Apparatus and method for support, and program

Abstract

To provide an apparatus and method for support and program allowing for relieving the workload of an operator.SOLUTION: An apparatus for support is adapted to acquire information posted on a network and calculate scores indicative of the possibility that a content thereof breaches the agreement previously determined, send information of the breach of agreement to an information terminal an operator operates based on calculated scores and acquire information of the breach of agreement selected by the operator, and then record the acquired information of the breach of agreement with correspondence to the acquired information posted on the network. The calculation of scores is performed using a content indicated by the acquired information and a learning model learnt such that scores become higher with a higher degree of relevancy to the breach of agreement previously determined.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] The present invention relates to an assistance device, an assistance method, and a program. [Background technology]

[0002] It is necessary to determine whether products, advertisements, etc. posted on a network meet predetermined standards. For this reason, various technologies for making such determinations have been proposed. For example, Patent Document 1 discloses a method for automatically inspecting whether online advertisements comply with system policies. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Special Publication No. 2008-527499 Summary of the Invention [Problem to be solved by the invention]

[0004] However, with the above method, it is difficult to apply it to compliance with various laws and regulations for products and services other than online advertising due to reasons such as amendments and new regulations, and ultimately the operator must deal with it. Therefore, there is still room for improvement in terms of reducing the operator's workload.

[0005] The present invention is intended to solve the above-mentioned problems, and has an object to provide an assistance device, an assistance method, and a program that can reduce the workload of an operator. [Means for solving the problem]

[0006] An assistance device according to a first aspect of the present invention comprises: an information acquisition unit that acquires information posted on a network; a score calculation unit that calculates a score indicating a possibility that the content indicated by the information acquired by the information acquisition unit corresponds to a violation of a predetermined rule; a transmitting unit that transmits information about the violation of the rules to an information terminal operated by an operator based on the score calculated by the score calculating unit; a regulation violation information acquisition unit that acquires information on the regulation violation selected by the operator from the information on the regulation violation transmitted by the transmission unit; a recording unit that records the information on the violation of the rules acquired by the information acquisition unit in association with the information acquired by the information acquisition unit, the score calculation unit calculates the score using a learning model that is trained so that the higher the degree of association between the content indicated by the information acquired by the information acquisition unit and the violation of the predetermined terms and conditions, the higher the score; It is characterized by:

[0007] In addition, in the support device according to the above aspect, an arrangement unit that arranges the order of the violations of the rules corresponding to the content indicated by the information acquired by the information acquisition unit based on the score calculated by the score calculation unit, the information acquisition unit acquires, from the information terminal, information selected by an operation of the operator from among the information posted on the network; the transmission unit transmits the information on the violation of the regulations arranged by the arrangement unit to the information terminal operated by the operator; the recording unit records the information on the violation of the regulations acquired by the information acquisition unit in association with the information selected by the operation of the operator acquired by the information acquisition unit. It is characterized by:

[0008] In addition, in the support device according to the above aspect, the information acquisition unit acquires information about products or services posted on a network, the score calculation unit sequentially calculates the score for each preset unit for the information on the product or service acquired by the information acquisition unit; the transmitting unit transmits, to the information terminal, as a special aspect, a background of a portion of the information on the product or service where the score for each unit calculated by the score calculating unit is equal to or greater than a predetermined value; the recording unit records the information on the violation of the terms and conditions acquired by the information acquisition unit on the violation of the terms and conditions in association with a portion of the information on the product or service for which the score for each unit calculated by the score calculation unit is equal to or greater than a predetermined value; It is characterized by:

[0009] In addition, in the support device according to the above aspect, The learning model further learns based on the information on the violation of the rules recorded by the recording unit and the information acquired by the information acquisition unit. It is characterized by:

[0010] In addition, in the support device according to the above aspect, The predetermined unit is a character string of a predetermined length extracted by shifting a predetermined number of characters from the text related to the information of the product or service acquired by the information acquisition unit. It is characterized by:

[0011] In addition, in the support device according to the above aspect, The predetermined unit is a predetermined range of data extracted by shifting the image related to the product or service information acquired by the information acquisition unit by a predetermined number of pixels. It is characterized by:

[0012] In addition, in the support device according to the above aspect, The predetermined unit is extracted by shifting character strings between spaces in units of character strings from the text related to the information on the product or service acquired by the information acquisition unit. It is characterized by:

[0013] In addition, in the support device according to the above aspect, the predetermined unit is obtained by morpheme analysis of the text related to the information on the product or service acquired by the information acquisition unit, and by shifting and extracting the morpheme from the text in morpheme units; It is characterized by:

[0014] In addition, in the support device according to the above aspect, the transmitting unit transmits to the information terminal, in a different manner depending on the score, a background of a portion of the information on the product or service where the score for each unit calculated by the score calculating unit is equal to or greater than a predetermined value. It is characterized by:

[0015] In addition, in the support device according to the above aspect, the information acquisition unit acquires description information and image information from the information on the product or service posted on the network, the score calculation unit sequentially calculates scores for each of the description information and the image information acquired by the information acquisition unit, for each predetermined unit; It is characterized by:

