invention comprises: a sound information acquisition unit that acquires sound information from the user's abdominal sound; a learning receiving unit that receives menstrual-related information; a teaching data composition unit that composes teaching data from the sound information and the menstrual-related information; a learning unit that performs machine learning processing on the teaching data composed by the teaching data composition unit to compose a learner, i.e., learning information; and an accumulation unit that accumulates the learner.

[0023] With this configuration, a learner capable of predicting menstrual-related information can be constructed using abdominal sounds and machine learning algorithms.

[0024] Beneficial effects

[0025] According to the menstrual-related information output device of the present invention, menstrual-related information can be predicted using abdominal sounds. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of information system A in implementation method 1.

[0027] Figure 2 This is a block diagram of information system A.

[0028] Figure 3 This is a block diagram of the menstrual-related information output device 2.

[0029] Figure 4 This is a flowchart illustrating the operation of the learning device 1.

[0030] Figure 5 This is a flowchart illustrating the first example of how this learning information is processed.

[0031] Figure 6 This is a flowchart illustrating the second example of how the learning information is processed.

[0032] Figure 7 This is a flowchart illustrating the operation of the menstrual-related information output device 2.

[0033] Figure 8 This is a flowchart illustrating an example of the predictive processing.

[0034] Figure 9 This is a flowchart illustrating the operation of the terminal device 3.

[0035] Figure 10 This is a diagram representing the teaching data management table.

[0036] Figure 11 This is a diagram representing the output example.

[0037] Figure 12 This is a diagram representing the output example.

[0038] Figure 13 This is a diagram representing the output example.

[0039] Figure 14 This is a diagram of the computer system's appearance.

[0040] Figure 15 This is a block diagram of the computer system. Detailed Implementation

[0041] Hereinafter, embodiments of the menstrual-related information output device, etc., will be described with reference to the accompanying drawings. Furthermore, since components marked with the same symbols in the embodiments perform the same function, repeated descriptions may be omitted.

[0042] Implementation Method 1

[0043] In this embodiment, a menstrual-related information output device is described that applies sound information about a user's abdominal sounds to learning information composed of two or more teaching data containing sound information about the user's abdominal sounds and menstrual-related information, and then obtains and outputs menstrual-related information.

[0044] Furthermore, abdominal sounds refer to sounds emanating from the user's abdomen. Abdominal sounds can also be considered to include sounds emanating from the periphery of the user's abdomen. For example, abdominal sounds may include intestinal sounds emanating from the intestines. Additionally, abdominal sounds may include sounds caused by blood flow in the abdomen (e.g., the sound of the abdominal aorta) or sounds emanating from organs such as the stomach. Menstrual-related information is information associated with menstruation, which will be described in detail later. Moreover, learning information includes, for example, learners composed of learning devices, correspondence tables described later, etc. Furthermore, learners can be described as classifiers, models, etc.

[0045] In addition, in this embodiment, a learning device is described that learns from two or more teaching data points containing sound information about the user's abdominal sounds and menstrual-related information, and constructs a learner by learning from them using a machine learning algorithm.

[0046] Furthermore, in this embodiment, an information system comprising a learning device, a menstrual-related information output device, and one or more terminal devices will be described.

[0047] Figure 1 This is a schematic diagram of information system A in this embodiment. Information system A includes a learning device 1, a menstrual-related information output device 2, and one or more terminal devices 3.

[0048] Learning device 1 is a device that learns from two or more teaching data points containing sound information and menstrual-related information by using machine learning algorithms to process and construct a learner.

[0049] Menstrual-related information output device 2 is a device that uses abdominal sounds to obtain and output menstrual-related information.

[0050] The learning device 1 and the menstrual-related information output device 2 are computers, such as servers. The learning device 1 and the menstrual-related information output device 2 are, for example, cloud servers, ASP servers, etc., and their types are not limited. Furthermore, the learning device 1 and the menstrual-related information output device 2 can also be independent devices.

[0051] Terminal device 3 is a terminal used by a user. The user is someone who wants to obtain menstrual-related information. Terminal device 3 is a terminal used to obtain learning information. Terminal device 3 is, for example, a personal computer, tablet computer, smartphone, etc., and its type is not limited.

[0052] Figure 2 This is a block diagram of information system A in this embodiment. Figure 3 This is a block diagram of the menstrual-related information output device 2.

[0053] The learning device 1 includes a teaching data storage unit 11, a sound collection unit 12, a sound information acquisition unit 13, a learning receiving unit 14, a teaching data composition unit 15, a learning unit 16, and an accumulation unit 17.

[0054] The menstrual-related information output device 2 includes a storage unit 21, a receiving unit 22, a processing unit 23, and an output unit 24. The storage unit 21 includes a learning information storage unit 211. The processing unit 23 includes a sound information acquisition unit 231 and a prediction unit 232.

[0055] The terminal device 3 includes a terminal storage unit 31, a terminal receiving unit 32, a terminal processing unit 33, a terminal transmitting unit 34, a terminal receiving unit 35, and a terminal output unit 36.

[0056] The teaching data storage unit 11, which constitutes the learning device 1, stores one or more teaching data. The teaching data includes audio information and menstrual-related information. Audio information refers to information obtained based on abdominal sounds. Audio information can be the recorded abdominal sound data itself, or data obtained by processing or editing the recorded data.

[0057] Sound information can be, for example, a spectral image representing the result of analyzing sound data (or processed data) obtained by Fourier transform or high-speed Fourier transform of recorded abdominal sounds in a prescribed form. Furthermore, sound information can be the sound data (or processed data) itself, or data transformed in other forms. Sound information can also be, for example, a set of features obtained by performing A / D conversion on abdominal sounds and cepstral analysis on the converted data. Additionally, sound information can also be, for example, a set of features obtained by performing A / D conversion on abdominal sounds and LPC analysis on the converted data. Sound information consists of two or more sound features obtained from the user's abdominal sounds.

[0058] The sound collection unit 12 collects abdominal sounds from or around a user's abdomen. The sound collection unit 12 is, for example, a microphone.

[0059] The sound information acquisition unit 13 acquires sound information. This sound information is obtained from abdominal sounds. The sound information acquisition unit 13 acquires sound information from the abdominal sounds used for predictive processing, which involves acquiring menstrual-related information described later. Furthermore, the sound information acquisition unit 13 can also acquire sound information from abdominal sounds received via the terminal device 3, and can also acquire sound information received from the terminal device 3. Additionally, the sound information acquisition unit 13 can also acquire sound information from abdominal sounds already acquired by the sound collection unit 12.

[0060] The sound information acquisition unit 13 performs A / D conversion on abdominal sounds, for example, and acquires sound information. The sound information acquisition unit 13 also performs cepstral analysis on abdominal sounds, for example, to obtain a vector of multidimensional feature quantities, i.e., sound information. Furthermore, the sound information acquisition unit 13 performs LPC analysis on abdominal sounds, for example, to obtain a vector of multidimensional feature quantities, i.e., sound information.

