JP2023072757A - Episode information storage/output apparatus, episode evocation system, and episode evocation program - Google Patents

Abstract

To provide an episode information storage/output apparatus which can simply and quickly present a forgotten episode of a user to the user, an episode evocation system, and an episode evocation program.SOLUTION: An episode information storage/output apparatus includes: an episode information receiving unit 11 which receives information on a past episode of a user; an episode information storage unit 12 which stores the episode information in association with a keyword related to the episode information; and an episode information output unit 13 which reads the episode information stored in the episode information storage unit 12 to output the information. In outputting predetermined episode information, the episode information output unit 13 simultaneously outputs episode information related to the predetermined episode information on the basis of a keyword associated with the predetermined episode information.SELECTED DRAWING: Figure 1

Description

Patent Law Article 30, Paragraph 2 Application Applied August 19, 2021 Apple Store https://apps. apple. com/jp/app/%E3%82%A8%E3%83%94%E3%82%BD%E3%83%BC%E3%83%89%E3%81%AE%E6%B3%89/id1570225293

The present invention relates to an episode information storage output device, an episode recall system, and an episode recall program for storing and outputting past episodes.

Conventionally, when you want to keep track of things that happen around you, things that you are interested in, or things you want to remember so that you don't forget them, you can record them in written form such as a diary or memo, or use application software. and stored in computers, smartphones, etc.

In particular, in the aging society of recent years, there is a strong demand for a support tool that draws out personal memories that have been forgotten due to aging.

As a device for recording personal matters, for example, Patent Document 1 proposes a device that improves convenience by recording using application software and recording in a preset format. It is

JP-A-2000-331089

Here, as described above, when reading data recorded in a computer or the like using application software, there is a method of reading desired data by, for example, a user specifying a search word and searching. Only data containing search words specified by the user can be extracted.

However, it is very difficult to set search words so that events that have already been forgotten are also searched. That is, since it is difficult for the user to consciously specify a search word or the like for an event that has already been forgotten, it is difficult to read it.

SUMMARY OF THE INVENTION In view of the above problems, the present invention provides an episode information storage output device, an episode recall system, and an episode recall program that can easily and quickly present to the user episodes that the user has already forgotten. With the goal.

The episode information storage and output device of the present invention includes an episode information reception unit that receives information on a user's past episodes, an episode information storage unit that associates and stores episode information and keywords related to the episode information, an episode information output unit for reading and outputting the information of the episode stored in the episode information storage unit, and when the episode information output unit outputs the information of the predetermined episode, the information is associated with the information of the predetermined episode. Based on the keyword, episode information related to predetermined episode information is output.

Further, the episode information storage/output device of the present invention can include a keyword information extractor for extracting keywords from episode information.

Further, in the episode information storage output device of the present invention, when outputting the information of the predetermined episode, the episode information output unit identifies a keyword linked to the information of the predetermined episode, and Information on other episodes linked with keywords can be output as information on related episodes.

Further, in the episode information storage output device of the present invention, when there are multiple keywords linked to predetermined episode information, the episode information output unit outputs the user's multiple episodes stored in the episode information storage unit. Based on the frequency of each keyword linked to information, a specific keyword is identified from among multiple keywords, and information on other episodes linked with the same keyword as the identified keyword is displayed in related episodes. It can be output as information.

Further, in the episode information storage and output device of the present invention, the episode information output unit specifies a keyword whose frequency is higher than a preset rank based on the frequency of each keyword, and the keyword having the same keyword as the specified keyword is specified. Information on other linked episodes can be output as related episode information.

Further, in the episode information storage and output device of the present invention, the episode information output unit identifies keywords whose frequency is lower than a preset rank based on the frequency of each keyword, and the keyword having the same frequency as the identified keyword Information on other linked episodes can be output as related episode information.

Further, in the episode information storage and output device of the present invention, the episode information output unit identifies the keywords of the order whose frequency is preset based on the frequency of each keyword, and associates the same keyword with the identified keyword. Information on other episodes that have been recorded can be output as information on related episodes.

Further, in the episode information storage output device of the present invention, the episode information storage unit can classify and store episode information for each category set in advance.

Further, the episode information storage/output device of the present invention can include an episode information classification unit that automatically classifies episode information into categories based on keywords linked to episode information.

In the episode information storage output device of the present invention, the episode information related to the predetermined episode information can be episode information belonging to a different category.

Further, in the episode information storage output device of the present invention, the episode information output unit can output a category related to the predetermined episode information when outputting the predetermined episode information.

In the episode information storage output device of the present invention, the episode information can be an image.

Further, in the episode information storage output device of the present invention, the episode information output unit can automatically output the episode information based on the keyword linked to the episode information.

Further, the episode information storage output device of the present invention can include an episode summary creation section for creating a summary of the episode information when the episode information is text, and the episode information storage section stores the episode information. The summary can be stored as

Further, the episode information storage output device of the present invention can include a keyword output unit that outputs a keyword associated with the episode information stored in the episode information storage unit to an external device.

Further, in the episode information storage output device of the present invention, when information of a predetermined episode is output by the episode information output unit, the search unit searches the Internet based on a keyword linked to the information of the predetermined episode. , and the episode information output unit can output the information retrieved by the Internet search together with the information of the predetermined episode.