[0016] A support method according to a second aspect of the present invention comprises: A support method using a support device, an information acquisition step of acquiring information posted on a network; a score calculation step of calculating a score indicating a possibility that the content indicated by the information acquired in the information acquisition step corresponds to a violation of a predetermined rule; a transmitting step of transmitting information about the violation of the rules to an information terminal operated by an operator based on the score calculated in the score calculation step; a rule violation information acquisition step of acquiring the rule violation information selected by the operator from the rule violation information transmitted in the transmission step; a recording step of recording the information on the violation of the rules acquired in the information acquisition step in association with the information acquired in the information acquisition step, In the score calculation step, the score is calculated using a learning model that is trained so that the higher the degree of association between the content indicated by the information acquired in the information acquisition step and the violation of the predetermined terms and conditions, the higher the score. It is characterized by:

[0017] A program according to a third aspect of the present invention comprises: Support equipment, an information acquisition unit that acquires information posted on the network; a score calculation unit that calculates a score indicating a possibility that the content indicated by the information acquired by the information acquisition unit corresponds to a violation of a predetermined rule; a transmission unit that transmits information about the violation of the rules to an information terminal operated by an operator based on the score calculated by the score calculation unit; a rules violation information acquisition unit that acquires the rules violation information selected by the operator from the rules violation information transmitted by the transmission unit; a recording unit that records the information on the violation of the rules acquired by the information acquisition unit in association with the information acquired by the information acquisition unit; the score calculation unit calculates the score using a learning model that is trained so that the higher the degree of association between the content indicated by the information acquired by the information acquisition unit and the violation of the predetermined terms and conditions, the higher the score; It is characterized by:

[0018] The program may be recorded on a non-transitory recording medium. The non-transitory recording medium can be distributed or sold independently of the computer. Here, a non-transitory recording medium refers to a tangible recording medium. Examples of non-transitory recording media include compact discs, flexible disks, hard disks, magneto-optical disks, digital video disks, magnetic tapes, and semiconductor memories. A transitory recording medium refers to the transmission medium (propagation signal) itself. Examples of transitory recording media include electrical signals, optical signals, and electromagnetic waves. A temporary storage area is an area for temporarily storing data and programs, such as volatile memory such as RAM (Random Access Memory). [Effects of the Invention]

[0019] According to the present invention, it is possible to provide an assistance device, an assistance method, and a program that can reduce the workload of an operator. [Brief explanation of the drawings]

[0020] [Figure 1] FIG. 2 is a diagram illustrating the relationship between a support device and an information terminal. [Figure 2] FIG. 1 is a block diagram illustrating an example of a support device. [Figure 3] FIG. 10 is an explanatory diagram illustrating an example of a learning model. [Figure 4] 10 is a flowchart illustrating an example of a candidate providing process. [Figure 5] FIG. 10 is an explanatory diagram showing an example of a region selected by an operator. [Figure 6] FIG. 10 is an explanatory diagram showing an example of score calculation. [Figure 7] FIG. 10 is an explanatory diagram showing an example of an arrangement of candidate orders. [Figure 8] FIG. 10 is an explanatory diagram showing an example of providing candidates to an operator. [Figure 9] FIG. 10 is an explanatory diagram showing an example of learning of a learning model after a candidate is selected by an operator. [Figure 10] FIG. 10 is an explanatory diagram showing an example of display of selection options for product descriptions in a modified example. [Figure 11] FIG. 10 is an explanatory diagram showing an example of displaying selection candidates for product images in a modified example. DETAILED DESCRIPTION OF THE INVENTION

[0021] (Overall composition) An assistance device, an assistance method, and a program according to an embodiment of the present invention will be described in detail with reference to the drawings. The same or corresponding parts in the drawings are denoted by the same reference numerals. As shown in Fig. 1, an assistance device 100 according to an embodiment of the present invention is communicably connected to an information terminal 200 via a computer communication network 400 such as the Internet.

[0022] The support device 100 is a computer such as a server, and has a function of providing an operator with information on which of multiple regulations a product or service posted on a network may violate. Specifically, the operator needs to identify the reasons why a product or service should not be posted on a network because it violates the regulations and present these reasons to the provider of the product or service. Therefore, the support device 100 in this embodiment has a function of appropriately presenting to the operator potential reasons why a product or service should not be posted on a network because it violates the regulations, such as being an illegal product or service or including inappropriate descriptions.

[0023] The information terminal 200 is an information terminal (so-called computer) such as a smartphone, tablet, or PC (Personal Computer) used by an operator who examines whether or not products and services posted on the network violate the terms and conditions. The information terminal 200 acquires pages of products and services to be posted on the network from a product management server (not shown), and examines whether or not the products and services violate the terms and conditions.

[0024] (Functional configuration of the support device) Next, the configuration of the assistance device 100 will be described with reference to FIG.

[0025] As shown in FIG. 2, the support device 100 includes a storage unit 110, a control unit 120, an input / output unit 130, a communication unit 140, and a system bus (not shown) that interconnects these units.

[0026] The storage unit 110 includes a ROM (Read Only Memory), a RAM (Random Access Memory), etc. The ROM stores a program 111 to be executed by the control unit 120, various data (not shown) required in advance for executing the program 111, a learning model 112, and past information 113.

[0027] The program 111 is a program for executing a candidate providing process, which will be described later, and is stored in the storage unit 110 in advance.

[0028] Learning model 112 is a learning model that has been machine-learned based on the selected area, which is the part selected by an operator in the product (service) description or product (service) image of a product or service posted on the network, and the reason for the violation of the terms and conditions selected by the operator.