[0061] The learning receiving unit 14 receives menstrual-related information. Typically, the learning receiving unit 14 receives user input, i.e., menstrual-related information. The learning receiving unit 14 usually receives menstrual-related information in conjunction with a user identifier. The user identifier is information that identifies the user. Examples of user identifiers include ID, email address, phone number, and name.

[0062] The learning receiver 14 can also receive abdominal sounds and menstrual-related information. In this case, the sound collection unit 12 is not needed in the learning device 1.

[0063] The learning receiving unit 14 can also receive teaching data containing sound information and menstrual-related information. In this case, the learning device 1 does not require a sound collection unit 12 and a sound information acquisition unit 13.

[0064] Preferably, the teaching data or information related to abdominal sounds and menstruation received by the learning receiving unit 14 is associated with a user identifier.

[0065] Menstrual-related information is information associated with menstruation. Examples of menstrual-related information include menstrual date relationship information and pain information. Menstrual date relationship information is information about the relationship between dates related to menstruation (e.g., start date of menstruation, ovulation date, end date of menstruation). Examples of menstrual date relationship information include whether the start date of menstruation is approaching, whether menstruation is currently underway, the number of days since the start date of menstruation, the number of days since ovulation, and the length of the menstrual period. Pain information is information about the pain of the next menstrual period. Examples of pain information include whether the pain is weak or strong, and the level of pain (e.g., any value from 1 to 5 or any value from 1 to 10).

[0066] In addition, the reception here is usually from the terminal device 3, but it can also include the reception from a microphone, the reception of information input from input devices such as a keyboard, mouse or touch screen, and the reception of information read from storage media such as optical disc or disk, semiconductor memory, etc.

[0067] Menstrual-related information can be input via any of the following methods: touchscreen, keyboard, mouse, or menu screen.

[0068] The teaching data composition unit 15 composes teaching data from voice information and menstrual-related information. For example, the teaching data composition unit 15 composes a vector containing voice information and menstrual-related information, i.e., teaching data. Alternatively, the teaching data composition unit 15 composes a vector that uses voice information (i.e., one or more feature quantities) and menstrual-related information as elements, i.e., teaching data. Preferably, the teaching data is associated with a user identifier.

[0069] The teaching data composition unit 15 can also use one or more previously received menstrual-related information to obtain other menstrual-related information. That is, the previously received menstrual-related information can also be information different from the menstrual-related information accumulated in relation to the sound information.

[0070] The teaching data composition unit 15 obtains menstrual-related information "menstrual period" by using menstrual-related information indicating the "start date of menstruation" and menstrual-related information indicating the "end date of menstruation," for example. That is, the teaching data composition unit 15 obtains date information indicating the date on which the menstrual-related information indicating the "start date of menstruation" is received. Additionally, the teaching data composition unit 15 obtains date information indicating the date on which the menstrual-related information indicating the "end date of menstruation" is received. Then, the teaching data composition unit 15 calculates the difference between the two date information to obtain the menstrual-related information "menstrual period." Furthermore, the teaching data composition unit 15 can obtain date information from a clock not shown, or it can obtain date information already received from the terminal device 3. The method of obtaining date information is not limited.

[0071] The teaching data composition unit 15, for example, uses menstrual-related information indicating "non-menstrual period" and menstrual-related information indicating "menstrual start date" to obtain menstrual-related information "days' information indicating the number of days since the start date of menstruation". In other words, the teaching data composition unit 15 obtains date information indicating the date on which the menstrual-related information indicating "non-menstrual period" is received. Additionally, the teaching data composition unit 15 obtains date information indicating the date on which the menstrual-related information indicating "menstrual start date" is received. Then, the teaching data composition unit 15 calculates the difference between the two date information to obtain the menstrual-related information "days' information indicating the number of days since the start date of menstruation".

[0072] The teaching data composition unit 15, for example, uses menstrual-related information indicating the "start date of menstruation" and menstrual-related information indicating "non-menstrual period" to obtain menstrual-related information "days' information indicating the number of days from the ovulation date." In other words, the teaching data composition unit 15 obtains date information indicating the date on which the menstrual-related information indicating the "start date of menstruation" is received. Additionally, the teaching data composition unit 15 obtains the usual number of days from the start date of menstruation to the ovulation date from the storage unit 21. Next, the teaching data composition unit 15 calculates date information indicating the ovulation date using the date information corresponding to the start date of menstruation and the number of days from the ovulation date. Next, the teaching data composition unit 15, for example, obtains the date information of the date on which the menstrual-related information is received. Next, the teaching data composition unit 15 calculates the difference between the date information indicating the ovulation date and the date information of the date on which the menstrual-related information is received, and obtains the number of days of this difference, which is the menstrual-related information "days' information indicating the number of days from the ovulation date."

[0073] Furthermore, the teach data composition unit 15 accumulates the composed teach data in the teach data storage unit 11. Preferably, the teach data composition unit 15 accumulates the composed teach data in association with user identifiers. Additionally, it is preferable that the teach data composition unit 15 accumulates the composed teach data in association with date information.

[0074] Learning Department 16 uses one or more teaching data to obtain learning information.

[0075] For example, for each user identifier, the learning unit 16 obtains learning information using one or more teaching data pairs that are paired with that user identifier.

[0076] For example, for each type of menstrual-related information (e.g., number of days since the start of menstruation, level of pain), the learning unit 16 uses one or more teaching data to obtain learning information.

[0077] For example, for each type of menstrual-related information and user identifier, the learning unit 16 uses one or more teaching data to obtain learning information.

[0078] For example, for two or more teaching data sets, the learning unit 16 performs learning processing using machine learning algorithms to obtain a learner, i.e., learning information. For the teaching data constructed by the teaching data composition unit 15, the learning unit 16 performs machine learning processing to construct a learner, i.e., learning information.

[0079] While machine learning algorithms can utilize deep learning, decision trees, random forests, SVM, SVR, etc., there is no limitation on their application. Furthermore, machine learning can utilize various machine learning functions and existing libraries, such as TensorFlow libraries, fastText, tinySVM, and the random forest module in R. Moreover, modules can be categorized as programs, software, functions, methods, etc.

[0080] Preferably, two or more teaching data sets are composed of teaching data that have acquired sound information and menstrual-related information from the daily abdominal sounds between a user's menstrual cycles.

[0081] Learning section 16, for example, constitutes a correspondence table. The correspondence table has two or more corresponding pieces of information. It can also be said that the corresponding information is teaching data. The corresponding information represents the correspondence between sound information and menstrual-related information. For example, the corresponding information represents the correspondence between sound information and one or more types of menstrual-related information. For example, the corresponding information represents the correspondence between sound information and one or more types of menstrual-related information and one or more other types of menstrual-related information.

[0082] The mapping table can also exist for each user identifier. The mapping table can also exist for each type of menstrual-related information. The mapping table can also exist for both each user identifier and each type of menstrual-related information.