Further, the episode information storage/output device of the present invention can include an episode information edit receiving section that receives editing of episode information.

Further, in the episode information storage output device of the present invention, the episode information edit receiving section can receive a change or addition of a category to which episode information belongs.

Further, in the episode information storage output device of the present invention, the episode information edit receiving unit can receive addition or deletion of a keyword linked to episode information.

The episode recall system of the present invention comprises the episode information storage output device of the present invention, and a control unit for displaying predetermined episode information output from the episode information storage output device and episode information related to the predetermined episode information. and a terminal device having

Further, in the episode recall system of the present invention, the control unit can include a control unit that displays a plurality of icons in which the names of a plurality of preset categories are listed and displayed on the screen.

Further, in the episode recall system of the present invention, when one of the plurality of icons displayed on the screen is selected, the control unit of the terminal device belongs to the category described in the selected icon. Episode information and related episode information can be displayed.

Further, in the episode recollection system of the present invention, the control unit of the terminal device receives input of a predetermined keyword, displays the information of the episode linked to the received keyword, and displays the information of the displayed episode. Information on related episodes can be displayed.

Further, an episode recall program of the present invention is an episode recall program installed in a terminal device in the episode recall system of the present invention, and includes predetermined episode information and predetermined episode information output from an episode information storage output device. Cause the terminal device to perform the step of displaying information of the episode related to the information.

According to the episode information storage output device, episode recall system, and episode recall program of the present invention, past episode information of the user and keywords related to the episode information are associated and stored. Then, when reading and outputting the stored information of the episode, the information of the episode related to the information of the episode is output based on the keyword linked to the information of the episode.

As a result, it is possible to easily and quickly present the user's own episode that the user has already forgotten.

1 is a block diagram showing a schematic configuration of an embodiment of an episode system of the present invention; FIG. Flowchart for explaining the flow of processing for generating and storing episode information A diagram showing an example of an initial screen displayed on a terminal device A diagram showing an example of an episode type selection screen displayed on the terminal device Flowchart for explaining the flow of processing for displaying and editing episode information A diagram showing an example of a category selection screen displayed on a terminal device A diagram showing an example of an episode information list display screen A diagram showing a display example of information on a given episode and information on related episodes. A diagram showing an example of an edit content selection screen for episode information A diagram showing an example of keywords and categories associated with episode information A diagram showing the frequency of keywords linked to information on episodes shown in FIG. 10 and the frequency of categories into which episodes are classified. A diagram showing an example of the search information input screen A diagram showing an example of a keyword selection screen A diagram showing an example of a list display of information on searched episodes 1 is a block diagram showing a schematic configuration of a first modified example of an embodiment of an episode system of the present invention; FIG. A diagram showing an example of a summary template and keywords associated with each summary template Block diagram showing a schematic configuration of a second modification of one embodiment of the episode system of the present invention Block diagram showing a schematic configuration of a third modification of one embodiment of the episode system of the present invention

An embodiment of the episode recall system of the present invention will be described in detail below with reference to the drawings. FIG. 1 is a block diagram showing a schematic configuration of an episode recall system 1 of this embodiment.

The episode recall system 1 of this embodiment is a system that can easily and quickly present to the user his or her own episode that the user has forgotten.

The episode recall system 1 specifically includes an episode information storage output device 10 and a terminal device 20, as shown in FIG. The episode information storage output device 10 and the terminal device 20 are connected via a communication line such as the Internet or a LAN (Local Area Network), and are configured to be able to communicate with each other.

As shown in FIG. 1, the episode information storage/output device 10 includes an episode information reception unit 11, an episode information storage unit 12, an episode information output unit 13, a keyword extraction unit 14, an episode information classification unit 15, an episode information and an information editing reception unit 16 .

The episode information storage/output device 10 is composed of a computer, and includes a CPU (Central Processing Unit), a semiconductor memory, a hard disk, a communication I/F, and the like.

An embodiment of the episode information storage output program of the present invention is installed in the semiconductor memory or hard disk of the episode information storage output device 10 . The functions of the above sections are realized by the episode information storage output program being executed by the CPU provided in the episode information storage output device 10 . In the present embodiment, the functions of the above units are realized by the episode information storage output program as described above. , FPGA (Field-Programmable Gate Array), and other hardware such as electric circuits.

The episode information reception unit 11 receives information on the user's past episodes output from the terminal device 20, which will be detailed later. The episode information in this embodiment is, as described above, information such as events that occurred in the user's surroundings, matters that the user is interested in, and matters that the user wants to remember.

Past episode information includes, for example, text data created by the user using the terminal device 20, image data and moving image data captured by the camera function of the terminal device 20, Data stored in the device, etc.

The text data may be created by activating an episode recall program installed in the terminal device 20 and described later, or may be created by another application installed in the terminal device 20 . Applications other than the episode recall program include, for example, "notepad" and "reminder", but are not limited to these applications, such as schedule management applications, to-do management applications, and other user episode recording applications. Any application can be used.

The data saved in a device communicatively connected to terminal device 20 is, for example, data saved in a cloud server communicatively connected to terminal device 20, and may be text data, image data, or the like. It may be video data.