[0029] Specifically, as shown in FIG. 3 , the learning model 112 is a learning model trained based on the content of past information 113 described below, in which the operator selected the phrase "You'll definitely lose weight" in the product description as the selection area and selected "Violation of the Act against Unjustifiable Premiums and Misleading Representations (5)" as the reason for the violation of the terms and conditions. In the illustrated example, the past information 113 also includes the content that the operator selected the phrase "Accommodation Fee: XXX Yen" and the phrase "Benefit Amount: YYY Yen" in the service description as the selection area and selected "Benefit Ratio Violation (4)" as the reason for the violation of the terms and conditions, and the learning model is based on this content. Note that violations of the terms and conditions are further subdivided, such as violations of the Act against Unjustifiable Premiums and Misleading Representations and violations of the benefit ratio. Therefore, "Violation of the Act against Unjustifiable Premiums and Misleading Representations (5)" indicates that it corresponds to the fifth type of violation, which is preset, among the multiple types of violations included in the Act against Unjustifiable Premiums and Misleading Representations. Similarly, "Violation of the Benefit Ratio (4)" indicates that it corresponds to the fourth type of violation, which is preset, among the multiple types of violations included in the benefit ratio violation.

[0030] Returning to Fig. 2, past information 113 is historical information in a database that stores combinations of information about the content selected by the operator and the reason why the operator selected that information about the content. That is, the part "You'll definitely lose weight" shown in Fig. 3 is information about the content selected by the operator (information about the selected area), and the part "Violation of the Act against Unjustifiable Premiums and Misleading Representations (5)" is information about the reason why the operator selected that information about the content. Past information 113 is data that stores past information about these combinations selected by the operator.

[0031] 2 is configured with a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), etc. The control unit 120 operates in accordance with a program 111 stored in the storage unit 110, and executes processing in accordance with the program 111. The control unit 120 includes a selected area acquisition unit 121, a score calculation unit 122, a candidate order arrangement unit 123, a candidate providing unit 124, a candidate acquisition unit 125, and an acquired information recording unit 126 as main functional units provided by the program 111 stored in the storage unit 110.

[0032] The selected area acquisition unit 121 is a functional unit that records information about a selected area selected by an operator. Specifically, the selected area acquisition unit 121 is a functional unit that acquires, from the information terminal 200, information about a selected area that is a portion that the operator has determined to be a possible violation of the terms and conditions and selected from explanatory information about a product or service, an image of the product or service, etc. For example, in the example shown in FIG. 5, the selected area acquisition unit 121 acquires, as the selected area, the portion of the product description that the operator selected by operating the information terminal 200 that reads, "Anyone will definitely lose weight."

[0033] The score calculation unit 122 is a functional unit that calculates a score for each preset violation of the rules for the information of the selection area selected by the selection area acquisition unit 121. Specifically, the score calculation unit 122 is a functional unit that calculates the degree of association between the information of the selection area selected by the selection area acquisition unit 121 and each preset violation of the rules, and calculates the score using the learning model 112 so that the more relevant the violation of the rules, the higher the score. For example, as shown in FIG. 5, when the selection area acquisition unit 121 acquires the part "Anyone can lose weight for sure" as the selection area, the score calculation unit 122 uses the learning model 112 to calculate the score so that "Violation of the Act against Unjustifiable Premiums and Misleading Representations (5)" has the highest score, as shown in FIG. 6. As shown in FIG. 3, the learning model 112 has learned that when the operator selected the selection area for "You'll Lose Weight Guaranteed," the operator selected the violation of the Act against Unjustifiable Premiums and Misleading Representations (5). Since "Anyone Will Lose Weight Guaranteed" acquired by the selection area acquisition unit 121 has a high similarity to "You'll Lose Weight Guaranteed," the score for "Violation of the Act against Unjustifiable Premiums and Misleading Representations (5)" is the highest, as shown in FIG. 6. While this embodiment illustrates an example in which a score is calculated on a 10-point scale from 1 to 10 as shown in FIG. 6, it may also be a 5-point scale or 20-point scale. Furthermore, while an example in which a score is calculated for all violations of the terms and conditions is illustrated, it may also be possible to calculate scores only for violations with the same name. That is, the name of the violation of the terms and conditions may be identified, and then a score may be calculated only for that type. Specifically, if the portion "Anyone Will Lose Weight Guaranteed" is acquired as a selection area, a violation of the Act against Unjustifiable Premiums and Misleading Representations may be identified, and then a score may be calculated only for that type (e.g., Violation of the Act against Unjustifiable Premiums and Misleading Representations (1) to Violation of the Act against Unjustifiable Premiums and Misleading Representations (5)). In this embodiment, an example has been shown in which score calculation unit 122 calculates the score by AI (artificial intelligence) using learning model 112, but differently, score calculation unit 122 may refer to past information 113 and calculate the score based on previously selected selection areas and violations of rules. Specifically, the score of a violation of rules may be calculated according to the similarity of the selection areas.For example, in the example shown in Figure 6, the similarity between the part "Anyone will definitely lose weight" and the selection area of ​​"You will definitely lose weight" as shown in Figure 3 can be calculated, and the value obtained by multiplying the similarity by 10 can be calculated as the score for violation of the terms and conditions indicated by the past information 113.

