Program, system, method, and terminal device

Abstract

To provide a mechanism for appropriately generating unique images related to a preference or community of a user as images used for user identification in SNS (Social Networking Service).SOLUTION: There is provided a program that causes a CPU of an image generation system 100 which enables a user to post to at least one SNS via a general-purpose terminal 103 to extract keywords by analyzing the user's post data on the SNS, and generate an image using a learned model on the basis of the extracted keywords.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a program, a system, a method, and a terminal device.

Background Art

[0002] Due to the spread of SNS (Social Networking Service), individuals actively send information. Identification of an individual (hereinafter referred to as "user") in SNS is performed not only by name but also by images such as, for example, an icon, a header, an avatar, etc. (hereinafter may be abbreviated as "icon etc."). In this regard, images such as icons are often set with images related to the user's hobbies and the communities to which the user belongs. In this regard, Patent Document 1 discloses a technique for generating poster preference information by analyzing SNS information and associating subjective preference polarity and degree. In recent years, when a keyword is input in document editing software, a service that provides an image generated by an image generation AI based on the input keyword is known. Such a service is convenient because it can prepare the material image required by the user immediately after the keyword is input.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, even if the user uses the technology of Patent Document 1, it is difficult to determine which word is appropriate as a keyword that clearly indicates the user's hobbies and the communities to which the user belongs. Therefore, it has been difficult for the user to cause an image generation AI to generate an image related to the user. In addition, many users do not have the skill to create an image such as an icon by themselves. Therefore, there are cases where the user cannot prepare an image related to the user as an image such as an icon used for user identification on SNS, even if the user wants to do so.

[0005] The present invention has been made in view of the above problems. An object of the present invention is to provide a mechanism that can appropriately generate an image based on a keyword extracted by performing an analysis on user posting data on an SNS as an image used for user identification on the SNS.

Means for Solving the Problems

[0006] In order to achieve the above object, the program of the present invention causes a computer of a system in which a user can post on at least one or more SNSs (Social Networking Services) using a terminal device to perform an analysis on user posting data on the SNS to extract keywords, and a generation means for generating an image using a learned model based on the keywords extracted by the extraction means.

Effects of the Invention

[0007] According to the present invention, as an image used for user identification on an SNS, an image based on a keyword extracted by performing an analysis on user posting data on the SNS can be appropriately generated.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Mode for Carrying Out the Invention

[0009] Hereinafter, each embodiment of the present invention will be described in detail with reference to the drawings. However, the configurations described in the following embodiments are merely examples, and the scope of the present invention is not limited by the configurations described in each embodiment. For example, each part constituting the present invention can be replaced with any configuration that can exhibit the same function. Also, an arbitrary component may be added. Also, any two or more configurations (features) among the embodiments can be combined. Furthermore, in the accompanying drawings, the same or similar configurations are denoted by the same reference numerals, and redundant explanations are omitted.

[0010] <First Embodiment> Hereinafter, a first embodiment will be described with reference to FIGS. 1 to 4. In the first embodiment, an image generation system for a short text posting type SNS will be described. Note that the short text posting type SNS in the first embodiment is assumed to be capable of posting images. Also, the image generation system of the first embodiment can be similarly realized for other types of SNSs such as image posting types. FIG. 1 is a diagram showing an example of an image generation system 100 in the first embodiment. As shown in FIG. 1, in the image generation system 100 (system), an SNS server 101, a general-purpose computer 102 (terminal device), and general-purpose terminals 103 and 104 (terminal devices) are connected to a Web network 105. The general-purpose computer 102 is a notebook personal computer, but may be a desktop personal computer. The general-purpose terminals 103 and 104 are smartphones or tablet terminals.

[0011] FIG. 2 is a block diagram of each component of the image generation system 100. FIG. 2(a) is a block diagram of the short text posting type SNS server 101. As shown in FIG. 2(a), the SNS server 101 is equipped with a CPU 201, a RAM 202, a ROM 203, a storage unit 204, a GPU 205, an input device 206, a display 207, a short text posting control unit 208, and a network I / F 209. The CPU 201 controls the operation of the SNS server 101 and operates based on a program developed in the RAM 202, and performs processing on, for example, the short text posting control unit 208. The ROM 203 is a boot ROM. The ROM 203 stores a boot program of the SNS server 101 and the like. The storage unit 204 assumes a non-volatile device such as an HDD or an SSD.

[0012] The memory unit 204 stores at least one or more of the learned model 210, the image generation AI program 211, and the SNS data 212. The SNS data 212 includes account data and posting data on the SNS. In the image generation AI implemented by the SNS server 101, any learned model 210 or an existing image generation AI program 211 such as Stable Diffusion is used, but other ones may also be used. The image generation AI program 211 is deployed in the RAM 202 and executed by the CPU 201. As a result, the image generation AI using the learned model 210 is implemented in the SNS server 101. Since the technology of the image generation AI is a known technology, a detailed description thereof is omitted.