Also, if the terminal device 20 has a GPS (Global Positioning System) function, the episode information may include the user's movement trajectory obtained by the GPS function. For example, if the episode information is image data or moving image data, information on the shooting location may be acquired by the GPS function and included in the episode information.

Also, if the terminal device 20 has a calendar function, the episode information may include date information obtained by the calendar function. For example, if the episode information is image data or moving image data, information on the shooting date and time may be acquired by the calendar function and included in the episode information. If the episode information is text data, information on the date and time when the text data was created may be acquired by the calendar function and included in the episode information.

The episode information storage unit 12 associates and stores the information of the episode received by the episode information receiving unit 11 and the keyword related to the information of the episode.

Keywords related to episode information include place names such as "Yokohama", "Tokyo" and "Osaka", "ramen shops", "hospitals", "hotels", "pet shops", "bookstores" and "restaurants". Facility names such as ", pet type names such as "dog" and "cat", words that express user behavior itself such as "eating out", "travel", "driving" and "reading", "novel", " Words related to the user's hobbies such as "book" and "running", words representing food such as "cake", "ramen", "meat" and "fish", and "train", "car", "service" and words related to user behavior such as "trimming". However, the words are not limited to these words, and any words related to the user's episode may be used.

Keywords related to the episode information may be automatically extracted from the episode information by the keyword extraction unit 14, or may be arbitrarily set and input by the user on the terminal device 20. FIG.

For example, if the episode information is image data and video data, the keyword extraction unit 14 recognizes subjects such as people, buildings such as facilities, and vehicles included in the image data and video data, and uses the names of the subjects as keywords. Extract. Existing techniques can be used for object recognition and object name specification. Further, the keyword extracting unit 14 may recognize the type of the subject and extract the keyword by inputting image data and video data into a learned model that has been machine-learned by deep learning, for example. A convolutional neural network, for example, can be used as the trained model.

Moreover, when the image data and moving image data contain text such as store names and facility names, the keyword extraction unit 14 extracts the text as a keyword.

Further, when the information of an episode is text data and a keyword is to be extracted from the text data, the keyword extracting unit 14 pre-stores, for example, a large number of reference words related to the user's episode and includes them in the text data. It is only necessary to extract, as a keyword, those words that match the reference word from among the words that appear in the reference word.

If the episode information includes location information obtained by the GPS function or date information obtained by the calendar function, the keyword extraction unit 14 also extracts these information as keywords.

Note that the keyword extraction unit 14 may extract keywords from image data and video data using an API (Application Programming Interface) on the cloud. As such an API, for example, Google Cloud's Vison API or Video Intelligence API can be used.

By automatically extracting the keyword by the keyword extraction unit 14 as described above, the user can save the trouble of setting and inputting the keyword.

The episode information storage unit 12 classifies the information of the episodes linked with the keywords as described above into categories set in advance and stores them. Categories include, for example, work, hobbies, trips, hospitals and schools, but can be arbitrarily set.

The episode information classification unit 15 automatically classifies the information of episodes by category. Specifically, the episode information classification unit 15 is set with a table in which keywords associated with episode information are associated with categories. The above table is referred to based on the obtained keyword, the category to which the episode information belongs is determined, and the episode information is classified.

The category to which the episode information belongs is not only automatically determined by the episode information classification unit 15, but can also be arbitrarily set by the user in the terminal device 20 as described below.

By classifying and storing information on episodes into categories as described above, the user can efficiently check a desired episode by specifying a category.

The episode information edit accepting unit 16 accepts editing of the episode information stored in the episode information storage unit 12 . For example, when the episode information is text data, the episode information editing reception unit 16 edits the text, adds and deletes keywords linked to the episode information, and changes and adds categories to which the episode information belongs. Or accept edits such as deletion.

For example, when the episode information is image data and video data, the episode information editing reception unit 16 adds and deletes keywords linked to the episode information, and changes, adds, or deletes the category to which the episode information belongs. Accepts edits such as deletion.

The episode information output unit 13 reads information of episodes stored in the episode information storage unit 12 and outputs the information to the terminal device 20 . In particular, when outputting predetermined episode information, the episode information output unit 13 of the present embodiment outputs episode information related to the predetermined episode information based on the keyword linked to the predetermined episode information. are output at the same time. A method for outputting episode information will be described in detail later.

Next, the terminal device 20 will be described.

The terminal device 20 includes a CPU, a semiconductor memory, a hard disk, a communication I/F (Interface), and the like. The terminal device 20 is composed of a mobile terminal device such as a tablet terminal or a smart phone, but may be a desktop personal computer or a notebook personal computer.

The terminal device 20 includes an episode information generator 21 and a controller 22, as shown in FIG.

An episode recall program is installed in the semiconductor memory or hard disk of the terminal device 20 . The functions of the episode information generation unit 21 and the control unit 22 are realized by executing the episode recall program by the CPU provided in the terminal device 20 . In the present embodiment, the functions of the episode information generation unit 21 and the control unit 22 are realized by the episode recall program as described above, but not limited to this, some or all of the functions may be implemented in an ASIC ( Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array), and other hardware such as electric circuits.

The episode information generation unit 21 generates the above-described episode information and outputs it to the episode information storage output device 10 . The episode information output from the episode information generation unit 21 is received by the episode information reception unit 11 of the episode information storage output device 10 and stored in the episode information storage unit 12 . A method of generating episode information will be described in detail later.