[0034] Returning to Fig. 2, the candidate order arrangement unit 123 is a functional unit that arranges the items violating the regulations in order of the scores calculated by the function of the score calculation unit 122. Specifically, for example, when the scores calculated by the score calculation unit 122 are as shown in Fig. 6, the candidate order arrangement unit 123 arranges the items violating the regulations, i.e., the candidates to be provided to the operator, in order from highest to lowest score (descending order) so that the violation of the Act against Unjustifiable Premiums and Misleading Representations (5) comes first, as shown in Fig. 7.

[0035] The candidate providing unit 124 shown in Fig. 2 is a functional unit that provides the operator with the rule violation items arranged by the candidate order arrangement unit 123. Specifically, the candidate providing unit 124 transmits the candidates, which are rule violation items arranged by the candidate order arrangement unit 123 as shown in Fig. 7, to the information terminal 200, thereby providing them to the operator as shown in Fig. 8. In this way, the violation candidates are provided to the operator. In this embodiment, the violation candidates are provided in the order of the scores calculated by the score calculation unit 122, so that there is a high degree of consistency with the rule violations corresponding to the selection area selected by the operator.

[0036] The candidate acquisition unit 125 is a functional unit that acquires a candidate selected by an operator from among the candidates that are violations of the regulations and that have been provided to the operator by the candidate providing unit 124. Specifically, the candidate acquisition unit 125 is a functional unit that acquires information about a candidate selected by the operator from among the candidates provided to the operator by the candidate providing unit 124, together with information about the selected area, from the information terminal 200. For example, when the content shown in FIG. 8 is provided to the operator by the function of the candidate providing unit 124 and "Violation of the Act against Unjustifiable Premiums and Misleading Representations (5)" is selected as shown in the figure, the candidate acquisition unit 125 acquires information about "Violation of the Act against Unjustifiable Premiums and Misleading Representations (5)" together with information about the selected area, such as "Anyone will definitely lose weight."

[0037] The acquired information recording unit 126 is a functional unit that associates candidate information acquired by the function of the candidate acquisition unit 125 with information in the selected area and records it in the past information 113. Specifically, when the acquired information recording unit 126 acquires information on "Violation of the Act against Unjustifiable Premiums and Misleading Representations (5)" shown in FIG. 8 together with information in the selected area such as "Anyone can lose weight," by the function of the candidate acquisition unit 125, the acquired information recording unit 126 records the acquired content in the last line of the past information 113, as shown in FIG. 9. This allows the learning model 112 to perform learning based on the newly recorded content.

[0038] The input / output unit 130 is a device that is configured with a keyboard, a mouse, a camera, a microphone, a liquid crystal display, an organic EL (Electro-Luminescence) display, and the like, and is used to input and output various types of data.

[0039] The communication unit 140 is a device that enables the assistance device 100 to communicate with other information terminals such as the information terminal 200 via the computer communication network 400. The configuration of the assistance device 100 has been described above.

[0040] (operation) Next, the operation of the support device 100 will be described. Fig. 4 is a flowchart showing an example of the candidate providing process in the support device 100. The candidate providing process in this embodiment starts to be executed upon receiving a request from an operator who operates the information terminal 200. Note that in this embodiment, for ease of understanding, the following description will be given taking as an example a case where the statement "Anyone will definitely lose weight" in the product description of product A shown in Fig. 5 is selected by the operator (i.e., when the selected area is "Anyone will definitely lose weight").

[0041] 4 starts, the control unit 120 acquires information on the selection area selected by the operator operating the information terminal 200 using the function of the selection area acquisition unit 121 (step S11). Specifically, in the processing of step S11, the selection area acquisition unit 121 acquires the information "Anyone will definitely lose weight" in the product description of product A shown in Fig. 5 from the information terminal 200 as the selection area information. Note that the selection area acquisition unit 121 that executes the processing of step S11 and step S11 correspond to the information acquisition unit and the information acquisition step.

[0042] After executing the process of step S11 shown in FIG. 4, the control unit 120 uses the function of the score calculation unit 122 to calculate a score for each preset violation of the rules for the information of the selected area acquired in the process of step S11 (step S12). Specifically, in the process of step S12, the score calculation unit 122 calculates the degree of association between the information of the selected area acquired in the process of step S11 and each preset violation of the rules, and calculates the score using the learning model 112 so that the more relevant the violation of the rules, the higher the score. That is, in the process of step S12, the score calculation unit 122 calculates a score corresponding to each violation (candidate) of the rules for the selected area "Anyone can lose weight" acquired in the process of step S11, using the learning model 112, so that the score for "Violation of the Act against Unjustifiable Premiums and Misleading Representations (5)" is the highest, as shown in FIG. 6. The score calculation unit 122 and step S12 that execute the process of step S12 correspond to the score calculation unit and the score calculation step.

[0043] After executing the process of step S12 shown in Fig. 4, the control unit 120 uses the function of the candidate order arrangement unit 123 to arrange the items violating the rules in order of the scores calculated in the process of step S12 (step S13). Specifically, in the process of step S13, the candidate order arrangement unit 123 arranges the items violating the rules, i.e., the candidates to be provided to the operator, in order from highest to lowest score calculated in the process of step S12 (descending order), as shown in Fig. 7. The candidate order arrangement unit 123 that executes the process of step S13 corresponds to the arrangement unit.