[0013] The GPU 205 is an image processing device. In the SNS server 101, according to an instruction from the CPU 201 in response to a request for the image generation AI, the GPU 205 performs the process of image generation. The data of the image generated by this image generation process is stored in the memory unit 204 and transmitted to the general-purpose computer 102, the general-purpose terminals 103, and 104 via the Web network 105. The input device 206 is an input device such as a keyboard and a mouse. The display 207 displays the screen of the SNS server 101 and the like. The short text posting control unit 208 performs controls related to the SNS, such as data transmission and reception with the general-purpose computer 102, the general-purpose terminals 103, and 104, and data storage in the memory unit 204. Since the control related to the SNS is a known technology, a detailed description thereof is omitted. The network I / F 209 is connected to the Web network 105 in a wireless or wired form and controls the input and output of various information via the Web network 105.

[0014] Figure 2(b) is a block diagram of the general-purpose computer 102. As shown in Figure 2(b), the general-purpose computer 102 is equipped with a CPU 213, a RAM 214, a ROM 215, an HDD 216, an input device 217, a display 218, and a network I / F 219. The CPU 213 controls the operation of the general-purpose computer 102 and operates based on a program expanded in the RAM 214. The ROM 215 is a boot ROM. The ROM 215 stores a boot program of the general-purpose computer 102 and the like. The HDD 216 stores programs of SNS applications and the like. The HDD 216 is a hard disk drive, but other non-volatile devices may also be used.

[0015] The input device 217 is an input device such as a keyboard or a mouse. The display 218 displays outputs for inputs from the input device 217 and the like. The network I / F 219 is connected to the Web network 105 in a wireless or wired format and controls the input / output of various information via the Web network 105. In the general-purpose computer 102, the CPU 213 executes the program of the SNS application stored in the HDD 216 by expanding it in the RAM 214, and stores text and image data in the HDD 216 as necessary.

[0016] Figure 2(c) is a block diagram of the general-purpose terminal 103. The configuration of the general-purpose terminal 104 is the same as that of the general-purpose terminal 103 described below. As shown in Figure 2(c), the general-purpose terminal 103 is equipped with a CPU 220, a RAM 221, an SSD 222, a touch panel display 223, and a network I / F 224. The CPU 220 controls the operation of the general-purpose terminal 103 and operates based on a program expanded in the RAM 221. The SSD 222 stores a system program, programs of SNS applications, and the like.

[0017] The touch panel display 223 performs input / output processing with the user. The network I / F 224 is connected to the Web network 105 in a wireless or wired format and is responsible for inputting and outputting various information via the Web network 105. However, in the case of the general-purpose terminal 103, the network I / F 224 is often in a wireless format. In the general-purpose terminal 103, the CPU 220 executes by expanding the program of the SNS application stored in the SSD 222 into the RAM 221, and stores text and image data in the SSD 222 as necessary.

[0018] Hereinafter, in the image generation system 100, a sequence in which the user operates the general-purpose terminal 103 to cause the SNS server 101 to generate an icon image will be described with reference to FIGS. 3 and 4. In this case, the image generation system 100 can implement the present invention even without the general-purpose computer 102 and the general-purpose terminal 104. Even if the user operates the general-purpose computer 102 or the general-purpose terminal 104, the SNS server 101 can be caused to generate an icon image in the same manner. FIG. 3 is a diagram showing a transition example of the UI screen of the general-purpose terminal 103.

[0019] FIG. 4 is a sequence diagram showing the processes executed by the SNS server 101 and the general-purpose terminal 103. The sequence (method) of the SNS server 101 shown in FIG. 4 is realized by the CPU 201 (computer) expanding and executing the image generation AI program 211 stored in the storage unit 204 into the RAM 202. At that time, the CPU 201 of the SNS server 101 performs data exchange with the general-purpose terminal 103 via the network I / F 209 as necessary. Similarly, the sequence (method) of the general-purpose terminal 103 shown in FIG. 4 is realized by the CPU 220 (computer) expanding and executing the program of the SNS application stored in the SSD 222 into the RAM 221. At that time, the CPU 220 of the general-purpose terminal 103 performs data exchange with the SNS server 101 via the network I / F 224 as necessary.