The control unit 22 controls the terminal device 20 as a whole. In particular, the control unit 22 of the present embodiment causes the terminal device 20 to display the episode information output from the episode information output unit 13 of the episode information storage output device 10 . The method of displaying episode information will be described later in detail.

Next, an example of the processing flow of the episode recall system 1 of this embodiment will be described. First, the process of generating episode information using the terminal device 20 and storing it in the episode information storage output device 10 will be described with reference to the flowchart shown in FIG.

First, an episode recall program installed in the terminal device 20 is started, and an initial screen as shown in FIG. 3 is displayed on the terminal device 20 (S10). As shown in FIG. 3, on the initial screen, an icon labeled "remember" and an icon labeled "remember" are displayed.

Then, when generating episode information, the icon of "store" is tapped and selected (S12).

When the "store" icon is selected, the episode type selection screen shown in FIG. 4 is displayed on the terminal device 20, and the user selects the type of episode (S14). Specifically, in the present embodiment, an icon labeled "camera", an icon labeled "image", and an icon labeled "character" are displayed on the episode type selection screen.

When the icon described as "camera" is tapped and selected, the camera function of the terminal device 20 is activated, and image data or moving image data is captured. After image data or moving image data is captured as episode information, category selection, title and predetermined text setting input, etc. are performed to generate episode information (S16).

After that, when the “registration” button is tapped, the captured image data or moving image data is output to the episode information storage output device 10 and received by the episode information receiving section 11 . Then, the keyword extraction unit 14 extracts a keyword from the captured image data or moving image data and associates it with the image data or moving image data (S18). Then, it is classified into the category designated by the user and the category determined by the episode information classification unit 15, and stored in the episode information storage unit 12 (S20).

Also, in S14, when the icon described as "image" is tapped and selected, an image selection screen for accepting selection of image data and moving image data prestored in the terminal device 20 is displayed. When predetermined image data or video data is selected on the image selection screen, category selection, title and predetermined text setting input, etc. are performed, and episode information is generated (S16). Note that selection of image data and moving image data is not limited to one, and a plurality of images can be selected.

After that, when the “Register” button is tapped, the selected image data or moving image data is output to the episode information storage output device 10 and accepted by the episode information accepting section 11 . Then, the keyword extraction unit 14 extracts a keyword from the selected image data or moving image data and associates it with the image data or moving image data (S18). Then, it is classified into the category designated by the user and the category determined by the episode information classification unit 15, and stored in the episode information storage unit 12 (S20).

In addition, in S14, when the icon described as "character" is tapped and selected, the text is extracted from the image data captured by the camera function, or the text is extracted from the image data already saved in the terminal device 20. A text input selection screen is displayed on the terminal device 20 to accept a selection of text extraction or direct input of text data.

Then, in the text input selection screen, when "extract text from image data photographed by camera function" is selected, the camera function of terminal device 20 is activated and image data is photographed. Then, text data is extracted from the photographed image data by the character recognition function. At this time, it is possible to set and input the range of the text included in the image data, and the text data included in the set and input range of the image data is extracted.

After the text data is extracted from the image data, the category is selected, the title and predetermined text are set, and the episode information is generated (S16).

After that, when the “Register” button is tapped, the text data extracted from the image data is output to the episode information storage output device 10 and accepted by the episode information accepting section 11 . Keywords are then extracted from the text data by the keyword extraction unit 14 and linked to the text data (S18). Then, it is classified into the category designated by the user and the category determined by the episode information classification unit 15, and stored in the episode information storage unit 12 (S20).

Further, when "extract text from image data already saved" is selected on the text input selection screen, an image selection screen for accepting selection of image data pre-saved in the terminal device 20 is displayed. . When predetermined image data is selected on the image selection screen, text data is extracted from the selected image data by the character recognition function. At this time, it is possible to set and input the range of the text included in the image data, and the text data included in the set and input range of the image data is extracted.

After the text data is extracted from the image data, the category is selected, the title and predetermined text are set, and the episode information is generated (S16).

After that, when the “Register” button is tapped, the text data extracted from the image data is output to the episode information storage output device 10 and accepted by the episode information accepting section 11 . Keywords are then extracted from the text data by the keyword extraction unit 14 and linked to the text data (S18). Then, it is classified into the category designated by the user and the category determined by the episode information classification unit 15, and stored in the episode information storage unit 12 (S20).

Further, when "directly input text data" is selected on the text input selection screen, a text input screen is displayed.

Then, after text data is input on the text input screen, category selection, title and predetermined text setting input, etc. are performed to generate episode information (S16).

After that, when the “Register” button is tapped, the text data input on the text input screen is output to the episode information storage output device 10 and received by the episode information reception unit 11 . Keywords are then extracted from the text data by the keyword extraction unit 14 and linked to the text data (S18). Then, it is classified into the category designated by the user and the category determined by the episode information classification unit 15, and stored in the episode information storage unit 12 (S20).

Next, processing for outputting and displaying episode information from the episode information storage output device 10 to the terminal device 20 and processing for editing the displayed episode information will be described with reference to the flowchart shown in FIG. .