[0044] After executing the process of step S13 shown in FIG. 4, the control unit 120, using the function of the candidate providing unit 124, provides the operator with the terms and conditions violation items as selection candidates for the operator in the order arranged in the process of step S13 (step S14). Specifically, in the process of step S14, the candidate providing unit 124 provides the operator with the terms and conditions violation items arranged in the process of step S13 by transmitting the candidates to the information terminal 200 in the arranged order, as shown in FIG. 8. This provides the operator with terms and conditions violation candidates. Note that in the process of step S14, as shown in FIG. 8, the information on the selected area, "Anyone can lose weight," acquired in step S11 of FIG. 4, is transmitted to the information terminal 200 along with the terms and conditions violation items (candidates), and is displayed superimposed on the display screen of the information terminal 200. This allows the operator to select a relevant terms and conditions violation from these terms and conditions violation candidates. Furthermore, in this embodiment, the violation candidates are provided in order of score, which provides a high degree of consistency with the terms and conditions violation corresponding to the selected area, and the terms and conditions violation displayed at the top is more likely to be the terms and conditions violation corresponding to the selected area. Therefore, the workload of the operator can be reduced. Note that in the process of step S14, the items (candidates) violating the rules may be selected based on the score, for example, by providing only items (candidates) violating the rules with a score of 3 or more. The candidate providing unit 124 that executes the process of step S14 and step S14 correspond to a transmitting unit and a transmitting step.

[0045] After executing the process of step S14 shown in FIG. 4, the control unit 120, using the function of the candidate acquisition unit 125, acquires information on the candidate selected by the operator from among the candidates that are violations of the rules provided to the operator in the process of step S14 (selected candidate information) (step S15). Specifically, in the process of step S15, if the operator selects "Violation of the Act against Unjustifiable Premiums and Misleading Representations (5)" shown in FIG. 8, the candidate acquisition unit 125 acquires information on the "Violation of the Act against Unjustifiable Premiums and Misleading Representations (5)" along with information on the selected area, such as "Anyone will lose weight." Note that the process of step S15 acquires only information on the selected candidate, and the information on the selected area may be information acquired in the process of step S11 in FIG. 4. The candidate acquisition unit 125 and step S15 that execute the process of step S15 correspond to the violation of rules information acquisition unit and the violation of rules information acquisition step.

[0046] After executing the process of step S15 shown in FIG. 4, the control unit 120 uses the function of the acquired information recording unit 126 to record the information acquired in the process of step S14, i.e., the information in the selection area and the selection candidate information, in association with each other in the past information 113 (step S16), and terminates the candidate providing process. Specifically, in the process of step S16, the acquired information recording unit 126 associates the information in the selection area, such as "Anyone can lose weight," with the information on the selection candidate information, "Violation of the Act against Unjustifiable Premiums and Misleading Representations (5)," in the last line of the past information 113, as shown in FIG. 9. This updates the content of the past information 113. Accordingly, the learning model 112 is retrained based on the updated content of the past information 113. In other words, the learning model 112 is trained every time the content of the past information 113 is updated. The acquired information recording unit 126, which executes the process of step S16, and step S16 correspond to a recording unit and a recording step.

[0047] The above is the operation of the assistance device 100. In this way, according to the assistance device 100 of this embodiment, rule violations that are likely to correspond to the selected area selected by the operator are presented in order of score. In other words, the rule violation displayed at the top is most likely to be consistent with the rule violation corresponding to the selected area. This reduces the workload of the operator.

[0048] (Variation) It should be noted that the present invention is not limited to the above-described embodiments, and various modifications and applications are possible. For example, the assistance device 100 and the information terminal 200 according to the above-described embodiments do not necessarily have all of the technical features described above, but may have some of the configurations described in the above-described embodiments so as to solve at least one problem in the prior art. Furthermore, at least a portion of each of the following modifications may be combined.

[0049] In the above embodiment, as shown in Fig. 5, an example is shown in which information on a selection area selected by an operator operating information terminal 200 is acquired and a score for the violation of the terms and conditions corresponding to the information on the selection area is calculated, but this is just one example. In addition, for example, before the operator selects a selection area, information on the product description or the entire product image may be acquired and a score for the acquired information may be calculated. In this case, if the score is higher than a certain value, the relevant part may be presented to the operator as a part that is likely to be relevant.

[0050] Specifically, when receiving a request from an operator operating the information terminal 200, the assistance device 100 acquires information on the product description shown in Fig. 10 and the entire product image shown in Fig. 11. In the case of the product description, as shown in Fig. 10(A) and Fig. 10(B), the target character string is shifted sequentially by a preset number of characters, and the score for each target character string is calculated in the same manner as in the above embodiment.

[0051] In the example shown in FIG. 10, the set number of characters is set to five. As shown in FIG. 10(A), the support device 100 first extracts the first character string of the product description, "ABCDE," and calculates the score for each violation of the rules in the same manner as when the character string is selected as the selection area in the above embodiment. Next, as shown in FIG. 10(B), the support device 100 shifts one character to extract the character string "BCDEF" as the target character string and calculates the score for each violation of the rules in the same manner as when "BCDEF" is selected as the selection area in the above embodiment. Note that, as shown in FIGS. 10(A) and 10(B), some of the target character strings overlap with each other. Also, this example shows an example in which a predetermined character length (set number of characters: five) is extracted by shifting a predetermined number of characters (one character at a time), but the predetermined number of characters and the predetermined character length can be changed as desired. In this way, the target character string is shifted and a score is calculated for the entire product description. Then, the assistance device 100 may provide the operator with a special background color for target character strings with a score equal to or higher than a preset score (e.g., 8 or higher), as shown in Fig. 10(C). In the example shown in Fig. 10(C), the target character string "PKKKK" has a score equal to or higher than 8, and the background color is displayed in a special mode. Note that, for example, if the score is 10, the background color may be displayed in a special mode different from the special mode.