[0083] The accumulation unit 17 accumulates the learning information already acquired by the learning unit 16. For example, the accumulation unit 17 accumulates the learning information already acquired by the learning unit 16. Furthermore, the location for accumulating the learning information in the accumulation unit 17 can be either a local storage medium or another device such as the menstrual-related information output device 2.

[0084] For example, for each user identifier, the accumulation unit 17 accumulates the learning information obtained by the learning unit 16 in correspondence with each user identifier. For example, for each type of menstrual-related information, the accumulation unit 17 accumulates the learning information obtained by the learning unit 16 in correspondence with the identifier of each type of menstrual-related information. For example, for each user identifier and each type of menstrual-related information, the accumulation unit 17 accumulates the learning information obtained by the learning unit 16 in correspondence with each user identifier and the identifier of the type. Furthermore, the identifier of the type is, for example, "whether the start date of menstruation is approaching", "whether it is during menstruation", "number of days from the start date of menstruation", "number of days from the date of ovulation", and "length of menstrual period".

[0085] Various types of information are stored in the storage unit 21 that constitutes the menstrual-related information output device 2. These various types of information include, for example, learning information.

[0086] One or more pieces of learning information are stored in the learning information storage unit 211. The learning information may be, for example, the learner or the mapping table described above. Preferably, the learning information is information already acquired by the learning device 1. Preferably, the learning information in the learning information storage unit 211 corresponds to a user identifier. That is, it is preferable to use different learning information for each user. However, common learning information can also be used for two or more users. The learning information may correspond to, for example, the user identifier and an identifier for the type of menstrual-related information.

[0087] The receiving unit 22, for example, receives the abdominal sound of a user. The receiving unit 22, for example, receives sound information obtained from the abdominal sound of a user. The receiving unit 22, for example, receives the abdominal sound or sound information in correspondence with a user identifier.

[0088] The receiving unit 22 receives, for example, an output indication. The output indication is an output indication of menstrual-related information. The output indication may, for example, contain data related to abdominal sounds. The output indication may, for example, contain sound information. Preferably, the output indication includes a user identifier.

[0089] The receiving unit 22 receives, for example, abdominal sounds or sound information or output instructions from the terminal device 3.

[0090] The information received by the receiving unit 22 is usually received from the terminal device 3, but it can also include receiving information from a microphone, receiving information input from input devices such as a keyboard, mouse or touch screen, and receiving information read from storage media such as optical discs, disks, semiconductor memory, etc.

[0091] The processing unit 23 performs various processes. These various processes include, for example, those performed by the sound information acquisition unit 231 and the prediction unit 232.

[0092] The sound information acquisition unit 231 acquires sound information. The sound information acquisition unit 231 can also acquire sound information from abdominal sounds already received by the receiving unit 22, and can also acquire sound information already received by the receiving unit 22. The sound information acquisition unit 231 performs the same function as the sound information acquisition unit 13. The sound information acquisition unit 231 can also acquire sound information already received by the receiving unit 22.

[0093] The prediction unit 232 applies the learning information to the sound information already obtained by the sound information acquisition unit 231 and obtains menstrual-related information.

[0094] The prediction unit 232 obtains menstrual-related information by using the sound information already obtained by the sound information acquisition unit 231 and the learning information stored in the learning information storage unit 211.

[0095] For example, the prediction unit 232 provides the machine learning prediction module with the sound information already acquired by the sound information acquisition unit 231 and the learning information stored in the learning information storage unit 211, executes the module, and obtains menstrual-related information. Furthermore, as described above, the machine learning algorithm can be deep learning, decision tree, random forest, SVM, SVR, etc., but is not limited thereto, and the learning processing and prediction processing are the same.

[0096] The prediction unit 232, for example, obtains learning information from the learning information storage unit 211 that corresponds to a user identifier of the voice information already obtained by the voice information acquisition unit 231, applies this learning information to the voice information already obtained by the voice information acquisition unit 231, and obtains menstrual-related information. That is, it is preferable that the prediction unit 232 uses user-specific learning information to obtain menstrual-related information. However, the prediction unit 232 may also use learning information common to two or more users or all users to obtain menstrual-related information.

[0097] For example, the prediction unit 232 obtains learning information from the learning information storage unit 211, which is an identifier corresponding to the type of menstrual-related information to be obtained, and applies the learning information to the sound information already obtained by the sound information acquisition unit 231 to obtain the menstrual-related information of that type.

[0098] For example, the prediction unit 232 obtains learning information from the learning information storage unit 211, which corresponds to the identifier of the type of menstrual-related information to be obtained and the user identifier, and applies the learning information to the voice information already obtained by the voice information acquisition unit 231 to obtain the menstrual-related information.

[0099] The prediction unit 232, for example, uses sound information and a learner to perform prediction processing through machine learning algorithms to obtain menstrual-related information.

[0100] For example, the prediction unit 232 selects the sound information that is closest to the sound information from the correspondence table, and obtains the menstrual association information that is paired with the selected sound information from the correspondence table.

[0101] For example, the prediction unit 232 selects from a correspondence table two or more pieces of audio information that have been acquired by the audio information acquisition unit 231 and that are more similar (e.g., similarity is above a threshold) the more they meet predetermined conditions. It then retrieves two or more menstrual-related information pieces corresponding to each of the selected two or more audio information pieces from the correspondence table, and obtains one menstrual-related information piece from these two or more menstrual-related information pieces. The prediction unit 232, for example, obtains representative values ​​(e.g., average, median, or value chosen according to the majority opinion) of these two or more menstrual-related information pieces.

[0102] The output unit 24 outputs the menstrual-related information obtained by the prediction unit 232. Here, the output is usually a message sent to the terminal device 3, but it can also include concepts such as display on a monitor, projection using a projector, typing using a printer, sound output, accumulation to an external storage medium, and transmission of processing results to other processing devices or other programs.

[0103] Various types of information are stored in the terminal storage unit 31 that constitutes the terminal device 3. These various types of information include, for example, a user identifier. The user identifier may also be the ID of the terminal device 3, etc.

[0104] The terminal receiving unit 32 receives various information and instructions. These information and instructions include, for example, abdominal sounds, menstrual-related information, and output instructions. The input method for these information and instructions can be any of a microphone, touchscreen, keyboard, mouse, menu screen, etc.

[0105] The terminal processing unit 33 performs various processes. These processes include, for example, A / D conversion of the abdominal sounds received by the terminal receiving unit 32 to generate abdominal sound data to be transmitted. Other processes include processing the instructions and information received by the terminal receiving unit 32 into data to be transmitted. Additionally, various processes include processing the information received by the terminal receiving unit 35 into data to be output.

[0106] The terminal transmitter 34 sends various information and instructions to the learning device 1 or the menstrual-related information output device 2. These various information and instructions include, for example, abdominal sounds, menstrual-related information, and output instructions.

[0107] The terminal receiving unit 35 receives various information from the menstrual-related information output device 2. This information includes, for example, menstrual-related information.