[0020] Each UI screen shown in FIG. 3 is drawn on the touch panel display 223 in the general-purpose terminal 103 and transitions based on user operations on the touch panel display 223. First, on the touch panel display 223 of the general-purpose terminal 103, an editing screen 301 shown in FIG. 3(a) is opened by a user operation. The editing screen 301 displays an image upload button 304, an icon image generation button 305, and the like. On the editing screen 301, when the user taps the icon image generation button 305, the sequence shown in FIG. 4 is started. When the sequence shown in FIG. 4 is started, in step S401, the CPU 220 of the general-purpose terminal 103 receives a setting input from the user. At that time, the CPU 220 of the general-purpose terminal 103 displays a setting input screen 302 shown in FIG. 3(b) on the touch panel display 223.

[0021] Boxes 306 to 310 for receiving setting inputs from the user are displayed on the setting input screen 302. In boxes 306 to 309, analysis parameters used in the analysis for keyword extraction in step S403 described later are set and input by the user. Thereby, the user can characterize the analysis for keyword extraction with the analysis parameters. In box 306, a setting input for the analysis period is made. Specifically, in box 306, the setting of the posting date period is performed in an input format such as "after YYMMDD", "before YYMMDD", "within YYMMDD~YYMMDD", "all", etc. Note that the setting of the posting date period in box 306 may be performed in a format in which a day of the week, a time zone, etc. are input.

[0022] In box 307, the setting input for the analysis range is performed. Specifically, in box 307, the setting of the follow relationship is performed in input formats such as "up to the FF of FF (follow follower)", "up to FF", and "within one's own account". In Fig. 3(b), in box 307, "up to the FF of FF" is set and input. This is because the SNS server 101 performs the analysis for extracting keywords indicating the user's hobbies, belonging communities, etc. not only on the user's own posting data but also on the posting data of people in the follow relationship, expecting to improve the accuracy of keyword extraction. This also applies when "up to FF" is set and input in box 307. In addition, by displaying a list of options for people in the follow relationship set and input in box 307, the user may be able to select a specific person from among the people in the set follow relationship.

[0023] In box 308, the setting input for the type of analysis target is performed. Specifically, in box 308, the setting of the type of analysis target is performed in input formats such as "text", "image", "hashtag", and combinations thereof. As described above, the short text posting type SNS in the first embodiment allows image posting. Therefore, in box 308, as shown in Fig. 3(b), the setting input of the combination of "text and image" is possible. In box 309, the additional setting input for influencers is performed. This is because famous accounts, entertainer accounts, corporate accounts, etc. that a user follows on the SNS are important elements indicating the user's tendencies such as hobbies, so regardless of the setting input of the analysis range, the additional setting input for influencers is made possible. Specifically, in box 309, the additional setting for influencers is performed in input formats such as "do" or "do not".

[0024] In box 310, generation parameters used in generating an image in step S404 described later are set and input by the user. Specifically, in box 310, the setting and input of the number of images to be generated are performed in a numeric input format. Thereby, the user can change the number of generated images with the generation parameters. Note that the items of setting and input using boxes 306 to 310 are merely examples, and even if they are added or deleted, it is possible to generate icon images. Also, boxes 306 to 310 may be in the form of a list box or a combo box.

[0025] When the CPU 220 of the general-purpose terminal 103 has the setting and input in boxes 306 to 310 and the generation button 311 is tapped on the setting input screen 302 of the touch panel display 223, it completes receiving the setting input in step S401 of FIG. 4. Thereafter, the process proceeds to step S402. In step S402, the CPU 220 of the general-purpose terminal 103 transmits the information of the setting input performed in boxes 306 to 310 and the information of the user's account to the SNS server 101 via the Web network 105. In step S403, the CPU 201 (extraction means) of the SNS server 101 extracts keywords by analyzing the user's posted data based on the information of the setting input from the general-purpose terminal 103 and the information of the user's account (extraction step). Thereby, keywords indicating the user's hobbies, interests, affiliated communities, etc. are extracted. Note that the posted data is data such as documents and images stored as SNS data 212 in the storage unit 204. Also, when analyzing image data, the CPU 201 of the SNS server 101 uses the GPU 205.

[0026] In addition, the CPU 201 of the SNS server 101 may limit the analysis target to the posted data with the posting date within the set period according to the information of the setting input of the analysis period. Also, the CPU 201 of the SNS server 101 may expand the analysis target to the posted data of the people in the set following relationship according to the information of the setting input of the analysis range. Further, the CPU 201 of the SNS server 101 may limit the analysis target to the posted data of the set type according to the information of the setting input of the type of the analysis target. Additionally, the CPU 201 of the SNS server 101 may expand the analysis target to the posted data of the influencers followed by the user according to the information of the additional setting input of the influencers. Note that the technology of extracting keywords from the analysis of a large amount of document data and the technology of determining what the object extracted from the analysis of a large amount of image data is and extracting keywords from the determination result are well-known technologies, so the detailed description thereof is omitted.