First, the history information display program installed in the terminal device 20 is started, and an initial screen as shown in FIG. 3 is displayed on the terminal device 20 (S20). As shown in FIG. 3, on the initial screen, an icon labeled "remember" and an icon labeled "remember" are displayed.

Then, when displaying the information of the episode, the icon of "remember" is tapped and selected (S22).

When the "remember" icon is tapped, a category selection screen as shown in FIG. 6 is displayed on the terminal device 20 (S24). On the category selection screen, a plurality of icons each having a category name are displayed side by side on the screen. In FIG. 6, the category names are "category 1" to "category 5". By displaying the icon with the category name described in this way, the user can easily reach the information of the desired episode. Further, in this embodiment, when the "remember" icon is tapped, the category selection screen shown in FIG. 6 is displayed. button and an "episode list" button may be displayed. When the "category list" button is tapped, the category selection screen shown in FIG. 6 is displayed. A list of information on all episodes stored in 12 may be displayed.

Then, when one of the plurality of icons displayed on the category selection screen is selected (S26), a list of information on episodes belonging to the category described in the selected icon is displayed (S28). . FIG. 7 is a diagram showing an example in which a predetermined icon is selected and icons of information 1 to 3 of three episodes belonging to the category described in that icon are displayed as a list. When any one of the icons of episode information 1 to 3 is selected (S30, YES), as shown in FIG. 8, the specific content of the episode information is displayed (S32). . As the specific content of the episode information, in addition to text, image data and video data, keywords linked to the episode information are displayed. Furthermore, when the selected episode information is displayed as described above, episode information related to the selected episode information is also displayed as shown in FIG. A method of determining related episode information will be described in detail later.

In addition, as shown in FIG. 7, the icons of episode information 1 to 3 each display an "edit" button. If the "edit" button is tapped in S28 (S34, YES), an edit content selection screen including an "edit" button, a "change category" button and a "delete" button as shown in FIG. 9 is displayed. displayed (S36).

When the "edit" button shown in FIG. 9 is tapped, the title, text data, and keywords of the episode information are displayed so that they can be edited (S38). For keywords, a keyword editing screen is displayed, and keywords can be deleted or new keywords can be added.

For example, when a keyword of predetermined episode information is deleted, the episode information classification unit 15 removes the predetermined episode information from the category corresponding to the deleted keyword. That is, the information of the predetermined episode is no longer classified into the category corresponding to the deleted keyword.

On the other hand, for example, when a new keyword is added to information of a predetermined episode, the episode information classification unit 15 adds the information of the predetermined episode to a category corresponding to the added keyword. That is, the episode information classification unit 15 newly classifies the information of the predetermined episode into a category corresponding to the added keyword.

Further, when the "change category" button shown in FIG. 9 is tapped, a screen for selecting the changed category is displayed, and the changed category is selected by the user (S40). At this time, the current category may be changed to a different category, or a new category different from the current category may be added.

Also, when the "delete" button shown in FIG. 9 is tapped, the information of the episode to be edited is deleted (S42).

By accepting editing of episode information as described above, keywords linked to episode information can be easily added or deleted, and the category to which the episode information belongs can be easily changed.

Next, a method for determining related episodes when displaying information on an episode selected by the user and information on episodes related to the information on the selected episode will be described in detail.

Specifically, when the episode information output unit 13 outputs information on a predetermined episode belonging to the category selected by the user, it identifies a keyword linked to the information on the predetermined episode.

For example, when the user selects the category of "Kanagawa", it is assumed that information of episodes linked with "Yokohama" as a keyword is displayed. In this case, the episode information output unit 13 searches for information on other episodes associated with the keyword "Yokohama". When the information of other episodes is classified, the information of other episodes is read out and output and displayed together with the information of the episodes of the "Kanagawa" category described above. That is, information of episodes that are associated with the same keyword of "Yokohama" and that are classified into the category "Kanagawa" and information of episodes that are classified into the category "gourmet" are displayed. By displaying the episode information related to the displayed episode information in this way, it is possible to automatically and efficiently display the episode information that the user has forgotten.

Further, when determining and displaying episode information related to predetermined episode information belonging to the category selected by the user as described above, if there are a plurality of keywords linked to the predetermined episode information. There is

In such a case, the episode information output unit 13 outputs the predetermined episode based on the frequency of each keyword linked to all episode information stored in the user's episode information storage unit 12 of the terminal device 20. A specific keyword is specified from a plurality of keywords linked to information.

Then, the episode information output unit 13 outputs and displays other episode information associated with the specified keyword as related episodes.

A specific example will be given for explanation. FIG. 10 shows all episode information 1 to 12 stored in the episode information storage unit 12, keywords associated with the episode information 1 to 12, and categories into which the episode information 1 to 12 are classified. FIG. 4 is a diagram showing an example; FIG. 11 is a diagram showing the frequency of each category to which the episode information 1 to 12 shown in FIG. 10 belongs and the frequency of keywords linked to the episode information 1 to 12. In FIG.