[0108] The terminal output unit 36 ​​outputs various types of information, such as menstrual-related information. Preferably, the terminal output unit 36 ​​outputs menstrual-related information for each type of menstrual-related information.

[0109] While it is preferred that the teaching data storage unit 11, storage unit 21, learning information storage unit 211 and terminal storage unit 31 be non-volatile storage media, it is also possible to achieve this using volatile storage media.

[0110] The process of storing information in the teaching data storage unit 11 is not limited. For example, information can be stored in the teaching data storage unit 11 by means of a storage medium, information can be transmitted by means of a communication line, or information can be input by means of an input device.

[0111] Typically, the sound information acquisition unit 13, teaching data composition unit 15, learning unit 16, accumulation unit 17, processing unit 23, sound information acquisition unit 231, prediction unit 232, and terminal processing unit 33 can be implemented using a processor or memory. The processing sequence of the sound information acquisition unit 13, etc., is usually implemented in software, which is recorded on a storage medium such as ROM. However, it can also be implemented in hardware (dedicated circuitry). Furthermore, the processor can be, for example, a CPU, MPU, GPU, etc., and its type is not limited.

[0112] For example, the learning receiving unit 14, receiving unit 22, and terminal receiving unit 35 can be implemented through wireless or wired communication methods.

[0113] For example, the output unit 24 and the terminal transmitter 34 can be implemented through wireless or wired communication methods.

[0114] The terminal receiving unit 32 can be implemented through device drivers or menu screen control software, which can be input via a microphone, touch screen, or keyboard.

[0115] The terminal output unit 36 ​​can be considered to include output devices such as displays or speakers, or it can be considered not to include them. The terminal output unit 36 ​​can be implemented through driver software for the output devices, or through the driver software for the output devices and the output devices themselves.

[0116] Next, we will explain the action examples of information system A. First, using... Figure 4 The flowchart illustrates the operation of the learning device 1.

[0117] (Step S401) The learning receiving unit 14 determines whether abdominal sounds, etc., have been received from the terminal device 3. If abdominal sounds, etc., have been received, the process proceeds to step S402; otherwise, it proceeds to step S403. Furthermore, abdominal sounds, etc., may include, for example, abdominal sounds and menstrual-related information. Examples of abdominal sounds include, for example, abdominal sounds, menstrual-related information, and a user identifier. Moreover, the learning receiving unit 14 does not need to receive both abdominal sounds and menstrual-related information simultaneously. It is sufficient that the abdominal sounds correspond to the menstrual-related information.

[0118] Furthermore, the learning receiving unit 14 can also receive teaching data. In this case, the teaching data constructing unit 15 accumulates the received teaching data in the teaching data storage unit 11. The learning receiving unit 14 can also receive teaching data in association with a user identifier.

[0119] (Step S402) The sound information acquisition unit 13 acquires sound information from the abdominal sound received in step S401. Then, the teaching data construction unit 15 constructs teaching data containing the sound information and the received menstrual-related information. Next, the teaching data construction unit 15 associates the teaching data with a user identifier and accumulates it in the teaching data storage unit 11. Returning to step S401.

[0120] (Step S403) The learning unit 16 determines whether it is an opportunity to constitute learning information. If it is an opportunity to constitute learning information, it proceeds to step S404; if it is not an opportunity to constitute learning information, it returns to step S401.

[0121] Furthermore, the learning unit 16 can determine the timing for constructing learning information based on instructions from the terminal device 3. Additionally, the learning unit 16 can determine the timing for constructing learning information when teaching data exceeding a threshold exists in the teaching data storage unit 11. Furthermore, the learning unit 16 can determine the timing for constructing learning information corresponding to a user identifier when teaching data corresponding to a user identifier exceeds a threshold. Furthermore, the learning unit 16 can determine the timing for constructing learning information when teaching data on predetermined irregular dates exists in the teaching data storage unit 11 during the menstrual cycle. Additionally, the learning unit 16 can determine the timing for constructing learning information when, during the menstrual cycle, the dates (any dates in the cycle) of multiple teaching data corresponding to a user identifier satisfy a predetermined irregularity condition. Furthermore, “predetermined irregularity” refers to irregularity in the dates of the menstrual cycle, such as teaching data for different dates from the start date of menstruation to the start date of the next menstruation having a threshold (e.g., more than 15 days) or more than a threshold (e.g., more than 18 days).

[0122] (Step S404) The learning unit 16 substitutes 1 into the counter i.

[0123] (Step S405) The learning unit 16 determines whether there is an i-th user identifier that constitutes the learning information. If the i-th user identifier exists, proceed to step S406; if the i-th user identifier does not exist, return to step S401.

[0124] (Step S406) The learning unit 16 substitutes 1 into the count j.

[0125] (Step S407) The learning unit 16 determines whether there is menstrual-related information of the j-th type that constitutes the learning information. If there is menstrual-related information of the j-th type, proceed to step S408; if there is no menstrual-related information of the j-th type, proceed to step S412.

[0126] (Step S408) The learning unit 16 obtains from the teaching data storage unit 11 one or more teaching data pairs that are paired with the i-th user identifier and contain menstrual-related information of the j-th type.

[0127] (Step S409) The learning unit 16 uses one or more teaching data points acquired in step S408 to construct learning information. Figure 5 , Figure 6 The flowchart illustrates an example of how such learning information is processed.

[0128] (Step S410) The accumulation unit 17 accumulates the learning information obtained in step S407 by associating the i-th user identifier with the j-th type identifier. Furthermore, the accumulation location of the learning information can be either the learning device 1 or the learning information storage unit 211 of the menstrual-related information output device 2.

[0129] (Step S411) The learning unit 16 increments the count j by 1. Return to step S407.

[0130] (Step S412) The learning unit 16 increments the count i by 1. Return to step S405.

[0131] Furthermore, in Figure 4 In the flowchart, learning information is constructed for each user identifier. However, common learning information can also be constructed for two or more users.

[0132] In addition, Figure 4 In the flowchart, when there is only one type of menstrual-related information, the learning information does not correspond to the type identifier of the menstrual-related information.

[0133] Moreover, in Figure 4 In the flowchart, the process ends by power-off or by embedding the end of the process.

[0134] Next, use Figure 5 The flowchart below illustrates the first example of the learning information formation process in step S409. The first example is the case where the learner, i.e., the learning information, is obtained through machine learning processing.

[0135] (Step S501) The learning unit 16 determines whether to construct a learner for multivariate classification, i.e., to learn the information. If a learner for multivariate classification is constructed, the process proceeds to step S502; if a learner for binary classification is constructed, the process proceeds to step S504. Furthermore, the choice between multivariate classification and binary classification can be predetermined, or determined by the learning unit 16 based on the amount of teaching data being processed. For example, if the amount of teaching data being processed is above or greater than a threshold, the learning unit 16 determines it to be "binary classification"; if the amount of teaching data is below or less than the threshold, it determines it to be "multivariate classification".

[0136] (Step S502) The learning unit 16 provides the machine learning learning module with one or more teaching data obtained in step S408 and executes the learning module.