[0027] In step S404, the CPU 201 (generation means) of the SNS server 101 performs the generation of the above image using the extracted keywords (generation step). In this way, the CPU 201 of the SNS server 101 uses the image generation AI of the learned model 210 to accurately generate an image related to the keywords indicating the user's hobbies, interests, affiliated communities, etc. At that time, the CPU 201 of the SNS server 101 generates the number of images set in the box 310 according to the information of the setting input of the number of images to be generated. Also, the CPU 201 of the SNS server 101 may use the keywords randomly selected from the extracted keywords, or the keywords selected according to the importance (frequency). Further, the CPU 201 of the SNS server 101 may use the synonyms (other keywords) obtained by converting the extracted keywords, or the text generated from the extracted keywords.

[0028] Note that, in the first embodiment, as described above, in the SNS server 101, from keyword extraction to image generation is automatically performed. However, the first embodiment may be in a form in which, on the general-purpose terminal 103, the keywords extracted by the SNS server 101 are displayed, and the user is required to select the keywords to be used for generating the image of the SNS server 101 from the displayed keywords. Further, the first embodiment may be in a form in which, on the general-purpose terminal 103 that displays the keywords extracted by the SNS server 101, the user is required to perform an operation of setting input for adding the keywords to be used for generating the image of the SNS server 101.

[0029] In step S405, the CPU 201 of the SNS server 101 transmits the generated image to the general-purpose terminal 103 via the Web network 105. When the CPU 220 of the general-purpose terminal 103 receives the image from the SNS server 101, in step S406, it displays the received image as a selection screen 303 shown in FIG. 3(c). As a result, each image generated in step S404 is displayed on the selection screen 303. Each image displayed on the selection screen 303 is an image related to the keywords indicating the user's hobbies, preferences, affiliated communities, etc. As a simple example, if there are many items related to a specific dog breed in the user's posted data, etc., images of various patterns related to that specific dog breed are independently generated by the image generation AI of the SNS server 101 and displayed on the selection screen 303 of the general-purpose terminal 103. Further, in step S406, the CPU 220 of the general-purpose terminal 103 accepts the user's selection of an image by tapping on the selection screen 303.

[0030] On the selection screen 303, when the user taps and selects an image and then taps the save button 312, in step S407, the CPU 220 of the general-purpose terminal 103 notifies the SNS server 101 of the selected image via the Web network 105. In step S408, the CPU 201 (update means) of the SNS server 101 updates the information of the user's account by storing the information regarding the image of the notification from the general-purpose terminal 103 in the SNS data 212 of the storage unit 204. As a result, the image selected by the user is used as an image for user identification on the SNS.

[0031] As described above, in the first embodiment, by analyzing the user's posted data on the SNS, keywords indicating the user's hobbies, belonging communities, etc. are extracted, and based on the extracted keywords, the image generation AI generates an icon image. Thus, in the first embodiment, as an icon image used for user identification on the SNS, a unique image related to the user's hobbies, belonging communities, etc. can be appropriately generated.

[0032] <Second Embodiment> Hereinafter, a second embodiment will be described with reference to FIGS. 5 to 8. In the second embodiment, an example of an image generation system for a plurality of posting-type SNSs will be described. FIG. 5 is a diagram showing an example of an image generation system 500 in the second embodiment. As shown in FIG. 5, in the image generation system 500 (system), a plurality of servers, namely a short text posting SNS server 501, an image posting SNS server 502, and an image generation server 503, are connected to a Web network 105. Further, in the image generation system 500, a general-purpose computer 102 (terminal device) and general-purpose terminals 103, 104 (terminal devices) are connected to the Web network 105. Note that the image generation system 500 of the second embodiment is for two types of SNSs, namely a short text posting type and an image posting type, but it can be similarly realized for two different short text posting type SNSs or two different image posting type SNSs. Also, the image generation system 500 of the second embodiment can be similarly realized for two SNSs replaced with other posting types or for three or more SNSs with other posting types added.

[0033] FIG. 6 is a block diagram of each component of the image generation system 500. FIG. 6(a) is a block diagram of the short text posting SNS server 501. FIG. 6(b) is a block diagram of the image generation server 503. Note that since the configuration of the image posting SNS server 502 is almost the same as that of the short text posting SNS server 501 shown in FIG. 6(a), its description will be omitted. As shown in FIG. 6(a), the short text posting SNS server 501 is equipped with a CPU 601, a RAM 602, a ROM 603, a storage unit 605 mainly storing SNS data 604, a GPU 606, an input device 607, and a display 608. The short text posting SNS server 501 is further equipped with an SNS control unit 609 and a network I / F 610.