Episode information 1 to 12 as shown in FIG. 10 are stored. For example, when the category "Kanagawa" is selected and episode information 6 is displayed, the keyword associated with the episode information 6 is There are three keywords: "Soba", "Japanese food" and "Yokohama". Therefore, the episode information output unit 13 refers to the frequencies of these keywords. As shown in FIG. 11, the frequency of "soba" is "1", the frequency of "Japanese food" is "2", and the frequency of "Yokohama" is "9". Therefore, the episode information output unit 13 specifies, for example, the keyword "Yokohama" with the highest frequency, and preliminarily outputs information on episodes other than the episode information 6 that is associated with the keyword "Yokohama". Extract and display only the set number. At this time, as the episode information other than the episode information 6, episode information belonging to categories other than the category "Kanagawa" to which the episode information 6 belongs (for example, "gourmet" and "television") is also extracted.

As described above, it is possible to extract more relevant episode information by identifying the most frequent keyword and determining the information of the episode associated with that keyword as the related episode information. can.

In the above description, as a method of specifying keywords when determining information on related episodes, not only the keywords with the highest frequency but also the keywords with the highest frequency and the keyword with the second highest frequency, for example, are specified. It is also possible to display the information of the episode linked to the keyword.

Contrary to the above example, a keyword with the lowest frequency may be specified, and information on episodes associated with the keyword may be extracted as information on related episodes. Alternatively, for example, a keyword with the lowest frequency and a keyword with the second lowest frequency may be specified, and information on episodes linked to these keywords may be displayed.

In this way, if a keyword with low frequency is specified and the information of the episode linked with that keyword is extracted as the information of the related episode, the information of the episode linked with the keyword with low frequency is more Since it can be said that it is an easy-to-forget episode, it is possible to retrieve and recall an episode that has been completely forgotten.

In addition, keywords whose frequencies are set in advance, such as the keyword with the third highest frequency and the keyword with the fourth highest frequency, are specified, and the episode information linked to the keyword is displayed as related episode information. You may make it

In addition, in the above description, the method of specifying the most frequently occurring keyword, the method of specifying the least frequently occurring keyword, and the method of specifying the keyword whose frequency is set in advance have been described. Selection of at least two identification methods out of keyword identification methods is accepted, and the episode information output unit 13 identifies a keyword based on the selected identification method, and outputs information on an episode linked to the identified keyword. may be determined as related episode information.

Note that the method of extracting and displaying a predetermined episode is not limited to the method of specifying a predetermined episode after selecting a category as described above. There is a method of extracting and displaying the recorded episodes. A method of displaying related episodes in this display method will be described below.

First, when the search button S shown in FIG. 6 is clicked, a search information input screen as shown in FIG. 12 is displayed. On the search information input screen shown in FIG. 12, the "title" of the episode, the "category" of the episode information, the "date" of the episode information, and the "keyword" associated with the episode information are set and input as search information. displayed as possible. When searching based on "keyword", "select keyword" shown in FIG. 12 is selected.

When "Select Keyword" shown in FIG. 12 is selected, a keyword selection screen shown in FIG. 13 is displayed, and all keywords linked to all episode information stored in the episode information storage unit 12 are selected. displayed as possible. The information of all episodes is information of all episodes of the user who has logged in to the terminal device 20 . Also, on the keyword selection screen, the keyword is selected by selecting the "select" button displayed next to the keyword. FIG. 13 shows a state in which "Yokohama" is selected as a keyword. When a predetermined keyword (for example, "Yokohama") is selected, the "select" button is switched to a "cancel" button, and the selection can be canceled. When the "delete" button is selected, the keyword next to it (for example, "Yokohama") is deselected.

Then, when a predetermined keyword is selected on the keyword selection screen shown in FIG. 13, a list of information of episodes associated with the selected keyword is displayed as shown in FIG. Then, in the list display of the information of the searched episodes shown in FIG. 14, buttons of "search result" and "related episode" are displayed. FIG. 14 shows a state in which the "search result" button is selected. When the "related episode" button shown in FIG. 14 is selected, the display is switched to a list of episode information related to the searched episode information.

Regarding the method of determining related episode information, related episode information is determined based on the frequency of keywords linked to all episodes retrieved as "search results." Specifically, in the same manner as described above, the frequency of each keyword linked to all retrieved episodes is calculated, and, for example, the keyword with the highest frequency is specified. Then, the information of the episode linked to the keyword with the highest frequency is determined as the display of related episodes and displayed as a list. Note that the keyword identification method is not limited to the method of identifying the keyword with the highest frequency, and as described above, the keyword with the lowest frequency or the keyword whose frequency is set in advance may be identified. .

Further, in the above description, when information about a predetermined episode is displayed, information about other episodes related to the information about the predetermined episode is displayed. You may make it display the category relevant to the information of a predetermined episode.

For example, when the category "Kanagawa" is selected and the episode information 6 is displayed, the category to which the episode information 6 belongs includes "Gourmet" in addition to "Kanagawa", as shown in FIG. Therefore, when the episode information 6 is displayed, the episode information output unit 13 simultaneously outputs and displays the category "gourmet" as a related category.

In the above example, the only category related to the episode information 6 is "Gourmet", but there are a plurality of categories to which the episode information 6 belongs, other than "Kanagawa" selected by the user. In that case, the categories to be displayed as related categories may be identified based on the frequency with which each category is used. The frequency of use of each category is the frequency of use of the category to which all the episode information stored in the user's episode information storage unit 12 of the terminal device 20 belongs. For example, when episode information 1 to 12 as shown in FIG. 9 is stored in the episode information storage unit 12, the frequency of use of the category "Kanagawa" is "9" and the frequency of use of the category "gourmet" is " 5”.