[0137] (Step S503) The learning unit 16 obtains the execution result of the module in step S502, i.e., the learner. It then returns to the host processing unit.

[0138] (Step S504) The learning unit 16 substitutes 1 into the counter i.

[0139] (Step S505) The learning unit 16 determines whether the i-th category exists. If the i-th category exists, proceed to step S506; if the i-th category does not exist, return to the previous processing step. Furthermore, the category is candidate data for menstrual-related information. For example, the category could be "menstrual start date is approaching" or "menstrual start date is far away".

[0140] (Step S506) The learning unit 16 obtains one or more teaching data points (positive examples) corresponding to the i-th category from the one or more teaching data points obtained in step S408. In addition, the learning unit 16 obtains one or more teaching data points (negative examples) that do not correspond to the i-th category from the one or more teaching data points obtained in step S406.

[0141] (Step S507) The learning unit 16 provides the teaching data of positive and negative examples obtained in step S506 to the machine learning learning module and executes the learning module.

[0142] (Step S508) The learning unit 16 is associated with the category identifier of the i-th category to obtain the execution result of the module in step S507, i.e., the learner.

[0143] (Step S509) The learning unit 16 increments the count i by 1. Return to step S505.

[0144] Next, use Figure 6The flowchart below illustrates a second example of the learning information processing in step S409. This second example involves obtaining the correspondence table, i.e., the learning information.

[0145] (Step S601) The learning unit 16 substitutes 1 into the counter i.

[0146] (Step S602) The learning unit 16 determines whether the i-th category exists. If the i-th category exists, proceed to step S603; if the i-th category does not exist, proceed to step S606.

[0147] (Step S603) The learning unit 16 acquires one or more teaching data points corresponding to the i-th category. That is, the learning unit 16 acquires, for example, one or two or more audio information points corresponding to the i-th category, and acquires representative values ​​of these audio information points (e.g., the average value, median value, or vector of the majority opinion of each feature). Next, the learning unit 16 acquires teaching data with the acquired representative values ​​and the i-th category.

[0148] (Step S604) The learning unit 16 uses one or more teaching data points obtained in step S603 to construct the i-th correspondence information. Furthermore, the correspondence information is information that maps sound information to menstrual-related information (category data).

[0149] (Step S605) The learning unit 16 increments the count i by 1. Return to step S602.

[0150] (Step S606) The learning unit 16 constructs a correspondence table containing two or more corresponding information items that were constructed in step S604. Return to the host computer for further processing.

[0151] Next, use Figure 7 The flowchart illustrates the operation of the menstrual-related information output device 2.

[0152] (Step S701) The receiving unit 22 determines whether an output instruction has been received from the terminal device 3. If an output instruction has been received, the process proceeds to step S702; otherwise, it returns to step S701. The output instruction may include, for example, abdominal sounds and a user identifier. The output instruction may also include sound information and a user identifier.

[0153] (Step S702) The sound information acquisition unit 231 acquires sound information from the abdominal sound contained in the output indication received in step S701.

[0154] (Step S703) The prediction unit 232 uses the sound information obtained in step S702 to perform prediction processing to obtain menstrual-related information. Figure 8The flowchart illustrates an example of prediction processing.

[0155] (Step S704) Output unit 24 sends the menstrual-related information obtained in step S703 to terminal device 3. Return to step S701.

[0156] Next, use Figure 8 The flowchart illustrates the first example of the prediction process in step S703.

[0157] (Step S801) Prediction unit 232 obtains the user identifier corresponding to the received abdominal sound.

[0158] (Step S802) The prediction unit 232 substitutes 1 into the count i.

[0159] (Step S803) Prediction unit 232 determines whether the i-th category exists. If the i-th category exists, proceed to step S804; if the i-th category does not exist, proceed to step S808.

[0160] (Step S804) The prediction unit 232 obtains the user identifier and the learner corresponding to the i-th category obtained in step S801 from the learning information storage unit 211.

[0161] (Step S805) The prediction unit 232 gives the learner information obtained in step S805 and the sound information obtained in step S702 to the prediction processing module for machine learning, and executes the module.

[0162] (Step S806) The prediction unit 232 obtains the execution result of the module in step S805, namely the prediction result and the score. Furthermore, the prediction result here indicates whether it belongs to the i-th category.

[0163] (Step S807) Prediction unit 232 increments count i by 1. Return to step S803.

[0164] (Step S808) The prediction unit 232 obtains menstrual-related information using the prediction results and scores obtained in step S806. It then returns to the host processing unit.

[0165] Furthermore, the prediction unit 232 obtains, for example, the category identifier of the category with the highest score that is the prediction result obtained in step S806 that is "belonging to the i-th category" as menstrual association information.

[0166] Furthermore, in Figure 8The flowchart also allows for predictive processing for each type of menstrual-related information. Types of menstrual-related information include, for example, information indicating whether the start date of menstruation is approaching, information indicating whether menstruation is currently underway, information indicating the number of days since the start date of menstruation, information indicating the number of days since ovulation, information indicating the length of the menstrual period, information indicating whether the next menstrual cramps will be weak or strong, and information indicating the level of pain.

[0167] In addition, Figure 8 In the flowchart, a second prediction process can also be performed. That is, the prediction unit 232 obtains a learner from the learning information storage unit 211 that corresponds to the user identifier of the received abdominal sound and is capable of multivariate classification. Next, the prediction unit 232 gives the learner and the sound information obtained in step S702 to the prediction processing module that performs machine learning, executes the module, and obtains menstrual-related information.

[0168] In addition, Figure 8 In the flowchart, a third prediction process can also be performed. That is, the prediction unit 232 obtains a correspondence table from the learning information storage unit 211 corresponding to the user identifier of the received abdominal sound. Next, the prediction unit 232 determines the sound information (e.g., a vector) that is closest to the sound information (e.g., a vector) obtained in step S702 from the correspondence table. Next, the prediction unit 232 obtains the menstrual association information paired with the closest sound information from the correspondence table.

[0169] Furthermore, in Figure 8 In the flowchart, the prediction unit 232 can also perform prediction processing using learning information shared by two or more users (a learner capable of multivariate classification, a learner capable of binary classification for each category, or a correspondence table).

[0170] Next, use Figure 9 The flowchart illustrates the operation of terminal device 3.

[0171] (Step S901) The terminal receiving unit 32 determines whether abdominal sounds, etc., have been received. If abdominal sounds, etc., have been received, the process proceeds to step S902; if abdominal sounds, etc., have not been received, the process proceeds to step S904. Furthermore, abdominal sounds, etc., may include, for example, abdominal sounds and menstrual-related information.

[0172] (Step S902) The terminal processing unit 33 constructs information to be sent to the learning device 1 using abdominal sounds, etc. That is, the terminal processing unit 33 obtains a user identifier from the terminal storage unit 31, for example. The terminal processing unit 33 performs A / D conversion on the abdominal sounds collected through the microphone. The terminal processing unit 33 constructs information to be sent, containing the A / D converted abdominal sound data, menstrual-related information, and the user identifier information.