[0034] The CPU 601 controls the operation of the short text posting SNS server 501, operates based on the program expanded in the RAM 602, and performs, for example, processing on the SNS data 604 and the SNS control unit 609. The ROM 603 is a boot ROM. The ROM 603 stores the boot program of the short text posting SNS server 501 and the like. The SNS data 604 includes account data, posting data, etc. in the short text posting type SNS. The storage unit 605 assumes a non-volatile device such as an HDD or an SSD. The GPU 606 is an image processing device.

[0035] The input device 607 is an input device such as a keyboard or a mouse. The display 608 displays the screen of the short text posting SNS server 501 and the like. The SNS control unit 609 controls the short text posting type SNS, such as data transmission and reception with the general-purpose computer 102, the general-purpose terminals 103 and 104, and data storage in the storage unit 204. Since the control of the short text posting type SNS is a known technology, its detailed description is omitted. The network I / F 610 is connected to the Web network 105 in a wireless or wired form and is responsible for input and output of various information via the Web network 105.

[0036] As shown in FIG. 6(b), the image generation server 503 is equipped with a CPU 611, a RAM 612, a ROM 613, a storage unit 614, a GPU 615, an input device 616, a display 617, and a network I / F 618. In the storage unit 614, at least one or more of the learned model 619, the image generation AI program 620, and the SNS data 621 are stored. Since the configuration of the image generation server 503 is substantially the same as that of the SNS server 101 shown in FIG. 2(a), the description of the configuration of the image generation server 503 other than the above is omitted.

[0037] The following describes, with reference to FIGS. 6 and 7, a sequence in which a user operates a general-purpose computer 102 to cause an image generation server 503 to generate an image such as an icon. In this case, the image generation system 500 can implement the present invention even without the general-purpose terminals 103 and 104. Note that even if the user operates the general-purpose terminal 103 or the general-purpose terminal 104, the image generation server 503 can similarly be caused to generate an image such as an icon. FIG. 7 is a diagram showing an example of transition of the UI screen of the general-purpose computer 102.

[0038] FIG. 8 is a sequence diagram showing processes executed by the general-purpose computer 102, the short text posting SNS server 501, the image posting SNS server 502, and the image generation server 503. The sequence (method) of the image generation server 503 shown in FIG. 8 is realized by the CPU 611 (computer) expanding and executing the image generation AI program 620 stored in the storage unit 614 in the RAM 612. At that time, the CPU 611 of the image generation server 503 exchanges data with the general-purpose computer 102, the short text posting SNS server 501, and the image posting SNS server 502 via the network I / F 618 as necessary. Similarly, the sequence (method) of the short text posting SNS server 501 shown in FIG. 8 is realized by the CPU 601 (computer) expanding and executing the program stored in the storage unit 605 in the RAM 602. At that time, the CPU 601 of the short text posting SNS server 501 exchanges data with the image generation server 503 via the network I / F 610 as necessary. Note that the sequence (method) of the image posting SNS server 502 shown in FIG. 8 is realized in substantially the same manner as the sequence of the short text posting SNS server 501. Also, the sequence (method) of the general-purpose computer 102 shown in FIG. 8 is realized by the CPU 213 (computer) expanding and executing the program stored in the HDD 216 in the RAM 214. At that time, the CPU 213 of the general-purpose computer 102 exchanges data with the image generation server 503 via the network I / F 219 as necessary.

[0039] Each UI screen shown in FIG. 7 is drawn on the display 218 in the general-purpose computer 102 and transitions based on user operations using the input device 217. First, on the display 218 of the general-purpose computer 102, the setting input screen 701 shown in FIG. 7(a) is opened by a user operation. When the setting input screen 701 is opened, the sequence shown in FIG. 8 is started. Note that the setting input screen 701 may be a web page screen, an application screen, or other forms. This also applies to the setting input screen 302 of the first embodiment. When the sequence shown in FIG. 8 is started, in step S801, the CPU 213 of the general-purpose computer 102 receives the setting input by the user on the setting input screen 701 shown in FIG. 7(a).

[0040] On the setting input screen 701, boxes 703 to 708 for receiving the setting input by the user are displayed. The setting input screen 701 is roughly composed of an analysis parameter setting area and a generation parameter setting area. In the analysis parameter setting area, boxes 703 and 704 are provided.

[0041] In boxes 703 and 704, the analysis parameters used in the analysis for keyword extraction in step S807 described later are set and input by the user. Thereby, the user can characterize the analysis for keyword extraction with the analysis parameters. In box 703, the setting of the user's account information in the short text posting type SNS is performed in an input format such as "@xxx xx". In box 704, the setting of the user's account information in the image posting type SNS is performed in an input format such as "user name".