The episode information output unit 13 may, for example, preferentially display categories in order of frequency of use, or may preferentially display categories in order of low frequency of use. , and identify related categories in the selected order of priority.

In addition, the episode information output unit 13 automatically outputs the episode information based on the keyword linked to the episode information stored in the episode information storage unit 12, and automatically outputs the episode information to the terminal device 20 of the user. You may make it display. Specifically, for example, if the episode information is image data or video data and the shooting location is included as a keyword, for example, the user can visit the shooting location with the terminal device 20 having a GPS function. If the GPS information includes the shooting location of the keyword, the terminal device 20 may automatically display image data or moving image data associated with the keyword (shooting location).

Further, when the date (for example, month and day) is included as a keyword linked to the episode information, for example, when the terminal device 20 has a calendar function, the keyword linked to the episode information is When the date is the same as the date, the terminal device 20 may automatically display the information of the episode associated with the date. Further, when time and day of the week are included as keywords linked to episode information, for example, when the terminal device 20 has a calendar function, the time or day of the week linked to the episode information is the same. When the time or the day of the week comes, the terminal device 20 may automatically display the information of the episode associated with the time or the day of the week.

By automatically displaying episode information on the terminal device 20 as described above, it is possible to present information on episodes likely to be relevant to the user without the user's designation.

In the episode recollection system 1 of the above-described embodiment, text data is stored and output as episode information. may be stored and output as Specifically, as in the first modified example shown in FIG. 15, an episode summary creation unit 17 is further provided for creating a summary of episode information, and the episode information storage unit 12 stores the summary as episode information. can be By creating and storing a summary in this way, the storage capacity of a storage medium such as a memory can be reduced. In the case of text data created by the user, it may take a long time to reread the text because it is too long. can be understood quickly.

An example of a summary creation method in the episode summary creation section 17 will be described below.

The episode summary creation unit 17 stores a plurality of pre-created summary templates, for example, for each category. Each summary template is associated with a keyword included in the summary template and stored. For example, six summary templates are pre-stored for category A, as shown in FIG. For example, a plurality of keywords A-1a to A-1g are associated with the summary template A-1 and stored. A plurality of keywords A-2a to A-2h are associated with the summary template A-2 and stored.

Then, when text data is stored as episode information for a predetermined category, the episode summary creation unit 17 stores the keywords linked to the plurality of summary templates belonging to the predetermined category and the text data. Compare with the associated keywords and identify the summary template with the most matching keywords. Then, the episode summary creating unit 17 determines the specified summary template as a summary of the text data.

The episode information storage unit 12 stores the summary template determined as described above as episode information.

Note that when the episode summary creation unit 17 determines a summary template, the number of keywords that match the keywords linked to the text data for all the summary templates belonging to the predetermined category is equal to or less than a preset threshold. may determine that there is no suitable summary template and not create a summary. That is, the text data may be stored in the episode information storage unit 12 as it is without summarizing the text data.

Further, the method of creating a summary in the episode summary creating unit 17 is not limited to the above-described method using a summary template, and for example, a learned model may be used. Specifically, the relationship between a plurality of keywords linked to the text data and a summary created in advance based on the text data is machine-learned in advance to generate a learned model, and the episode summary creation unit 17 Remember. Then, when text data is stored as episode information for a predetermined category, the episode summary creation unit 17 inputs keywords linked to the text data to the learned model, Get the output summary. Then, the episode information storage unit 12 stores the acquired summary as episode information.

When a summary is created using a trained model as described above, it is preferable to generate and store a trained model for each category.

As a machine learning method, a known method can be used, such as a deep neural network (DNN), a convolutional neural network (CNN), a recurrent neural network (RNN), and a denoising stack autoencoder (DSA). can be used.

Further, in the episode recall system 1 of the above-described embodiment, when information of a predetermined episode is output by the episode information output unit 13 as in the second modification shown in FIG. A search unit 18 for searching the Internet based on the attached keyword may be further provided. Then, the episode information output unit 13 may simultaneously output the information searched by the Internet search by the search unit 18 together with the predetermined episode and display the information on the terminal device 20 . As a result, for example, Internet news that was playing during past episodes can also be checked together.

Further, in the episode recall system 1 of the above-described embodiment, as in the third modification shown in FIG. An output unit 19 may be further provided. The external device may be, for example, a terminal device owned by a user's friend. The keyword output from the keyword output unit 19 can be received and confirmed by the friend's terminal device. In this way, by making it possible to output keywords to a terminal device of a friend or the like, keywords can be exchanged, and mutual hobbies and activities can be confirmed.

Further, when displaying the icons of each category on the category selection screen shown in FIG. 6, each category may be ranked, and the icon with the higher rank may be displayed at a higher rank. For two icons arranged side by side, for example, the icon on the left may be set as the category with the highest rank. When information on a new episode is registered in the new arrivals category, the information on the episode is stored, and when the content of the stored information on the episode is confirmed, the information is deleted. The New Arrivals category icon is always displayed at the top on the left, and the Others category icons are displayed in order from the top on the right.