[0173] (Step S903) The terminal transmitting unit 34 sends the information that has been generated in step S902 to the learning device 1.

[0174] (Step S904) The terminal receiving unit 32 determines whether an output instruction containing abdominal sounds has been received. If an output instruction has been received, the process proceeds to step S905; if no output instruction has been received, the process returns to step S901.

[0175] (Step S905) The terminal processing unit 33 configures an output instruction to be sent. That is, the terminal processing unit 33 obtains a user identifier from the terminal storage unit 31, for example. The terminal processing unit 33 performs A / D conversion on the abdominal sound. The terminal processing unit 33 configures an output instruction having the A / D converted abdominal sound data and the user identifier.

[0176] Furthermore, the terminal processing unit 33 can also obtain sound information from the abdominal sound and form an output instruction having the sound information and the user identifier.

[0177] (Step S906) The terminal transmitting unit 34 sends the output instruction formed in step S905 to the menstrual-related information output device 2.

[0178] (Step S907) The terminal receiving unit 35 determines whether one or more types of menstrual-related information have been received based on the output instruction sent in step S906. If menstrual-related information has been received, proceed to step S908; if no menstrual-related information has been received, return to step S907.

[0179] (Step S908) The terminal processing unit 33 uses the menstrual-related information received in step S907 to construct the menstrual-related information to be output. The terminal output unit 36 ​​outputs the menstrual-related information. Return to step S901.

[0180] Furthermore, in Figure 9 In the flowchart, the process ends by power-off or by embedding the end of the process.

[0181] The following describes a specific example of the operation of information system A in this embodiment. A schematic diagram of information system A is shown below. Figure 1 .

[0182] At this time, the teaching data storage unit 11 of the learning device 1 has been filled with data containing... Figure 10 The teaching data management table shown is a structured table that manages one or more records with "ID", "user identifier", "voice information", "date and time information", and "menstrual association information". Here, "menstrual association information" includes "menstrual flag", "number of days", "period information", and "level".

[0183] Here, "voice information" refers to a set of two or more feature quantities, i.e., a feature vector, from which abdominal sounds have been acquired. "Date and time information" refers to the date and time corresponding to the voice information. This "date and time information" can be the date and time the abdominal sounds were acquired, the date and time the learning device 1 received the abdominal sounds or voice information, or the date and time the terminal device 3 transmitted the abdominal sounds or voice information, etc. The "date and time information" includes date information that determines the date. Furthermore, "date and time information" can also be simply date information.

[0184] The "Menstrual Flag" indicates whether you are currently menstruating. It is set to "1" if you are menstruating and "0" if you are not. The "Days Information" indicates the number of days until the start date of your next menstrual period. The "Period Information" indicates the number of days in your menstrual period. If you are menstruating, the "Days Information" indicates the duration of your current menstrual period; otherwise, it indicates the duration of your next menstrual period. The "Level" indicates the level of menstrual pain and is a value entered by the user.

[0185] Furthermore, the teaching data composition unit 15 acquires "day information" as shown below. That is, the teaching data composition unit 15 acquires a user identifier corresponding to the voice information already acquired by the voice information acquisition unit 13. The teaching data composition unit 15 acquires teaching data paired with this user identifier, and the first date information is the date and time information paired with teaching data containing the menstrual flag "1" and the day information "28" (teaching data for the start date of menstruation). Additionally, the teaching data composition unit 15 acquires second date information (second date information < first date information) corresponding to the date and time information of the voice information already acquired by the voice information acquisition unit 13. Next, the teaching data composition unit 15 acquires the difference between the first date information and the second date information as "day information." Here, the menstrual cycle is taken as 28 days.

[0186] Furthermore, the teaching data composition unit 15 acquires "period information" as shown below. That is, the teaching data composition unit 15 acquires a user identifier corresponding to the voice information acquired by the voice information acquisition unit 13. The teaching data composition unit 15 acquires the voice information paired with the user identifier, and the first date information is formed by date and time information paired with the voice information of the menstrual start date. The teaching data composition unit 15 acquires the date and time information paired with the user identifier, and the date and time information representing a date later than the first date information, and the date and time information representing the date closest to the first date information, and the second date information is formed by date and time information paired with the menstrual flag "0". The teaching data composition unit 15 calculates the "period information" using the formula "period information = second date information - first date information". Then, the teaching data composition unit 15 accumulates this "period information" as an attribute value that is paired with the user identifier, paired with date and time information having a date earlier than the first date information, and where the "period information" is NULL. In other words, when the teaching data composition unit 15 determines the period information representing the menstrual period, it substitutes the calculated "period information" into the teaching data that is present from the end of the last menstrual period to the start date of the current menstrual period.

[0187] In this context, two specific examples will be explained. Specific example 1 illustrates the learning process of learning device 1. Specific example 2 illustrates the predictive processing of menstrual-related information in menstrual-related information output device 2.

[0188] Specific example 1

[0189] At this time, the user identified as "U02" activates the application (hereinafter, appropriately referred to as "application") on terminal device 3 in order to learn menstrual-related information. An example of the output of such an application is... Figure 11 .

[0190] against Figure 11 On the screen, the user selected the menstrual date associated information "Menstrual Start Date" and the pain information "3". The learning receiving unit 14 then receives this menstrual-related information.

[0191] In addition, the user presses Figure 11 By pressing the recording button 1101, the microphone 1102 of the terminal device 3 is brought close to the abdomen to collect sound information. In this way, the terminal receiver 32 of the terminal device 3 receives the sound from the abdomen.

[0192] Next, the user presses... Figure 11The send button 1103 is pressed. The terminal processing unit 33 then reads the user identifier "U02" from the terminal storage unit 31. Next, the terminal processing unit 33 obtains the menstrual-related information "<Menstrual Date Related Information> Menstrual Start Date <Pain Information> 3". Additionally, the terminal processing unit 33 digitizes the abdominal sound. Furthermore, the terminal processing unit 33 constructs information containing the menstrual-related information, the abdominal sound, and the user identifier "U02". Next, the terminal transmission unit 34 sends the constructed information to the learning device 1.

[0193] Next, the learning receiving unit 14 of the learning device 1 receives menstrual-related information, abdominal sounds, and user identifiers from the learning device 1.

[0194] Next, the teaching data composition unit 15 obtains the menstrual flag "1", the number of days "28", and the level "3" from the received menstrual-related information "<menstrual date related information> menstrual start date <pain information> 3". Next, the teaching data composition unit 15 obtains the date and time information "9 / 10 8:15" from the unillustrated clock. Additionally, the teaching data composition unit 15 obtains various characteristic quantities from abdominal sounds and constructs sound information (x... 981 x 982 , ..., x 98n Next, the teaching data composition unit 15 constructs a record to be accumulated in the teaching data management table. This record is... Figure 10 The record with "ID=99".

[0195] Through the accumulation and processing of teaching data as described above, multiple teaching data sets are accumulated for each user.