[0042] In the case of a modified example where the short text posting SNS server 501 is replaced with a server of another SNS, in box 703, the information of the user's account in the replaced other SNS is set. Similarly, in the case of a modified example where the image posting SNS server 502 is replaced with a server of another SNS, in box 704, the information of the user's account in the replaced other SNS is set. Also, in the case of a modified example where a server of another SNS is added to the image generation system 500, a box enabling the setting of the information of the user's account in the added other SNS is added to the analysis parameter setting area. Further, not only the setting input of the account information but also a box for authenticating the account may be introduced into the analysis parameter setting area.

[0043] Boxes 705 to 708 are provided in the generation parameter setting area. In boxes 705 to 708, the generation parameters used in the image generation in step S808 described later are set and input by the user. Thereby, the user can change the type, color, symmetry, and number of the generated images with the generation parameters. In box 705, the type setting input is performed. Specifically, in box 705, the setting of the image type is performed in an input format such as "icon", "header", "avatar", or a combination thereof. Note that box 705 may be in the form of a list box or a combo box. This also applies to boxes 706 to 708.

[0044] In box 706, color settings are input. Specifically, in box 705, the color settings of the image are input in input formats such as "automatic", "color", and "monochrome". In box 707, symmetry settings are input. Specifically, in box 707, the symmetry settings of the image are input in input formats such as "left - right symmetry", "up - down symmetry", and "asymmetry". In box 708, the setting input for the number of images to be generated is performed in a numeric input format. Note that the items of setting input using boxes 703 to 708 are only examples, and even if they are added or deleted, it is possible to generate images such as icons.

[0045] When the CPU 213 of the general - purpose computer 102 has the setting inputs in boxes 703 to 708 performed on the setting input screen 701 of the display 218 and the generate button 709 is clicked, it completes receiving the setting input in step S801 of FIG. 8. Then, the process proceeds to step S802. In step S802, the CPU 213 of the general - purpose computer 102 transmits the information of the setting inputs performed in boxes 703 to 708 to the image generation server 503 via the Web network 105. In steps S803 and S804, the CPU 611 of the image generation server 503 exchanges information with the short - text posting SNS server 501 via the Web network 105 based on the information from the general - purpose computer 102. At that time, the CPU 611 of the image generation server 503 acquires the information of the user's account set and input in box 703 from the short - text posting SNS server 501, and if there is information of the user that has been analyzed by the short - text posting SNS server 501, that information may also be acquired.

[0046] Similarly, in step S805 and step S806, the CPU 611 of the image generation server 503 exchanges information with the image posting SNS server 502 via the Web network 105 based on the information from the general-purpose computer 102. At this time, the CPU 611 of the image generation server 503 obtains the information of the user's account set and input in the box 704 from the image posting SNS server 502. If there is information of the user that has been analyzed by the image posting SNS server 502, that information may also be obtained. For example, if the image posting SNS server 502 determines what the image posted by the user is and the determination result is converted into words, the CPU 611 of the image generation server 503 may obtain the information of the words instead of the image.

[0047] The CPU 611 (extraction means) (generation means) of the image generation server 503 performs keyword extraction in step S807 and image generation in step S808 based on the information obtained by step S806 (extraction process) (generation process). Note that since the processes of keyword extraction in step S807 and image generation in step S808 are substantially the same as the processes of step S403 and step S404 in the first embodiment, their descriptions are omitted. In step S809, the CPU 611 of the image generation server 503 transmits the generated image to the general-purpose computer 102 via the Web network 105.

[0048] When the CPU 213 of the general-purpose computer 102 receives an image from the image generation server 503, in step S810, it displays the received image as a selection screen 702 shown in FIG. 7(b). As a result, each image generated in step S808 is displayed on the selection screen 702. Each image displayed on the selection screen 702 is an image related to keywords indicating the user's hobbies and tastes, the community to which the user belongs, etc., in the same manner as each image displayed on the selection screen 303 of the first embodiment. As a simple example, when there are many items related to a specific regional brand in the user's posted data, etc., images of various patterns related to that specific regional brand are independently generated by the image generation AI of the image generation server 503 and displayed on the selection screen 702 of the general-purpose computer 102. Note that in the second embodiment, although it is the case where an image of an icon set and input in the box 705 of the setting input screen 701 is generated, even when a header image or an avatar image is generated, the display in step S810 is performed in the same manner.

[0049] Furthermore, in step S810, the CPU 213 of the general-purpose computer 102 accepts the user's selection of an image by a click on the selection screen 702. After that, although not shown, processing substantially the same as each of the processes in steps S407 and S408 of the first embodiment is performed. At that time, the CPU 601 (update means) of the short text posting SNS server 501 updates the information of the user's account by storing the information related to the notified image from the general-purpose computer 102 in the SNS data 604 of the storage unit 605. This is the same for the image posting SNS server 502 and the image generation server 503. As a result, the image selected by the user is used as an image for user identification in two types of SNSs, the short text posting type and the image posting type.