[0052] Also, for example, when the product description is in English or the like, the target character string may be shifted word by word, and the score may be calculated for each word. Furthermore, in the case of Japanese, the text of the product description may be morphologically analyzed and broken down into word strings (morphemes), and the word strings may be shifted word by word (shifted morpheme by morpheme), that is, the target character string may be shifted word by word, and a score for each word string (score for each morpheme) may be calculated. Note that when the product description is in English or the like, it is sufficient to determine one word for each space, and when in Japanese, it is sufficient to determine one word string using a dictionary and grammar. In other words, when the product description is in English or the like, the character string between spaces is considered to be one word, and the score may be calculated by shifting one word by word.

[0053] On the other hand, in the case of product images, as shown in Figures 11(A) and 11(B), the target area is shifted sequentially by a predetermined number of pixels, and the score for each target area is calculated in the same manner as in the above embodiment.

[0054] In the example shown in FIG. 11, the set pixel count is set to 5×5 pixels. As shown in FIG. 11(A), the support device 100 first calculates the score for each violation of the rules in the same manner as when the first region of the product description, "XXXXX," is selected as the selected region in the above embodiment. Next, as shown in FIG. 11(B), the support device 100 shifts the region by one unit (5 pixels), extracts the "YYYPP" portion as the target region, and calculates the score for each violation of the rules in the same manner as when "YYYPP" is selected as the selected region in the above embodiment. Note that this example shows an example in which a predetermined size (5×5 pixels) is extracted by shifting a predetermined number of pixels (5 pixels), but the predetermined number of pixels and the predetermined size can be changed as desired. Also, as with "BCDE" shown in FIGS. 10(A) and 10(B), portions of the selected regions may overlap each other. In this manner, the target region is shifted and scores are calculated for the entire product image. Then, the support device 100 may provide the operator with a special background color for target areas with a score equal to or higher than a preset score (e.g., 8 or higher), as shown in Fig. 11(C). In the example shown in Fig. 11(C), the score of "You will definitely lose weight" written in the target area is 8 or higher, and the background color is displayed in a special mode. Note that, for example, if the score is 10, the background color may be displayed in a special mode different from the special mode, as in the case of product descriptions.

[0055] In this way, by acquiring information on the entire product description and product images and providing the operator with the parts that are likely to violate the terms and conditions in advance, the operator's workload can be further reduced. In addition, the display may be in a gradual manner, for example, by using a background color that becomes darker as the score increases, such as yellow for a score of 5 to 7, orange for 8 or above, and red for 10. In other words, different display modes may be used depending on the score. Furthermore, in the case of a product page that displays both a product description and product images, the operator may be provided with the parts of each that are likely to violate the terms and conditions in advance. In other words, the processing of this modified example may be performed for the entire page, including the product description section and the product image section. The unit of shift for the product description section and the product image section may be arbitrarily changed.

[0056] Furthermore, if the product description and the entire product image information are acquired and the parts that are likely to be in violation of the terms and conditions are provided to the operator in advance, the candidate providing process is not performed, and when the operator selects a part that is likely to be in violation of the terms and conditions (i.e., a part where the background color is displayed in a special manner), a display screen such as that shown in Fig. 8 may be displayed. Then, the content of the violation of the terms and conditions selected by the operator may be associated with the content of the part where the background color is displayed in a special manner, and recorded in past information 113.

[0057] In the above embodiment, the support device 100 provides support for determining whether or not there is a violation of terms and conditions for products and services posted on a network, but this is just one example, and it is also possible to provide support for determining whether or not there is a violation of terms and conditions for advertisements, news articles, blog articles, etc. In other words, the support device 100 can provide support for determining whether or not there is a violation of terms and conditions for information posted on a network.

[0058] Furthermore, in the above embodiment, an example was given in which the order of arrangement is changed in the processing of step S13 shown in FIG. 4 and then items violating the regulations, i.e., candidates, are provided to the operator, but the processing of step S13 may not be executed and the items may be provided to the operator without changing the order of arrangement.

[0059] The assistance device 100 and the information terminal 200 according to the above-described embodiments can be realized using a normal computer, not a dedicated device. For example, the assistance device 100 that executes the above-described processes may be configured by installing a program for executing any of the above-described processes on a computer from a recording medium storing the program. Also, one assistance device 100 and one information terminal 200 may be configured by multiple computers operating in cooperation with each other.

[0060] Furthermore, when the above-mentioned functions are realized by sharing the functions between an OS (Operating System) and an application, or by cooperation between the OS and the application, only the parts other than the OS may be stored on the medium.

[0061] It is also possible to superimpose the program on a carrier wave and distribute it via a communication network. For example, the program may be posted on a bulletin board system (BBS) on the communication network and distributed via the network. These programs may then be started and run under the control of an operating system in the same way as other application programs, thereby enabling the above-mentioned processing to be performed.

[0062] Various aspects of the present disclosure are summarized below as appendices.