[0196] Next, as shown below, the learning unit 16 constructs a learner for each user and each menstrual-related information. Furthermore, when constructing a learner for each menstrual-related information, the learning unit 16 may or may not use other menstrual-related information. For example, when constructing a learner for outputting the "level" of menstrual-related information, the learning unit 16 may use teaching data containing one or more of the other menstrual-related information (here, "menstrual flag," "days information," and "period information") for learning processing, or it may use teaching data that does not contain other menstrual-related information for learning processing.

[0197] In other words, for each user, the learning unit 16 retrieves all teaching data from the teaching data management table, consisting of voice information paired with the user's identifier and monthly association information (e.g., "level"). Next, the learning unit 16 performs learning processing using a machine learning algorithm (e.g., random forest) to construct a learner that takes voice information as input and monthly association information (e.g., "level") as output. Then, the accumulation unit 17 accumulates the learning data obtained by the learning unit 16 by pairing the learner with the user identifier.

[0198] Furthermore, the learning unit 16 processes each user's menstrual-related information ("menstrual flag", "days information", "period information") in the same way as described above, and constructs a learner for each user's menstrual-related information. Next, the accumulation unit 17 accumulates information by pairing the learner obtained by the learning unit 16 with the user identifier.

[0199] Through the above processing, a learner is accumulated that corresponds to four identifiers: the user identifier and the identifier of the type of menstrual-related information for each user.

[0200] Specific example 2

[0201] Next, the user identified as "U02" used the application as shown below to predict the number of days until the next menstrual period, the duration of the next menstrual period, and the level of menstrual pain.

[0202] In other words, the user launches the application on terminal device 3 to predict menstrual-related information. An example of the output of such an application is... Figure 12 .

[0203] Next, the user presses... Figure 12 The recording button 1201 is pressed, and the microphone 1202 of the terminal device 3 is brought close to the abdomen to collect sound information. In this way, the terminal receiver 32 of the terminal device 3 receives the sound from the abdomen.

[0204] Next, the user presses... Figure 12 The send button 1203 is pressed. The terminal processing unit 33 then reads the user identifier "U02" from the terminal storage unit 31. Next, the terminal processing unit 33 digitizes the abdominal sound. Furthermore, the terminal processing unit 33 generates an output instruction containing the abdominal sound and the user identifier "U02". Next, the terminal transmission unit 34 sends this output instruction to the menstrual-related information output device 2.

[0205] Next, the receiving unit 22 of the menstrual-related information output device 2 receives the output instruction. Then, the sound information acquisition unit 231 acquires sound information, i.e., the feature vector, from the abdominal sound contained in the received output instruction.

[0206] Next, the prediction unit 232 uses the acquired sound information to perform prediction processing to obtain menstrual-related information as shown below.

[0207] In other words, the prediction unit 232 obtains the user identifier "U02" from the output instruction. Next, the prediction unit 232 obtains the learner corresponding to the user identifier "U02" and the "menstrual flag" from the learning information storage unit 211. Next, the prediction unit 232 gives the learner and the sound information, i.e., the feature vector, to the machine learning module (e.g., a random forest module), executes the module, and obtains the menstrual flag "0".

[0208] Additionally, the prediction unit 232 retrieves the learner corresponding to the user identifier "U02" and the "day information" from the learning information storage unit 211. Next, the prediction unit 232 provides the learner and the sound information, i.e., the feature vector, to the machine learning module (e.g., the deep learning module), executes the module, and obtains the day information "3".

[0209] Additionally, the prediction unit 232 retrieves the learner corresponding to the user identifier "U02" and "period information" from the learning information storage unit 211. Next, the prediction unit 232 provides the learner and the sound information, i.e., the feature vector, to the machine learning module (e.g., the SVM module), executes the module, and retrieves the period information "4.5".

[0210] Furthermore, the prediction unit 232 retrieves the learner corresponding to the user identifier "U02" and "level" from the learning information storage unit 211. Next, the prediction unit 232 provides the learner and the sound information, i.e., the feature vector, to the machine learning module (e.g., the random forest module), executes the module, and obtains the level "3".

[0211] Next, the prediction unit 232 uses the menstrual flag "0", the number of days "3", the period information "4.5", and the level "3" to construct the menstrual correlation information to be sent. Then, the output unit 24 sends the constructed menstrual correlation information to the terminal device 3.

[0212] Next, according to the output instruction, the terminal receiving unit 35 of the terminal device 3 receives menstrual-related information. Next, the terminal processing unit 33 uses the received menstrual-related information to construct the menstrual-related information to be output. The terminal output unit 36 ​​outputs this menstrual-related information. Such an output example is... Figure 13 .

[0213] According to this embodiment, menstrual-related information can be obtained using abdominal sounds.

[0214] Furthermore, in this embodiment, when creating the learner, the learning device 1 can employ different machine learning algorithms depending on the type of menstrual-related information. For example, the learning device 1 can use a random forest module when constructing a learner for outputting a "menstrual flag," a deep learning module when constructing a learner for outputting "day information," and an SVR module when constructing a learner for outputting "period information," etc.

[0215] In addition, in this embodiment, the learning device 1 can also be a standalone device. In such a case, the learning device 1 includes: a sound collection unit that collects abdominal sounds from or around the abdomen of a user; a sound information acquisition unit that acquires sound information from the abdominal sounds; a learning receiving unit that receives menstrual-related information; a teaching data construction unit that constructs teaching data from the sound information and the menstrual-related information; a learning unit that performs machine learning processing on the teaching data constructed by the teaching data construction unit to construct a learner, i.e., learning information; and an accumulation unit that accumulates the learner.

[0216] In addition, in this embodiment, the menstrual-related information output device 2 can also be a stand-alone device. In such a case, the menstrual-related information output device 2 includes: a learning information storage unit that stores learning information consisting of two or more teaching data points, including sound information obtained from abdominal sounds received from a user's abdomen or surrounding area and menstrual-related information associated with menstruation; a sound information acquisition unit that acquires sound information used for predictive processing of menstrual-related information from sounds received from the user's abdomen or surrounding area; a prediction unit that applies the learning information to the sound information acquired by the sound information acquisition unit and acquires menstrual-related information; and an output unit that outputs the menstrual-related information acquired by the prediction unit.

[0217] In addition, in this embodiment, the menstrual-related information output device 2 may also have a structure that includes the learning device 1.

[0218] Furthermore, the processing in this embodiment can also be implemented using software. Moreover, the software can be distributed via software download or other means. Additionally, the software can be recorded on a storage medium such as a CD-ROM and then distributed. This is also true in other embodiments described in this specification. Furthermore, the software implementing the learning device 1 in this embodiment is a program that enables the computer to function as a sound information acquisition unit, a learning receiving unit, a teaching data composition unit, a learning unit, and an accumulation unit. The sound information acquisition unit acquires sound information from the user's abdominal voice; the learning receiving unit receives menstrual-related information; the teaching data composition unit composes teaching data from the sound information and the menstrual-related information; the learning unit performs machine learning processing on the teaching data composed by the teaching data composition unit to form a learner, i.e., learning information; and the accumulation unit accumulates the learner.