[0050] As described above, in the second embodiment, by analyzing the posted data of users on two SNSs, keywords indicating the hobbies and communities to which the users belong are extracted, and based on the extracted keywords, the image generation AI generates images. As a result, in the second embodiment, as an image used for user identification on two SNSs, a unique image related to the hobbies and communities to which the user belongs can be appropriately generated. Note that in the second embodiment, by analyzing the posted data on the SNSs of a plurality of people including the user, keywords indicating the hobbies and communities to which the user belongs may be extracted.

[0051] <Others> As described above, the preferred embodiments of the present invention have been described. However, the present invention is not limited to the above-described embodiments, and various modifications and changes are possible within the scope of the gist. For example, in the first embodiment, the CPU 201 of the SNS server 101 generates an icon image in step S408, but it is also possible to similarly generate a header image, an avatar image, and the like. Further, in the first embodiment, in the same manner as in the second embodiment, the type, color, and symmetry of the generated image may be changed by setting and inputting generation parameters. Furthermore, in the first and second embodiments, in addition to the type, color, and symmetry of the generated image, for example, the size and aspect ratio of the image may also be changed by setting and inputting generation parameters. Also, in the second embodiment, in the same manner as in the first embodiment, by expanding or limiting the analysis target by setting and inputting additional analysis parameters such as the analysis period, analysis range, type of analysis target, and influencer, the analysis for keyword extraction may be characterized.

[0052] Also, in the first embodiment, the analysis for keyword extraction may be characterized by adding the setting input of analysis parameters (that is, the information of the account of the other person) that specify other people with common hobbies, communities to which they belong, etc. This is because by performing the analysis for extracting keywords indicating the user's hobbies, communities to which they belong, etc. not only on the user's own posting data but also on the posting data of the other person, an improvement in the accuracy of keyword extraction is expected. This point is the same in the second embodiment as well.

[0053] Also, in the first embodiment, each process of keyword extraction in step S407 and image generation in step S408 is executed by the SNS server 101, but if the arithmetic processing ability and memory capacity are sufficient, it may be executed by the general-purpose terminal 103. Also, among the processes of keyword extraction in step S407 and image generation in step S408, one process may be executed by the SNS server 101 and the other process may be executed by the general-purpose terminal 103. These points are the same when the user operates the general-purpose computer 102 or the general-purpose terminal 104.

[0054] Also, in the second embodiment, each process of keyword extraction in step S807 and image generation in step S808 is executed by the image generation server 503, but it may be executed by the short text posting SNS server 501 or the image posting SNS server 502. Also, each process of keyword extraction in step S807 and image generation in step S808 may be executed by the general-purpose computer 102 if the arithmetic processing ability and memory capacity are sufficient. Also, among the processes of keyword extraction in step S807 and image generation in step S808, one process may be executed on the server side such as the image generation server 503 and the other process may be executed by the general-purpose computer 102. These points are the same when the user operates the general-purpose terminal 103 or the general-purpose terminal 104.

[0055] Furthermore, the present invention can also be realized by supplying a program that implements one or more functions of each of the above-described embodiments to a system or apparatus via a network or a storage medium, and causing one or more processors of a computer of the system or apparatus to read and execute the program. Additionally, the present invention can also be realized by a circuit (e.g., ASIC) that implements one or more functions.