[0063] (Appendix 1) an information acquisition unit that acquires information posted on a network; a score calculation unit that calculates a score indicating a possibility that the content indicated by the information acquired by the information acquisition unit corresponds to a violation of a predetermined rule; a transmitting unit that transmits information about the violation of the rules to an information terminal operated by an operator based on the score calculated by the score calculating unit; a regulation violation information acquisition unit that acquires information on the regulation violation selected by the operator from the information on the regulation violation transmitted by the transmission unit; a recording unit that records the information on the violation of the rules acquired by the information acquisition unit in association with the information acquired by the information acquisition unit, the score calculation unit calculates the score using a learning model that is trained so that the higher the degree of association between the content indicated by the information acquired by the information acquisition unit and the violation of the predetermined terms and conditions, the higher the score; 1. A support device comprising:

[0064] (Appendix 2) an arrangement unit that arranges the order of the violations of the rules corresponding to the content indicated by the information acquired by the information acquisition unit based on the score calculated by the score calculation unit, the information acquisition unit acquires, from the information terminal, information selected by an operation of the operator from among the information posted on the network; the transmission unit transmits the information on the violation of the regulations arranged by the arrangement unit to the information terminal operated by the operator; the recording unit records the information on the violation of the regulations acquired by the information acquisition unit in association with the information selected by the operation of the operator acquired by the information acquisition unit. 2. The assistance device according to claim 1,

[0065] (Appendix 3) the information acquisition unit acquires information about products or services posted on a network, the score calculation unit sequentially calculates the score for each preset unit for the information on the product or service acquired by the information acquisition unit; the transmitting unit transmits, to the information terminal, as a special aspect, a background of a portion of the information on the product or service where the score for each unit calculated by the score calculating unit is equal to or greater than a predetermined value; the recording unit records the information on the violation of the terms and conditions acquired by the information acquisition unit on the violation of the terms and conditions in association with a portion of the information on the product or service for which the score for each unit calculated by the score calculation unit is equal to or greater than a predetermined value; 2. The assistance device according to claim 1,

[0066] (Appendix 4) The learning model further learns based on the information on the violation of the rules recorded by the recording unit and the information acquired by the information acquisition unit. 4. The support device according to any one of claims 1 to 3.

[0067] (Appendix 5) The predetermined unit is a character string of a predetermined length extracted by shifting a predetermined number of characters from the text related to the information of the product or service acquired by the information acquisition unit. 4. The assistance device according to claim 3,

[0068] (Appendix 6) The predetermined unit is a predetermined range of data extracted by shifting the image related to the product or service information acquired by the information acquisition unit by a predetermined number of pixels. 4. The assistance device according to claim 3,

[0069] (Appendix 7) The predetermined unit is extracted by shifting character strings between spaces in units of character strings from the text related to the information on the product or service acquired by the information acquisition unit. 8. The assistance device according to claim 7,

[0070] (Appendix 8) the predetermined unit is obtained by morpheme analysis of the text related to the information on the product or service acquired by the information acquisition unit, and by shifting and extracting the morpheme from the text in morpheme units; 4. The assistance device according to claim 3,

[0071] (Appendix 9) the transmitting unit transmits to the information terminal, in a different manner depending on the score, a background of a portion of the information on the product or service where the score for each unit calculated by the score calculating unit is equal to or greater than a predetermined value. 5. The assistance device according to claim 3 or 4.

[0072] (Appendix 10) the information acquisition unit acquires description information and image information from the information on the product or service posted on the network, the score calculation unit sequentially calculates scores for each of the description information and the image information acquired by the information acquisition unit, for each predetermined unit; 6. The support device according to any one of Supplementary Notes 3 to 5.

[0073] (Appendix 11) A support method using a support device, an information acquisition step of acquiring information posted on a network; a score calculation step of calculating a score indicating a possibility that the content indicated by the information acquired in the information acquisition step corresponds to a violation of a predetermined rule; a transmitting step of transmitting information about the violation of the rules to an information terminal operated by an operator based on the score calculated in the score calculation step; a rule violation information acquisition step of acquiring the rule violation information selected by the operator from the rule violation information transmitted in the transmission step; a recording step of recording the information on the violation of the rules acquired in the information acquisition step in association with the information acquired in the information acquisition step, In the score calculation step, the score is calculated using a learning model that is trained so that the higher the degree of association between the content indicated by the information acquired in the information acquisition step and the violation of the predetermined terms and conditions, the higher the score. A support method characterized by:

[0074] (Appendix 12) Support equipment, an information acquisition unit that acquires information posted on the network; a score calculation unit that calculates a score indicating a possibility that the content indicated by the information acquired by the information acquisition unit corresponds to a violation of a predetermined rule; a transmission unit that transmits information about the violation of the rules to an information terminal operated by an operator based on the score calculated by the score calculation unit; a rules violation information acquisition unit that acquires the rules violation information selected by the operator from the rules violation information transmitted by the transmission unit; a recording unit that records the information on the violation of the rules acquired by the information acquisition unit in association with the information acquired by the information acquisition unit; the score calculation unit calculates the score using a learning model that is trained so that the higher the degree of association between the content indicated by the information acquired by the information acquisition unit and the violation of the predetermined terms and conditions, the higher the score; A program characterized by:

[0075] The present disclosure allows various embodiments and modifications without departing from the broad spirit and scope of the present disclosure. Furthermore, the above-described embodiments are intended to illustrate the present disclosure and do not limit the scope of the present disclosure. That is, the scope of the present disclosure is defined by the claims, not the embodiments. Various modifications made within the scope of the claims and the meaning of equivalent disclosures are considered to be within the scope of the present disclosure. [Industrial Applicability]