As for the ranking of the categories, for example, the frequency of registration of episode information in the category or the frequency of reference of episode information may be counted, and ranking may be made in descending order of these counts. Alternatively, the sum of the registration frequency and the reference frequency may be counted and ranked in descending order of the total value. Alternatively, the user may be allowed to arbitrarily set the order for each category.

1 episode recollection system 10 episode information storage output device 11 episode information reception unit 12 episode information storage unit 13 episode information output unit 14 keyword extraction unit 15 episode information classification unit 16 episode information edit reception unit 17 episode summary creation unit 18 search unit 19 keyword Output unit 20 Terminal device 21 Episode information generation unit 22 Control unit

Claims (24)

an episode information reception unit that receives information on the user's past episodes;
an episode information storage unit that associates and stores information on the episode and keywords related to the information on the episode;
an episode information output unit that reads out and outputs the information of the episode stored in the episode information storage unit;
An episode in which, when the episode information output unit outputs the predetermined information of the episode, the information of the episode related to the information of the predetermined episode is output based on the keyword linked to the information of the predetermined episode. Information storage output device. 2. The episode information storage output device according to claim 1, further comprising a keyword information extractor for extracting said keyword from said episode information. When the episode information output unit outputs the information of the predetermined episode, the keyword linked to the information of the predetermined episode is specified, and other episodes linked with the same keyword as the specified keyword. 3. The episode information storage output device according to claim 1, wherein the information of the episode is outputted as the information of the related episode. When there are a plurality of keywords linked to the information of the predetermined episode, the episode information output unit outputs each keyword linked to the information of the plurality of episodes of the user stored in the episode information storage unit. 1. A specific keyword is specified from among the plurality of keywords based on frequency, and information of other episodes linked with the same keyword as the specified keyword is output as information of the related episode. 3. The episode information storage output device according to any one of 3. The episode information output unit identifies a keyword whose frequency is higher than a preset ranking based on the frequency of each keyword, and outputs information of other episodes linked to the same keyword as the identified keyword. 5. The episode information storage and output device according to claim 4, wherein the episode information is output as information of the related episode. The episode information output unit identifies a keyword whose frequency is lower than a preset rank based on the frequency of each keyword, and outputs information of other episodes linked to the same keyword as the identified keyword. 5. The episode information storage and output device according to claim 4, wherein the episode information is output as information of the related episode. The episode information output unit identifies a keyword having a predetermined frequency based on the frequency of each keyword, and outputs information on other episodes linked to the same keyword as the identified keyword. 5. The episode information storage output device according to claim 4, which outputs information as related episode information. 8. The episode information storage output device according to any one of claims 1 to 7, wherein the episode information storage unit stores the information of the episodes by classifying them according to preset categories. 9. The episode information storage output device according to claim 8, further comprising an episode information classification unit that automatically classifies the episode information into the categories based on keywords linked to the episode information. 10. The episode information storage/output device according to claim 8, wherein said predetermined episode information and said related episode information are information of episodes belonging to said different categories. 11. The episode information storage output device according to any one of claims 8 to 10, wherein said episode information output unit outputs said category related to said predetermined episode information when outputting said predetermined episode information. 12. The episode information storage output device according to any one of claims 1 to 11, wherein said episode information is an image. 13. The episode information storage output device according to any one of claims 1 to 12, wherein the episode information output unit automatically outputs the episode information based on a keyword linked to the episode information. an episode summary creation unit that creates a summary of the episode information when the information of the episode is text;
14. The episode information storage output device according to any one of claims 1 to 13, wherein said episode information storage unit stores said summary as said episode information. 15. The episode information storage output device according to any one of claims 1 to 14, further comprising a keyword output unit that outputs the keyword associated with the episode information stored in the episode information storage unit to an external device. a search unit for performing an Internet search based on the keyword linked to the predetermined episode information when the episode information output unit outputs predetermined episode information;
16. The episode information storage and output device according to any one of claims 1 to 15, wherein said episode information output unit outputs information retrieved by said internet search together with information of said predetermined episode. 17. The episode information storage output device according to any one of claims 1 to 16, further comprising an episode information edit receiving section that receives editing of the information of the episode. 18. The episode information storage/output device according to claim 17, wherein said episode information edit reception unit receives change or addition of a category to which said episode information belongs. 19. The episode information storage output device according to claim 17 or 18, wherein the episode information edit reception unit receives addition or deletion of a keyword linked to the information of the episode. an episode information storage output device according to any one of claims 1 to 19;
and an episode recall system comprising a terminal device having a control unit for displaying the predetermined episode information output from the episode information storage output device and the episode information related to the predetermined episode information. 21. The episode recollection system according to claim 20, wherein the control unit displays a plurality of icons in which the names of a plurality of categories set in advance are displayed side by side on the screen. When one of a plurality of icons displayed on the screen is selected, the control unit displays information on episodes belonging to a category described in the selected icon and information on the related episodes. 22. The episodic recall system of claim 21, wherein the system allows 22. The control unit according to claim 21, wherein the control unit receives an input of the predetermined keyword, displays information on an episode linked to the received keyword, and displays information on an episode related to the information on the displayed episode. Episodic recall system. An episode recall program installed in the terminal device in the episode recall system according to any one of claims 20 to 23,
An episode recall program for causing the terminal device to display the predetermined episode information output from the episode information storage output device and the episode information related to the predetermined episode information.

Images