[0056] The disclosure of each embodiment includes the following programs, configurations, and methods. (Program 1) A program characterized in that, in a computer of a system that enables a user to post to at least one or more SNSs (Social Networking Services) using a terminal device, extraction means for extracting keywords by performing analysis on user posting data in the SNS, and generation means for generating an image using a learned model based on the keywords extracted by the extraction means are executed. (Program 2) The program according to Program 1, wherein the extraction means expands or limits the analysis target according to analysis parameters set and input by the user in the terminal device. (Program 3) The program according to Program 2, wherein the analysis parameter is a period of posting dates. (Program 4) The program according to Program 2 or 3, wherein the analysis parameter is a follow relationship. (Program 5) The program according to any one of Programs 2 to 4, wherein the analysis parameter is a type of analysis target. (Program 6) The program according to any one of Programs 2 to 5, wherein the analysis parameter is the addition of an influencer. (Program 7) The program according to Program 1, wherein the generation means generates an image based on another keyword obtained by converting the keyword extracted by the extraction means. (Program 8) The program according to Program 1, wherein the generating means generates an image based on a keyword randomly selected from the keywords extracted by the extracting means. (Program 9) The program according to Program 1, wherein the generating means generates an image based on a keyword selected according to importance from the keywords extracted by the extracting means. (Program 10) The program according to Program 1, wherein the generating means generates an image based on a keyword selected by the user on the terminal device from the keywords extracted by the extracting means. (Program 11) The program according to any one of Programs 1 to 10, wherein the generating means generates an image based on a keyword set and input by the user on the terminal device. (Program 12) The program according to any one of Programs 1 to 11, wherein the generating means generates an image according to generation parameters set and input by the user on the terminal device. (Program 13) In the computer of the system, The program according to any one of Programs 1 to 12, wherein an updating means for updating information of the user's account on the SNS with an image selected by the user on the terminal device from the images generated using the learned model is executed. (Program 14) The program according to any one of Programs 1 to 13, wherein the extracting means extracts keywords by performing an analysis targeting at least the posting data of the user on a plurality of SNSs. (Program 15) The program according to any one of Programs 1 to 14, wherein the image generated using the learned model is an icon, a header, or an avatar used for user identification on the SNS. (Configuration 1) A system in which a user can post on at least one or more SNSs using a terminal device, Extraction means for extracting keywords by analyzing user-posted data on SNS, and generation means for generating an image using a learned model based on the keywords extracted by the extraction means, characterized by a system comprising the same. (Method 1) A method executed in a system in which a user can post to at least one or more SNSs using a terminal device, an extraction step of extracting keywords by analyzing user-posted data on SNS, and a generation step of generating an image using a learned model based on the keywords extracted in the extraction step, characterized by a method comprising the same. (Configuration 2) A terminal device in which a user can post to at least one or more SNSs, extraction means for extracting keywords by analyzing user-posted data on SNS, and generation means for generating an image using a learned model based on the keywords extracted by the extraction means, characterized by a terminal device comprising the same.

Explanation of symbols

[0057] 100 Image generation system (system) 102 General-purpose computer (terminal device) 103, 104 General-purpose terminals (terminal devices) 201, 213, 220 CPUs (computer) 500 Image generation system (system) 601, 611 CPUs (computer)

Claims

1. A computer of a system that enables a user to post on a terminal device to at least one or more SNSs (Social Networking Services), an extraction means for extracting keywords by performing an analysis on user post data in the SNS, a generation means for generating an image using a learned model based on the keywords extracted by the extraction means, wherein the program is characterized by causing the above to be executed.

2. The program according to claim 1, wherein the extraction means expands or limits the analysis target according to analysis parameters set and input by the user on the terminal device.

3. The program according to claim 2, wherein the analysis parameter is a period of posting dates.

4. The program according to claim 2, wherein the analysis parameter is a following relationship.

5. The program according to claim 2, wherein the analysis parameter is a type of analysis target.

6. The program according to claim 2, wherein the analysis parameter is the addition of an influencer.

7. The program according to claim 1, wherein the generation means generates an image based on another keyword obtained by converting the keyword extracted by the extraction means.

8. The program according to claim 1, wherein the generation means generates an image based on a keyword randomly selected from the keywords extracted by the extraction means.

9. The program according to claim 1, wherein the generation means generates an image based on a keyword selected according to importance from the keywords extracted by the extraction means.

10. The program according to claim 1, wherein the generation means generates an image based on a keyword selected by the user on the terminal device from the keywords extracted by the extraction means.

11. The program according to claim 1, wherein the generation means generates an image based also on a keyword set and input by the user on the terminal device.

12. The program according to claim 1, wherein the generation means generates an image according to generation parameters set and input by the user on the terminal device.

13. On the computer of the system, The program according to claim 1, characterized in that it causes an updating means to update information of a user's account on an SNS with an image selected by the user on the terminal device from among the images generated using the learned model.

14. The program according to claim 1, characterized in that the extracting means extracts keywords by performing an analysis targeting at least the posting data of the user on a plurality of SNSs.

15. The program according to claim 1, characterized in that the image generated using the learned model is an icon, a header, or an avatar used for user identification on an SNS.

16. A system in which a user can post on at least one or more SNSs using a terminal device, an extracting means for extracting keywords by performing an analysis targeting the posting data of the user on an SNS, and a generating means for generating an image using a learned model based on the keywords extracted by the extracting means.

17. A method executed in a system in which a user can post on at least one or more SNSs using a terminal device, an extracting step for extracting keywords by performing an analysis targeting the posting data of the user on an SNS, and a generating step for generating an image using a learned model based on the keywords extracted in the extracting step.

18. A terminal device in which a user can post on at least one or more SNSs, an extracting means for extracting keywords by performing an analysis targeting the posting data of the user on an SNS, and a generating means for generating an image using a learned model based on the keywords extracted by the extracting means.

Images