[0219] Furthermore, the software for implementing the menstrual-related information output device 2 is a program as follows. That is, this program enables a computer connected to a learning information storage unit, which stores learning information consisting of two or more teaching data points—including sound information obtained from the user's abdominal voice and menstrual-related information—to function as a sound information acquisition unit, a prediction unit, and an output unit. The sound information acquisition unit acquires sound information from the user's abdominal voice; the prediction unit applies the learning information to the sound information acquired by the sound information acquisition unit and acquires menstrual-related information; and the output unit outputs the menstrual-related information acquired by the prediction unit.

[0220] in addition, Figure 14 This describes the appearance of the computer that executes the programs described in this specification to implement the various embodiments described above, including the learning device 1, the menstrual-related information output device 2, and the terminal device 3. The above embodiments can be implemented using computer hardware and computer programs executed thereon. Figure 14 This is an external view of the computer system 300. Figure 15 This is a block diagram of system 300.

[0221] Figure 14 In the computer system 300, there are computer 301 with CD-ROM drive, keyboard 302, mouse 303, monitor 304, and microphone 305.

[0222] Figure 15In addition to the CD-ROM drive 3012, the computer 301 also includes: an MPU 3013; a bus 3014 connected to the CD-ROM drive 3012, etc.; a ROM 3015 for storing programs such as boot programs; a RAM 3016 connected to the MPU 3013 and used to provide temporary storage space while temporarily storing application program commands; and a hard disk 3017 for storing application programs, system programs, and data. Although not shown here, the computer 301 may also include a network card providing a connection to a LAN.

[0223] In computer system 300, the program that performs the functions of the learning device 1, etc., as described above, can also be stored in CD-ROM 3101 and further transferred to hard disk 3017 by being inserted into CD-ROM drive 3012. Alternatively, the program can also be sent to computer 301 and stored in hard disk 3017 via a network (not shown). The program is loaded into RAM 3016 during execution. The program can also be loaded directly from CD-ROM 3101 or via a network.

[0224] The program may not necessarily include an operating system (OS) or third-party programs that enable the computer 301 to perform the functions of the learning device 1 described above. It is sufficient that the program invokes the appropriate functions (modules) in a controlled manner and contains only the command portion that yields the desired result. Since the operation of the computer system 300 is already well understood, detailed explanations are omitted.

[0225] Furthermore, the above procedure does not include hardware-based processing in the steps of sending or receiving information, such as processing via a modem or interface circuit board in the sending step (processing that can only be performed by hardware).

[0226] Furthermore, the computer executing the above program can be a single unit or multiple units. That is, it can perform centralized processing or distributed processing. In other words, the information processing device 5 can be a standalone device or it can be composed of two or more devices.

[0227] Furthermore, in the above embodiments, it goes without saying that two or more communication methods existing in a device can be physically implemented through a single medium.

[0228] Furthermore, in the above embodiments, each process can be implemented either through centralized processing by a single device or through decentralized processing by multiple devices.

[0229] The present invention is not limited to the above-described embodiments, but can be modified in various ways, and these modifications are also included within the scope of the present invention.

[0230] Industrial applicability

[0231] As described above, the menstrual-related information output device of the present invention has the effect of predicting menstrual-related information by utilizing abdominal sounds from the abdomen or the periphery of the abdomen, and is useful as a device for outputting menstrual-related information.

Claims

1. A menstrual-related information output device, characterized in that, It has: a learning information storage unit that stores learning information consisting of two or more teaching data, including sound information obtained from the user's abdominal voice and menstrual-related information. The sound information acquisition unit acquires sound information from the user's abdominal sounds; The prediction unit applies the learned information to the sound information already acquired by the sound information acquisition unit and obtains menstrual-related information; The prediction unit outputs menstrual-related information obtained by the prediction unit.

2. The menstrual-related information output device according to claim 1, characterized in that, The menstrual association information refers to menstrual date relationship information concerning the relationship between dates related to menstruation.

3. The menstrual-related information output device according to claim 1, characterized in that, The menstrual-related information refers to pain information related to menstrual cramps.

4. The menstrual-related information output device according to claim 1, characterized in that, The two or more teaching data consist of teaching data containing sound information obtained from abdominal sounds and menstrual-related information, the abdominal sounds being obtained from the user's abdomen daily between menstrual cycles.

5. The menstrual-related information output device according to claim 1, characterized in that, It also has a learning unit, which uses machine learning algorithms to process the two or more teaching data sets to obtain learning information. The prediction unit uses the sound information already obtained by the sound information acquisition unit and the learning information to perform prediction processing through a machine learning algorithm to obtain menstrual-related information.

6. The menstrual-related information output device according to claim 1, characterized in that, The sound information consists of two or more features of the user's abdominal voice.

7. A learning device, characterized in that, Equipped with: a sound information acquisition unit to acquire sound information from the user's abdominal sounds; The learning reception department receives menstrual-related information. The teaching data composition unit composes teaching data from the sound information and the menstrual-related information; The learning unit performs machine learning processing on the teaching data already constructed by the teaching data construction unit to construct a learner, i.e., learning information. And the accumulation section, which accumulates the learner.

8. A method for generating learning information, implemented through a sound information acquisition unit, a learning reception unit, a teaching data composition unit, a learning unit, and an accumulation unit, characterized in that: It includes: a sound information acquisition step, wherein the sound information acquisition unit acquires sound information from the user's abdominal voice; The learning and receiving step involves the learning and receiving unit receiving menstrual-related information; The teaching data composition step involves the teaching data composition unit composing teaching data from the audio information and the menstrual-related information. In the learning step, the learning unit performs machine learning processing on the teaching data that has been constructed in the teaching data construction step to construct a learner, i.e., learning information. The accumulation step involves accumulating the learner.

9. A storage medium characterized in that it records a program for enabling a computer connectable to a learning information storage unit to function as a sound information acquisition unit, a prediction unit, and an output unit. The learning information storage unit stores learning information composed of two or more teaching data, including sound information obtained from the user's abdominal voice and menstrual-related information. The sound information acquisition unit obtains sound information from the user's abdominal sounds. The prediction unit applies the learned information to the sound information already acquired by the sound information acquisition unit and obtains menstrual-related information. The output unit outputs the menstrual correlation information obtained by the prediction unit.

10. A storage medium, characterized in that, The records contain programs for enabling the computer to function as a sound information acquisition unit, a learning reception unit, a teaching data composition unit, a learning unit, and an accumulation unit. The sound information acquisition unit obtains sound information from the user's abdominal sounds. The learning reception department receives menstrual-related information. The teaching data composition unit composes teaching data from the sound information and the menstrual correlation information. The learning unit performs machine learning processing on the teaching data already constructed by the teaching data construction unit to construct a learner, i.e., learning information. The accumulation unit accumulates the learner.

Images