[0076] According to the present invention, it is possible to provide an assistance device, an assistance method, and a program that can reduce the workload of an operator. [Explanation of symbols]

[0077] 100 Support equipment 110 Storage section 111 Program 112 Learning Model 113 Past information 120 control section 121 Selection area acquisition section 122 Score Calculation Section 123 Candidate Order Arrangement Section 124 Candidate Providing Department 125 Candidate acquisition section 126 Acquired Information Recording Unit 130 Input / output section 140 Communications Department 200 Information terminal 400 Computer Network

Claims

1. an information acquisition unit that acquires information posted on a network; a score calculation unit that calculates a score indicating a possibility that the content indicated by the information acquired by the information acquisition unit corresponds to a violation of a predetermined rule; a transmitting unit that transmits information about the violation of the rules to an information terminal operated by an operator based on the score calculated by the score calculating unit; a regulation violation information acquisition unit that acquires information on the regulation violation selected by the operator from the information on the regulation violation transmitted by the transmission unit; a recording unit that records the information on the violation of the rules acquired by the information acquisition unit in association with the information acquired by the information acquisition unit, the score calculation unit calculates the score using a learning model that is trained so that the higher the degree of association between the content indicated by the information acquired by the information acquisition unit and the violation of the predetermined terms and conditions, the higher the score; 1. A support device comprising:

2. an arrangement unit that arranges the order of the violations of the rules corresponding to the content indicated by the information acquired by the information acquisition unit based on the score calculated by the score calculation unit, the information acquisition unit acquires, from the information terminal, information selected by an operation of the operator from among the information posted on the network; the transmission unit transmits the information on the violation of the regulations arranged by the arrangement unit to the information terminal operated by the operator; the recording unit records the information on the violation of the regulations acquired by the information acquisition unit in association with the information selected by the operation of the operator acquired by the information acquisition unit.

2. The support device according to claim 1.

3. the information acquisition unit acquires information about products or services posted on a network, the score calculation unit sequentially calculates the score for each preset unit for the information on the product or service acquired by the information acquisition unit; the transmitting unit transmits, to the information terminal, as a special aspect, a background of a portion of the information on the product or service where the score for each unit calculated by the score calculating unit is equal to or greater than a predetermined value; the recording unit records the information on the violation of the terms and conditions acquired by the information acquisition unit on the violation of the terms and conditions in association with a portion of the information on the product or service for which the score for each unit calculated by the score calculation unit is equal to or greater than a predetermined value; 2. The support device according to claim 1.

4. The learning model further learns based on the information on the violation of the rules recorded by the recording unit and the information acquired by the information acquisition unit.

4. The support device according to claim 1, wherein the support device is a device for supporting a user.

5. The predetermined unit is a character string of a predetermined length extracted by shifting a predetermined number of characters from the text related to the information of the product or service acquired by the information acquisition unit.

4. The support device according to claim 3.

6. The predetermined unit is a predetermined range of data extracted by shifting the image related to the product or service information acquired by the information acquisition unit by a predetermined number of pixels.

4. The support device according to claim 3.

7. The predetermined unit is extracted by shifting character strings between spaces in units of character strings from the text related to the information on the product or service acquired by the information acquisition unit.

4. The support device according to claim 3.

8. the predetermined unit is obtained by morpheme analysis of the text related to the information on the product or service acquired by the information acquisition unit, and by shifting and extracting the morpheme from the text in morpheme units; 4. The support device according to claim 3.

9. the transmitting unit transmits to the information terminal, in a different manner depending on the score, a background of a portion of the information on the product or service where the score for each unit calculated by the score calculating unit is equal to or greater than a predetermined value.

4. The support device according to claim 3.

10. the information acquisition unit acquires description information and image information from the information on the product or service posted on the network, the score calculation unit sequentially calculates scores for each of the description information and the image information acquired by the information acquisition unit in units of a predetermined value; 4. The support device according to claim 3.

11. A support method using a support device, an information acquisition step of acquiring information posted on a network; a score calculation step of calculating a score indicating a possibility that the content indicated by the information acquired in the information acquisition step corresponds to a violation of a predetermined rule; a transmitting step of transmitting information about the violation of the rules to an information terminal operated by an operator based on the score calculated in the score calculation step; a rule violation information acquisition step of acquiring the rule violation information selected by the operator from the rule violation information transmitted in the transmission step; a recording step of recording the information on the violation of the rules acquired in the information acquisition step in association with the information acquired in the information acquisition step, In the score calculation step, the score is calculated using a learning model that is trained so that the higher the degree of association between the content indicated by the information acquired in the information acquisition step and the violation of the predetermined terms and conditions, the higher the score. A support method characterized by:

12. Support equipment, an information acquisition unit that acquires information posted on the network; a score calculation unit that calculates a score indicating a possibility that the content indicated by the information acquired by the information acquisition unit corresponds to a violation of a predetermined rule; a transmission unit that transmits information about the violation of the rules to an information terminal operated by an operator based on the score calculated by the score calculation unit; a rules violation information acquisition unit that acquires the rules violation information selected by the operator from the rules violation information transmitted by the transmission unit; a recording unit that records the information on the violation of the rules acquired by the information acquisition unit in association with the information acquired by the information acquisition unit; the score calculation unit calculates the score using a learning model that is trained so that the higher the degree of association between the content indicated by the information acquired by the information acquisition unit and the violation of the predetermined terms and conditions, the higher the score; A program characterized by:

Images